Add PyPI package (#576)

* Add PyPI package

* fixes

* fixes

.github/workflows/basic-tests-linux-uv.yml

@@ -66,3 +66,9 @@ jobs:
         run: |
           source .venv/bin/activate
           pytest ch02/05_bpe-from-scratch/tests/tests.py
+
+      - name: Test Selected Bonus Materials
+        shell: bash
+        run: |
+          source .venv/bin/activate
+          pytest pkg/llms_from_scratch/tests/

appendix-D/01_main-chapter-code/appendix-D.ipynb

@@ -68,6 +68,11 @@
     "\n",
     "\n",
     "from previous_chapters import GPTModel\n",
+    "# If the `previous_chapters.py` file is not available locally,\n",
+    "# you can import it from the `llms-from-scratch` PyPI package.\n",
+    "# For details, see: https://github.com/rasbt/LLMs-from-scratch/tree/main/pkg\n",
+    "# E.g.,\n",
+    "# from llms_from_scratch.ch04 import GPTModel\n",
     "\n",
     "GPT_CONFIG_124M = {\n",
     "    \"vocab_size\": 50257,   # Vocabulary size\n",
@@ -139,6 +144,9 @@
    "outputs": [],
    "source": [
     "from previous_chapters import create_dataloader_v1\n",
+    "# Alternatively:\n",
+    "# from llms_from_scratch.ch02 import create_dataloader_v1\n",
+    "\n",
     "\n",
     "# Train/validation ratio\n",
     "train_ratio = 0.90\n",
@@ -454,6 +462,9 @@
    "outputs": [],
    "source": [
     "from previous_chapters import calc_loss_batch\n",
+    "# Alternatively:\n",
+    "# from llms_from_scratch.ch05 import calc_loss_batch\n",
+    "\n",
     "\n",
     "torch.manual_seed(123)\n",
     "model = GPTModel(GPT_CONFIG_124M)\n",
@@ -551,6 +562,9 @@
    "outputs": [],
    "source": [
     "from previous_chapters import evaluate_model, generate_and_print_sample\n",
+    "# Alternatively:\n",
+    "# from llms_from_scratch.ch05 import evaluate_model, generate_and_print_samplee\n",
+    "\n",
     "\n",
     "ORIG_BOOK_VERSION = False\n",
     "\n",
@@ -790,6 +804,9 @@
    ],
    "source": [
     "from previous_chapters import plot_losses\n",
+    "# Alternatively:\n",
+    "# from llms_from_scratch.ch05 import plot_losses\n",
+    "\n",
     "\n",
     "epochs_tensor = torch.linspace(1, n_epochs, len(train_losses))\n",
     "plot_losses(epochs_tensor, tokens_seen, train_losses, val_losses)\n",
@@ -823,7 +840,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.11.4"
+   "version": "3.10.16"
   }
  },
  "nbformat": 4,

appendix-E/01_main-chapter-code/appendix-E.ipynb

@@ -198,6 +198,16 @@
     "    create_balanced_dataset,\n",
     "    random_split\n",
     ")\n",
+    "# If the `previous_chapters.py` file is not available locally,\n",
+    "# you can import it from the `llms-from-scratch` PyPI package.\n",
+    "# For details, see: https://github.com/rasbt/LLMs-from-scratch/tree/main/pkg\n",
+    "# E.g.,\n",
+    "# from llms_from_scratch.ch06 import (\n",
+    "#     download_and_unzip_spam_data,\n",
+    "#     create_balanced_dataset,\n",
+    "#     random_split\n",
+    "# )\n",
+    "\n",
     "\n",
     "\n",
     "url = \"https://archive.ics.uci.edu/static/public/228/sms+spam+collection.zip\"\n",
@@ -409,6 +419,10 @@
    "source": [
     "from gpt_download import download_and_load_gpt2\n",
     "from previous_chapters import GPTModel, load_weights_into_gpt\n",
+    "# Alternatively:\n",
+    "# from llms_from_scratch.ch04 import GPTModel\n",
+    "# from llms_from_scratch.ch05 import load_weights_into_gpt\n",
+    "\n",
     "\n",
     "\n",
     "CHOOSE_MODEL = \"gpt2-small (124M)\"\n",
@@ -577,6 +591,9 @@
    ],
    "source": [
     "from previous_chapters import calc_accuracy_loader\n",
+    "# Alternatively:\n",
+    "# from llms_from_scratch.ch06 import calc_accuracy_loader\n",
+    "\n",
     "\n",
     "\n",
     "torch.manual_seed(123)\n",
@@ -1387,6 +1404,8 @@
    "source": [
     "import time\n",
     "from previous_chapters import train_classifier_simple\n",
+    "# Alternatively:\n",
+    "# from llms_from_scratch.ch06 import train_classifier_simple\n",
     "\n",
     "\n",
     "start_time = time.time()\n",
@@ -1442,6 +1461,8 @@
    ],
    "source": [
     "from previous_chapters import plot_values\n",
+    "# Alternatively:\n",
+    "# from llms_from_scratch.ch06 import plot_values\n",
     "\n",
     "epochs_tensor = torch.linspace(0, num_epochs, len(train_losses))\n",
     "examples_seen_tensor = torch.linspace(0, examples_seen, len(train_losses))\n",

ch04/01_main-chapter-code/ch04.ipynb

@@ -970,6 +970,12 @@
    "metadata": {},
    "outputs": [],
    "source": [
+    "# If the `previous_chapters.py` file is not available locally,\n",
+    "# you can import it from the `llms-from-scratch` PyPI package.\n",
+    "# For details, see: https://github.com/rasbt/LLMs-from-scratch/tree/main/pkg\n",
+    "# E.g.,\n",
+    "# from llms_from_scratch.ch03 import MultiHeadAttention\n",
+    "\n",
     "from previous_chapters import MultiHeadAttention\n",
     "\n",
     "\n",

ch04/02_performance-analysis/README.md

@@ -1,5 +1,4 @@
 # Chapter 4: Implementing a GPT Model from Scratch To Generate Text
 
 - [flops-analysis.ipynb](flops-analysis.ipynb) analyses the floating point operations per second (FLOPS) of the GPT model(s) implemented in the main chapter. 
-- [previous_chapters.py](previous_chapters.py) is a Python module containing the `GPTModel` code we implemented in chapter 4 and other code implemented in previous chapters, which we import in the analysis notebook.
 - `requirements-extra.txt` includes additional Python libraries that need to be installed (via `pip install -r requirements-extra.txt`.

ch04/02_performance-analysis/flops-analysis.ipynb

@@ -127,7 +127,9 @@
     "import torch\n",
     "from thop import profile\n",
     "\n",
-    "from previous_chapters import GPTModel\n",
+    "# For installation instructions, see:\n",
+    "# https://github.com/rasbt/LLMs-from-scratch/tree/main/pkg\n",
+    "from llms_from_scratch.ch04 import GPTModel\n",
     "\n",
     "\n",
     "BASE_CONFIG = {\n",
@@ -550,7 +552,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.11.4"
+   "version": "3.10.16"
   }
  },
  "nbformat": 4,

ch05/01_main-chapter-code/ch05.ipynb

@@ -147,6 +147,11 @@
    "source": [
     "import torch\n",
     "from previous_chapters import GPTModel\n",
+    "# If the `previous_chapters.py` file is not available locally,\n",
+    "# you can import it from the `llms-from-scratch` PyPI package.\n",
+    "# For details, see: https://github.com/rasbt/LLMs-from-scratch/tree/main/pkg\n",
+    "# E.g.,\n",
+    "# from llms_from_scratch.ch04 import GPTModel\n",
     "\n",
     "GPT_CONFIG_124M = {\n",
     "    \"vocab_size\": 50257,   # Vocabulary size\n",
@@ -212,6 +217,9 @@
     "import tiktoken\n",
     "from previous_chapters import generate_text_simple\n",
     "\n",
+    "# Alternatively:\n",
+    "# from llms_from_scratch.ch04 import generate_text_simple\n",
+    "\n",
     "def text_to_token_ids(text, tokenizer):\n",
     "    encoded = tokenizer.encode(text, allowed_special={'<|endoftext|>'})\n",
     "    encoded_tensor = torch.tensor(encoded).unsqueeze(0) # add batch dimension\n",
@@ -924,6 +932,8 @@
    "outputs": [],
    "source": [
     "from previous_chapters import create_dataloader_v1\n",
+    "# Alternatively:\n",
+    "# from llms_from_scratch.ch02 import create_dataloader_v1\n",
     "\n",
     "# Train/validation ratio\n",
     "train_ratio = 0.90\n",
@@ -2548,7 +2558,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.12.8"
+   "version": "3.10.16"
   }
  },
  "nbformat": 4,

ch05/02_alternative_weight_loading/previous_chapters.py

@@ -1,293 +0,0 @@
-# Copyright (c) Sebastian Raschka under Apache License 2.0 (see LICENSE.txt).
-# Source for "Build a Large Language Model From Scratch"
-#   - https://www.manning.com/books/build-a-large-language-model-from-scratch
-# Code: https://github.com/rasbt/LLMs-from-scratch
-#
-# This file collects all the relevant code that we covered thus far
-# throughout Chapters 2-4.
-# This file can be run as a standalone script.
-
-import tiktoken
-import torch
-import torch.nn as nn
-from torch.utils.data import Dataset, DataLoader
-
-#####################################
-# Chapter 2
-#####################################
-
-
-class GPTDatasetV1(Dataset):
-    def __init__(self, txt, tokenizer, max_length, stride):
-        self.input_ids = []
-        self.target_ids = []
-
-        # Tokenize the entire text
-        token_ids = tokenizer.encode(txt, allowed_special={"<|endoftext|>"})
-
-        # Use a sliding window to chunk the book into overlapping sequences of max_length
-        for i in range(0, len(token_ids) - max_length, stride):
-            input_chunk = token_ids[i:i + max_length]
-            target_chunk = token_ids[i + 1: i + max_length + 1]
-            self.input_ids.append(torch.tensor(input_chunk))
-            self.target_ids.append(torch.tensor(target_chunk))
-
-    def __len__(self):
-        return len(self.input_ids)
-
-    def __getitem__(self, idx):
-        return self.input_ids[idx], self.target_ids[idx]
-
-
-def create_dataloader_v1(txt, batch_size=4, max_length=256,
-                         stride=128, shuffle=True, drop_last=True, num_workers=0):
-    # Initialize the tokenizer
-    tokenizer = tiktoken.get_encoding("gpt2")
-
-    # Create dataset
-    dataset = GPTDatasetV1(txt, tokenizer, max_length, stride)
-
-    # Create dataloader
-    dataloader = DataLoader(
-        dataset, batch_size=batch_size, shuffle=shuffle, drop_last=drop_last, num_workers=num_workers)
-
-    return dataloader
-
-
-#####################################
-# Chapter 3
-#####################################
-class MultiHeadAttention(nn.Module):
-    def __init__(self, d_in, d_out, context_length, dropout, num_heads, qkv_bias=False):
-        super().__init__()
-        assert d_out % num_heads == 0, "d_out must be divisible by n_heads"
-
-        self.d_out = d_out
-        self.num_heads = num_heads
-        self.head_dim = d_out // num_heads  # Reduce the projection dim to match desired output dim
-
-        self.W_query = nn.Linear(d_in, d_out, bias=qkv_bias)
-        self.W_key = nn.Linear(d_in, d_out, bias=qkv_bias)
-        self.W_value = nn.Linear(d_in, d_out, bias=qkv_bias)
-        self.out_proj = nn.Linear(d_out, d_out)  # Linear layer to combine head outputs
-        self.dropout = nn.Dropout(dropout)
-        self.register_buffer('mask', torch.triu(torch.ones(context_length, context_length), diagonal=1))
-
-    def forward(self, x):
-        b, num_tokens, d_in = x.shape
-
-        keys = self.W_key(x)  # Shape: (b, num_tokens, d_out)
-        queries = self.W_query(x)
-        values = self.W_value(x)
-
-        # We implicitly split the matrix by adding a `num_heads` dimension
-        # Unroll last dim: (b, num_tokens, d_out) -> (b, num_tokens, num_heads, head_dim)
-        keys = keys.view(b, num_tokens, self.num_heads, self.head_dim)
-        values = values.view(b, num_tokens, self.num_heads, self.head_dim)
-        queries = queries.view(b, num_tokens, self.num_heads, self.head_dim)
-
-        # Transpose: (b, num_tokens, num_heads, head_dim) -> (b, num_heads, num_tokens, head_dim)
-        keys = keys.transpose(1, 2)
-        queries = queries.transpose(1, 2)
-        values = values.transpose(1, 2)
-
-        # Compute scaled dot-product attention (aka self-attention) with a causal mask
-        attn_scores = queries @ keys.transpose(2, 3)  # Dot product for each head
-
-        # Original mask truncated to the number of tokens and converted to boolean
-        mask_bool = self.mask.bool()[:num_tokens, :num_tokens]
-
-        # Use the mask to fill attention scores
-        attn_scores.masked_fill_(mask_bool, -torch.inf)
-
-        attn_weights = torch.softmax(attn_scores / keys.shape[-1]**0.5, dim=-1)
-        attn_weights = self.dropout(attn_weights)
-
-        # Shape: (b, num_tokens, num_heads, head_dim)
-        context_vec = (attn_weights @ values).transpose(1, 2)
-
-        # Combine heads, where self.d_out = self.num_heads * self.head_dim
-        context_vec = context_vec.reshape(b, num_tokens, self.d_out)
-        context_vec = self.out_proj(context_vec)  # optional projection
-
-        return context_vec
-
-
-#####################################
-# Chapter 4
-#####################################
-class LayerNorm(nn.Module):
-    def __init__(self, emb_dim):
-        super().__init__()
-        self.eps = 1e-5
-        self.scale = nn.Parameter(torch.ones(emb_dim))
-        self.shift = nn.Parameter(torch.zeros(emb_dim))
-
-    def forward(self, x):
-        mean = x.mean(dim=-1, keepdim=True)
-        var = x.var(dim=-1, keepdim=True, unbiased=False)
-        norm_x = (x - mean) / torch.sqrt(var + self.eps)
-        return self.scale * norm_x + self.shift
-
-
-class GELU(nn.Module):
-    def __init__(self):
-        super().__init__()
-
-    def forward(self, x):
-        return 0.5 * x * (1 + torch.tanh(
-            torch.sqrt(torch.tensor(2.0 / torch.pi)) *
-            (x + 0.044715 * torch.pow(x, 3))
-        ))
-
-
-class FeedForward(nn.Module):
-    def __init__(self, cfg):
-        super().__init__()
-        self.layers = nn.Sequential(
-            nn.Linear(cfg["emb_dim"], 4 * cfg["emb_dim"]),
-            GELU(),
-            nn.Linear(4 * cfg["emb_dim"], cfg["emb_dim"]),
-        )
-
-    def forward(self, x):
-        return self.layers(x)
-
-
-class TransformerBlock(nn.Module):
-    def __init__(self, cfg):
-        super().__init__()
-        self.att = MultiHeadAttention(
-            d_in=cfg["emb_dim"],
-            d_out=cfg["emb_dim"],
-            context_length=cfg["context_length"],
-            num_heads=cfg["n_heads"],
-            dropout=cfg["drop_rate"],
-            qkv_bias=cfg["qkv_bias"])
-        self.ff = FeedForward(cfg)
-        self.norm1 = LayerNorm(cfg["emb_dim"])
-        self.norm2 = LayerNorm(cfg["emb_dim"])
-        self.drop_shortcut = nn.Dropout(cfg["drop_rate"])
-
-    def forward(self, x):
-        # Shortcut connection for attention block
-        shortcut = x
-        x = self.norm1(x)
-        x = self.att(x)   # Shape [batch_size, num_tokens, emb_size]
-        x = self.drop_shortcut(x)
-        x = x + shortcut  # Add the original input back
-
-        # Shortcut connection for feed-forward block
-        shortcut = x
-        x = self.norm2(x)
-        x = self.ff(x)
-        x = self.drop_shortcut(x)
-        x = x + shortcut  # Add the original input back
-
-        return x
-
-
-class GPTModel(nn.Module):
-    def __init__(self, cfg):
-        super().__init__()
-        self.tok_emb = nn.Embedding(cfg["vocab_size"], cfg["emb_dim"])
-        self.pos_emb = nn.Embedding(cfg["context_length"], cfg["emb_dim"])
-        self.drop_emb = nn.Dropout(cfg["drop_rate"])
-
-        self.trf_blocks = nn.Sequential(
-            *[TransformerBlock(cfg) for _ in range(cfg["n_layers"])])
-
-        self.final_norm = LayerNorm(cfg["emb_dim"])
-        self.out_head = nn.Linear(cfg["emb_dim"], cfg["vocab_size"], bias=False)
-
-    def forward(self, in_idx):
-        batch_size, seq_len = in_idx.shape
-        tok_embeds = self.tok_emb(in_idx)
-        pos_embeds = self.pos_emb(torch.arange(seq_len, device=in_idx.device))
-        x = tok_embeds + pos_embeds  # Shape [batch_size, num_tokens, emb_size]
-        x = self.drop_emb(x)
-        x = self.trf_blocks(x)
-        x = self.final_norm(x)
-        logits = self.out_head(x)
-        return logits
-
-
-def generate_text_simple(model, idx, max_new_tokens, context_size):
-    # idx is (B, T) array of indices in the current context
-    for _ in range(max_new_tokens):
-
-        # Crop current context if it exceeds the supported context size
-        # E.g., if LLM supports only 5 tokens, and the context size is 10
-        # then only the last 5 tokens are used as context
-        idx_cond = idx[:, -context_size:]
-
-        # Get the predictions
-        with torch.no_grad():
-            logits = model(idx_cond)
-
-        # Focus only on the last time step
-        # (batch, n_token, vocab_size) becomes (batch, vocab_size)
-        logits = logits[:, -1, :]
-
-        # Get the idx of the vocab entry with the highest logits value
-        idx_next = torch.argmax(logits, dim=-1, keepdim=True)  # (batch, 1)
-
-        # Append sampled index to the running sequence
-        idx = torch.cat((idx, idx_next), dim=1)  # (batch, n_tokens+1)
-
-    return idx
-
-
-#####################################
-# Chapter 5
-#####################################
-
-
-def text_to_token_ids(text, tokenizer):
-    encoded = tokenizer.encode(text)
-    encoded_tensor = torch.tensor(encoded).unsqueeze(0)  # add batch dimension
-    return encoded_tensor
-
-
-def token_ids_to_text(token_ids, tokenizer):
-    flat = token_ids.squeeze(0)  # remove batch dimension
-    return tokenizer.decode(flat.tolist())
-
-
-def generate(model, idx, max_new_tokens, context_size, temperature=0.0, top_k=None, eos_id=None):
-
-    # For-loop is the same as before: Get logits, and only focus on last time step
-    for _ in range(max_new_tokens):
-        idx_cond = idx[:, -context_size:]
-        with torch.no_grad():
-            logits = model(idx_cond)
-        logits = logits[:, -1, :]
-
-        # New: Filter logits with top_k sampling
-        if top_k is not None:
-            # Keep only top_k values
-            top_logits, _ = torch.topk(logits, top_k)
-            min_val = top_logits[:, -1]
-            logits = torch.where(logits < min_val, torch.tensor(float('-inf')).to(logits.device), logits)
-
-        # New: Apply temperature scaling
-        if temperature > 0.0:
-            logits = logits / temperature
-
-            # Apply softmax to get probabilities
-            probs = torch.softmax(logits, dim=-1)  # (batch_size, context_len)
-
-            # Sample from the distribution
-            idx_next = torch.multinomial(probs, num_samples=1)  # (batch_size, 1)
-
-        # Otherwise same as before: get idx of the vocab entry with the highest logits value
-        else:
-            idx_next = torch.argmax(logits, dim=-1, keepdim=True)  # (batch_size, 1)
-
-        if idx_next == eos_id:  # Stop generating early if end-of-sequence token is encountered and eos_id is specified
-            break
-
-        # Same as before: append sampled index to the running sequence
-        idx = torch.cat((idx, idx_next), dim=1)  # (batch_size, num_tokens+1)
-
-    return idx

ch05/02_alternative_weight_loading/weight-loading-hf-safetensors.ipynb

@@ -95,7 +95,9 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "from previous_chapters import GPTModel"
+    "from llms_from_scratch.ch04 import GPTModel\n",
+    "# For llms_from_scratch installation instructions, see:\n",
+    "# https://github.com/rasbt/LLMs-from-scratch/tree/main/pkg"
    ]
   },
   {
@@ -270,7 +272,8 @@
    ],
    "source": [
     "import tiktoken\n",
-    "from previous_chapters import generate, text_to_token_ids, token_ids_to_text\n",
+    "from llms_from_scratch.ch05 import generate, text_to_token_ids, token_ids_to_text\n",
+    "\n",
     "\n",
     "torch.manual_seed(123)\n",
     "\n",

ch05/02_alternative_weight_loading/weight-loading-hf-transformers.ipynb

@@ -230,7 +230,9 @@
    "outputs": [],
    "source": [
     "import torch\n",
-    "from previous_chapters import GPTModel\n",
+    "from llms_from_scratch.ch04 import GPTModel\n",
+    "# For llms_from_scratch installation instructions, see:\n",
+    "# https://github.com/rasbt/LLMs-from-scratch/tree/main/\n",
     "\n",
     "\n",
     "gpt = GPTModel(BASE_CONFIG)\n",
@@ -258,7 +260,8 @@
    ],
    "source": [
     "import tiktoken\n",
-    "from previous_chapters import generate, text_to_token_ids, token_ids_to_text\n",
+    "from llms_from_scratch.ch05 import generate, text_to_token_ids, token_ids_to_text\n",
+    "\n",
     "\n",
     "torch.manual_seed(123)\n",
     "\n",

ch05/03_bonus_pretraining_on_gutenberg/pretraining_simple.py

@@ -17,14 +17,12 @@ from pathlib import Path
 import time
 import tiktoken
 import torch
-from previous_chapters import (
+
-    create_dataloader_v1,
+# For llms_from_scratch installation instructions, see:
-    GPTModel,
+# https://github.com/rasbt/LLMs-from-scratch/tree/main/pkg
-    generate_and_print_sample,
+from llms_from_scratch.ch02 import create_dataloader_v1
-    calc_loss_batch,
+from llms_from_scratch.ch04 import GPTModel, generate_and_print_sample
-    evaluate_model,
+from llms_from_scratch.ch05 import calc_loss_batch, evaluate_model, plot_losses
-    plot_losses
-)
 
 
 def read_text_file(file_path):

ch05/03_bonus_pretraining_on_gutenberg/previous_chapters.py

@@ -1,317 +0,0 @@
-# Copyright (c) Sebastian Raschka under Apache License 2.0 (see LICENSE.txt).
-# Source for "Build a Large Language Model From Scratch"
-#   - https://www.manning.com/books/build-a-large-language-model-from-scratch
-# Code: https://github.com/rasbt/LLMs-from-scratch
-
-# This file collects all the relevant code that we covered thus far
-# throughout Chapters 2-4.
-# This file can be run as a standalone script.
-
-import tiktoken
-import torch
-import torch.nn as nn
-from torch.utils.data import Dataset, DataLoader
-import matplotlib.pyplot as plt
-from matplotlib.ticker import MaxNLocator
-
-#####################################
-# Chapter 2
-#####################################
-
-
-class GPTDatasetV1(Dataset):
-    def __init__(self, txt, tokenizer, max_length, stride):
-        self.input_ids = []
-        self.target_ids = []
-
-        token_ids = tokenizer.encode(txt, allowed_special={'<|endoftext|>'})
-
-        for i in range(0, len(token_ids) - max_length, stride):
-            input_chunk = token_ids[i:i + max_length]
-            target_chunk = token_ids[i + 1: i + max_length + 1]
-            self.input_ids.append(torch.tensor(input_chunk))
-            self.target_ids.append(torch.tensor(target_chunk))
-
-    def __len__(self):
-        return len(self.input_ids)
-
-    def __getitem__(self, idx):
-        return self.input_ids[idx], self.target_ids[idx]
-
-
-def create_dataloader_v1(txt, batch_size=4, max_length=256,
-                         stride=128, shuffle=True, drop_last=True, num_workers=0):
-    tokenizer = tiktoken.get_encoding("gpt2")
-    dataset = GPTDatasetV1(txt, tokenizer, max_length, stride)
-    dataloader = DataLoader(
-        dataset, batch_size=batch_size, shuffle=shuffle, drop_last=drop_last, num_workers=num_workers)
-
-    return dataloader
-
-
-#####################################
-# Chapter 3
-#####################################
-
-class MultiHeadAttention(nn.Module):
-    def __init__(self, d_in, d_out, context_length, dropout, num_heads, qkv_bias=False):
-        super().__init__()
-        assert d_out % num_heads == 0, "d_out must be divisible by n_heads"
-
-        self.d_out = d_out
-        self.num_heads = num_heads
-        self.head_dim = d_out // num_heads  # Reduce the projection dim to match desired output dim
-
-        self.W_query = nn.Linear(d_in, d_out, bias=qkv_bias)
-        self.W_key = nn.Linear(d_in, d_out, bias=qkv_bias)
-        self.W_value = nn.Linear(d_in, d_out, bias=qkv_bias)
-        self.out_proj = nn.Linear(d_out, d_out)  # Linear layer to combine head outputs
-        self.dropout = nn.Dropout(dropout)
-        self.register_buffer('mask', torch.triu(torch.ones(context_length, context_length), diagonal=1))
-
-    def forward(self, x):
-        b, num_tokens, d_in = x.shape
-
-        keys = self.W_key(x)  # Shape: (b, num_tokens, d_out)
-        queries = self.W_query(x)
-        values = self.W_value(x)
-
-        # We implicitly split the matrix by adding a `num_heads` dimension
-        # Unroll last dim: (b, num_tokens, d_out) -> (b, num_tokens, num_heads, head_dim)
-        keys = keys.view(b, num_tokens, self.num_heads, self.head_dim)
-        values = values.view(b, num_tokens, self.num_heads, self.head_dim)
-        queries = queries.view(b, num_tokens, self.num_heads, self.head_dim)
-
-        # Transpose: (b, num_tokens, num_heads, head_dim) -> (b, num_heads, num_tokens, head_dim)
-        keys = keys.transpose(1, 2)
-        queries = queries.transpose(1, 2)
-        values = values.transpose(1, 2)
-
-        # Compute scaled dot-product attention (aka self-attention) with a causal mask
-        attn_scores = queries @ keys.transpose(2, 3)  # Dot product for each head
-
-        # Original mask truncated to the number of tokens and converted to boolean
-        mask_bool = self.mask.bool()[:num_tokens, :num_tokens]
-
-        # Use the mask to fill attention scores
-        attn_scores.masked_fill_(mask_bool, -torch.inf)
-
-        attn_weights = torch.softmax(attn_scores / keys.shape[-1]**0.5, dim=-1)
-        attn_weights = self.dropout(attn_weights)
-
-        # Shape: (b, num_tokens, num_heads, head_dim)
-        context_vec = (attn_weights @ values).transpose(1, 2)
-
-        # Combine heads, where self.d_out = self.num_heads * self.head_dim
-        context_vec = context_vec.reshape(b, num_tokens, self.d_out)
-        context_vec = self.out_proj(context_vec)  # optional projection
-
-        return context_vec
-
-
-#####################################
-# Chapter 4
-#####################################
-
-class LayerNorm(nn.Module):
-    def __init__(self, emb_dim):
-        super().__init__()
-        self.eps = 1e-5
-        self.scale = nn.Parameter(torch.ones(emb_dim))
-        self.shift = nn.Parameter(torch.zeros(emb_dim))
-
-    def forward(self, x):
-        mean = x.mean(dim=-1, keepdim=True)
-        var = x.var(dim=-1, keepdim=True, unbiased=False)
-        norm_x = (x - mean) / torch.sqrt(var + self.eps)
-        return self.scale * norm_x + self.shift
-
-
-class GELU(nn.Module):
-    def __init__(self):
-        super().__init__()
-
-    def forward(self, x):
-        return 0.5 * x * (1 + torch.tanh(
-            torch.sqrt(torch.tensor(2.0 / torch.pi)) *
-            (x + 0.044715 * torch.pow(x, 3))
-        ))
-
-
-class FeedForward(nn.Module):
-    def __init__(self, cfg):
-        super().__init__()
-        self.layers = nn.Sequential(
-            nn.Linear(cfg["emb_dim"], 4 * cfg["emb_dim"]),
-            GELU(),
-            nn.Linear(4 * cfg["emb_dim"], cfg["emb_dim"]),
-        )
-
-    def forward(self, x):
-        return self.layers(x)
-
-
-class TransformerBlock(nn.Module):
-    def __init__(self, cfg):
-        super().__init__()
-        self.att = MultiHeadAttention(
-            d_in=cfg["emb_dim"],
-            d_out=cfg["emb_dim"],
-            context_length=cfg["context_length"],
-            num_heads=cfg["n_heads"],
-            dropout=cfg["drop_rate"],
-            qkv_bias=cfg["qkv_bias"])
-        self.ff = FeedForward(cfg)
-        self.norm1 = LayerNorm(cfg["emb_dim"])
-        self.norm2 = LayerNorm(cfg["emb_dim"])
-        self.drop_shortcut = nn.Dropout(cfg["drop_rate"])
-
-    def forward(self, x):
-        # Shortcut connection for attention block
-        shortcut = x
-        x = self.norm1(x)
-        x = self.att(x)   # Shape [batch_size, num_tokens, emb_size]
-        x = self.drop_shortcut(x)
-        x = x + shortcut  # Add the original input back
-
-        # Shortcut connection for feed-forward block
-        shortcut = x
-        x = self.norm2(x)
-        x = self.ff(x)
-        x = self.drop_shortcut(x)
-        x = x + shortcut  # Add the original input back
-
-        return x
-
-
-class GPTModel(nn.Module):
-    def __init__(self, cfg):
-        super().__init__()
-        self.tok_emb = nn.Embedding(cfg["vocab_size"], cfg["emb_dim"])
-        self.pos_emb = nn.Embedding(cfg["context_length"], cfg["emb_dim"])
-        self.drop_emb = nn.Dropout(cfg["drop_rate"])
-
-        self.trf_blocks = nn.Sequential(
-            *[TransformerBlock(cfg) for _ in range(cfg["n_layers"])])
-
-        self.final_norm = LayerNorm(cfg["emb_dim"])
-        self.out_head = nn.Linear(cfg["emb_dim"], cfg["vocab_size"], bias=False)
-
-    def forward(self, in_idx):
-        batch_size, seq_len = in_idx.shape
-        tok_embeds = self.tok_emb(in_idx)
-        pos_embeds = self.pos_emb(torch.arange(seq_len, device=in_idx.device))
-        x = tok_embeds + pos_embeds  # Shape [batch_size, num_tokens, emb_size]
-        x = self.drop_emb(x)
-        x = self.trf_blocks(x)
-        x = self.final_norm(x)
-        logits = self.out_head(x)
-        return logits
-
-
-def generate_text_simple(model, idx, max_new_tokens, context_size):
-    # idx is (B, T) array of indices in the current context
-    for _ in range(max_new_tokens):
-
-        # Crop current context if it exceeds the supported context size
-        # E.g., if LLM supports only 5 tokens, and the context size is 10
-        # then only the last 5 tokens are used as context
-        idx_cond = idx[:, -context_size:]
-
-        # Get the predictions
-        with torch.no_grad():
-            logits = model(idx_cond)
-
-        # Focus only on the last time step
-        # (batch, n_token, vocab_size) becomes (batch, vocab_size)
-        logits = logits[:, -1, :]
-
-        # Get the idx of the vocab entry with the highest logits value
-        idx_next = torch.argmax(logits, dim=-1, keepdim=True)  # (batch, 1)
-
-        # Append sampled index to the running sequence
-        idx = torch.cat((idx, idx_next), dim=1)  # (batch, n_tokens+1)
-
-    return idx
-
-
-#####################################
-# Chapter 5
-####################################
-
-
-def calc_loss_batch(input_batch, target_batch, model, device):
-    input_batch, target_batch = input_batch.to(device), target_batch.to(device)
-    logits = model(input_batch)
-    loss = torch.nn.functional.cross_entropy(logits.flatten(0, 1), target_batch.flatten())
-    return loss
-
-
-def calc_loss_loader(data_loader, model, device, num_batches=None):
-    total_loss = 0.
-    if len(data_loader) == 0:
-        return float("nan")
-    elif num_batches is None:
-        num_batches = len(data_loader)
-    else:
-        num_batches = min(num_batches, len(data_loader))
-    for i, (input_batch, target_batch) in enumerate(data_loader):
-        if i < num_batches:
-            loss = calc_loss_batch(input_batch, target_batch, model, device)
-            total_loss += loss.item()
-        else:
-            break
-    return total_loss / num_batches
-
-
-def evaluate_model(model, train_loader, val_loader, device, eval_iter):
-    model.eval()
-    with torch.no_grad():
-        train_loss = calc_loss_loader(train_loader, model, device, num_batches=eval_iter)
-        val_loss = calc_loss_loader(val_loader, model, device, num_batches=eval_iter)
-    model.train()
-    return train_loss, val_loss
-
-
-def generate_and_print_sample(model, tokenizer, device, start_context):
-    model.eval()
-    context_size = model.pos_emb.weight.shape[0]
-    encoded = text_to_token_ids(start_context, tokenizer).to(device)
-    with torch.no_grad():
-        token_ids = generate_text_simple(
-            model=model, idx=encoded,
-            max_new_tokens=50, context_size=context_size)
-        decoded_text = token_ids_to_text(token_ids, tokenizer)
-        print(decoded_text.replace("\n", " "))  # Compact print format
-    model.train()
-
-
-def plot_losses(epochs_seen, tokens_seen, train_losses, val_losses, output_dir):
-    fig, ax1 = plt.subplots()
-
-    # Plot training and validation loss against epochs
-    ax1.plot(epochs_seen, train_losses, label="Training loss")
-    ax1.plot(epochs_seen, val_losses, linestyle="-.", label="Validation loss")
-    ax1.set_xlabel("Epochs")
-    ax1.set_ylabel("Loss")
-    ax1.legend(loc="upper right")
-    ax1.xaxis.set_major_locator(MaxNLocator(integer=True))
-
-    # Create a second x-axis for tokens seen
-    ax2 = ax1.twiny()  # Create a second x-axis that shares the same y-axis
-    ax2.plot(tokens_seen, train_losses, alpha=0)  # Invisible plot for aligning ticks
-    ax2.set_xlabel("Tokens seen")
-
-    fig.tight_layout()  # Adjust layout to make room
-    plt.savefig(output_dir / "losses.pdf")
-
-
-def text_to_token_ids(text, tokenizer):
-    encoded = tokenizer.encode(text, allowed_special={'<|endoftext|>'})
-    encoded_tensor = torch.tensor(encoded).unsqueeze(0)  # Add batch dimension
-    return encoded_tensor
-
-
-def token_ids_to_text(token_ids, tokenizer):
-    flat = token_ids.squeeze(0)  # Remove batch dimension
-    return tokenizer.decode(flat.tolist())

ch05/05_bonus_hparam_tuning/hparam_search.py

@@ -8,7 +8,11 @@ import math
 import os
 import tiktoken
 import torch
-from previous_chapters import GPTModel, create_dataloader_v1
+
+# For llms_from_scratch installation instructions, see:
+# https://github.com/rasbt/LLMs-from-scratch/tree/main/pkg
+from llms_from_scratch.ch02 import create_dataloader_v1
+from llms_from_scratch.ch04 import GPTModel
 
 
 # Define a grid of hyperparameters to search over

ch05/05_bonus_hparam_tuning/previous_chapters.py

@@ -1,279 +0,0 @@
-# Copyright (c) Sebastian Raschka under Apache License 2.0 (see LICENSE.txt).
-# Source for "Build a Large Language Model From Scratch"
-#   - https://www.manning.com/books/build-a-large-language-model-from-scratch
-# Code: https://github.com/rasbt/LLMs-from-scratch
-
-# This file collects all the relevant code that we covered thus far
-# throughout Chapters 2-4.
-# This file can be run as a standalone script.
-
-import tiktoken
-import torch
-import torch.nn as nn
-from torch.utils.data import Dataset, DataLoader
-
-#####################################
-# Chapter 2
-#####################################
-
-
-class GPTDatasetV1(Dataset):
-    def __init__(self, txt, tokenizer, max_length, stride):
-        self.input_ids = []
-        self.target_ids = []
-
-        # Tokenize the entire text
-        token_ids = tokenizer.encode(txt, allowed_special={"<|endoftext|>"})
-
-        # Use a sliding window to chunk the book into overlapping sequences of max_length
-        for i in range(0, len(token_ids) - max_length, stride):
-            input_chunk = token_ids[i:i + max_length]
-            target_chunk = token_ids[i + 1: i + max_length + 1]
-            self.input_ids.append(torch.tensor(input_chunk))
-            self.target_ids.append(torch.tensor(target_chunk))
-
-    def __len__(self):
-        return len(self.input_ids)
-
-    def __getitem__(self, idx):
-        return self.input_ids[idx], self.target_ids[idx]
-
-
-def create_dataloader_v1(txt, batch_size=4, max_length=256,
-                         stride=128, shuffle=True, drop_last=True, num_workers=0):
-    # Initialize the tokenizer
-    tokenizer = tiktoken.get_encoding("gpt2")
-
-    # Create dataset
-    dataset = GPTDatasetV1(txt, tokenizer, max_length, stride)
-
-    # Create dataloader
-    dataloader = DataLoader(
-        dataset, batch_size=batch_size, shuffle=shuffle, drop_last=drop_last, num_workers=num_workers)
-
-    return dataloader
-
-
-#####################################
-# Chapter 3
-#####################################
-class MultiHeadAttention(nn.Module):
-    def __init__(self, d_in, d_out, context_length, dropout, num_heads, qkv_bias=False):
-        super().__init__()
-        assert d_out % num_heads == 0, "d_out must be divisible by num_heads"
-
-        self.d_out = d_out
-        self.num_heads = num_heads
-        self.head_dim = d_out // num_heads  # Reduce the projection dim to match desired output dim
-
-        self.W_query = nn.Linear(d_in, d_out, bias=qkv_bias)
-        self.W_key = nn.Linear(d_in, d_out, bias=qkv_bias)
-        self.W_value = nn.Linear(d_in, d_out, bias=qkv_bias)
-        self.out_proj = nn.Linear(d_out, d_out)  # Linear layer to combine head outputs
-        self.dropout = nn.Dropout(dropout)
-        self.register_buffer('mask', torch.triu(torch.ones(context_length, context_length), diagonal=1))
-
-    def forward(self, x):
-        b, num_tokens, d_in = x.shape
-
-        keys = self.W_key(x)  # Shape: (b, num_tokens, d_out)
-        queries = self.W_query(x)
-        values = self.W_value(x)
-
-        # We implicitly split the matrix by adding a `num_heads` dimension
-        # Unroll last dim: (b, num_tokens, d_out) -> (b, num_tokens, num_heads, head_dim)
-        keys = keys.view(b, num_tokens, self.num_heads, self.head_dim)
-        values = values.view(b, num_tokens, self.num_heads, self.head_dim)
-        queries = queries.view(b, num_tokens, self.num_heads, self.head_dim)
-
-        # Transpose: (b, num_tokens, num_heads, head_dim) -> (b, num_heads, num_tokens, head_dim)
-        keys = keys.transpose(1, 2)
-        queries = queries.transpose(1, 2)
-        values = values.transpose(1, 2)
-
-        # Compute scaled dot-product attention (aka self-attention) with a causal mask
-        attn_scores = queries @ keys.transpose(2, 3)  # Dot product for each head
-
-        # Original mask truncated to the number of tokens and converted to boolean
-        mask_bool = self.mask.bool()[:num_tokens, :num_tokens]
-
-        # Use the mask to fill attention scores
-        attn_scores.masked_fill_(mask_bool, -torch.inf)
-
-        attn_weights = torch.softmax(attn_scores / keys.shape[-1]**0.5, dim=-1)
-        attn_weights = self.dropout(attn_weights)
-
-        # Shape: (b, num_tokens, num_heads, head_dim)
-        context_vec = (attn_weights @ values).transpose(1, 2)
-
-        # Combine heads, where self.d_out = self.num_heads * self.head_dim
-        context_vec = context_vec.contiguous().view(b, num_tokens, self.d_out)
-        context_vec = self.out_proj(context_vec)  # optional projection
-
-        return context_vec
-
-
-#####################################
-# Chapter 4
-#####################################
-class LayerNorm(nn.Module):
-    def __init__(self, emb_dim):
-        super().__init__()
-        self.eps = 1e-5
-        self.scale = nn.Parameter(torch.ones(emb_dim))
-        self.shift = nn.Parameter(torch.zeros(emb_dim))
-
-    def forward(self, x):
-        mean = x.mean(dim=-1, keepdim=True)
-        var = x.var(dim=-1, keepdim=True, unbiased=False)
-        norm_x = (x - mean) / torch.sqrt(var + self.eps)
-        return self.scale * norm_x + self.shift
-
-
-class GELU(nn.Module):
-    def __init__(self):
-        super().__init__()
-
-    def forward(self, x):
-        return 0.5 * x * (1 + torch.tanh(
-            torch.sqrt(torch.tensor(2.0 / torch.pi)) *
-            (x + 0.044715 * torch.pow(x, 3))
-        ))
-
-
-class FeedForward(nn.Module):
-    def __init__(self, cfg):
-        super().__init__()
-        self.layers = nn.Sequential(
-            nn.Linear(cfg["emb_dim"], 4 * cfg["emb_dim"]),
-            GELU(),
-            nn.Linear(4 * cfg["emb_dim"], cfg["emb_dim"]),
-        )
-
-    def forward(self, x):
-        return self.layers(x)
-
-
-class TransformerBlock(nn.Module):
-    def __init__(self, cfg):
-        super().__init__()
-        self.att = MultiHeadAttention(
-            d_in=cfg["emb_dim"],
-            d_out=cfg["emb_dim"],
-            context_length=cfg["context_length"],
-            num_heads=cfg["n_heads"],
-            dropout=cfg["drop_rate"],
-            qkv_bias=cfg["qkv_bias"])
-        self.ff = FeedForward(cfg)
-        self.norm1 = LayerNorm(cfg["emb_dim"])
-        self.norm2 = LayerNorm(cfg["emb_dim"])
-        self.drop_shortcut = nn.Dropout(cfg["drop_rate"])
-
-    def forward(self, x):
-        # Shortcut connection for attention block
-        shortcut = x
-        x = self.norm1(x)
-        x = self.att(x)   # Shape [batch_size, num_tokens, emb_size]
-        x = self.drop_shortcut(x)
-        x = x + shortcut  # Add the original input back
-
-        # Shortcut connection for feed-forward block
-        shortcut = x
-        x = self.norm2(x)
-        x = self.ff(x)
-        x = self.drop_shortcut(x)
-        x = x + shortcut  # Add the original input back
-
-        return x
-
-
-class GPTModel(nn.Module):
-    def __init__(self, cfg):
-        super().__init__()
-        self.tok_emb = nn.Embedding(cfg["vocab_size"], cfg["emb_dim"])
-        self.pos_emb = nn.Embedding(cfg["context_length"], cfg["emb_dim"])
-        self.drop_emb = nn.Dropout(cfg["drop_rate"])
-
-        self.trf_blocks = nn.Sequential(
-            *[TransformerBlock(cfg) for _ in range(cfg["n_layers"])])
-
-        self.final_norm = LayerNorm(cfg["emb_dim"])
-        self.out_head = nn.Linear(cfg["emb_dim"], cfg["vocab_size"], bias=False)
-
-    def forward(self, in_idx):
-        batch_size, seq_len = in_idx.shape
-        tok_embeds = self.tok_emb(in_idx)
-        pos_embeds = self.pos_emb(torch.arange(seq_len, device=in_idx.device))
-        x = tok_embeds + pos_embeds  # Shape [batch_size, num_tokens, emb_size]
-        x = self.drop_emb(x)
-        x = self.trf_blocks(x)
-        x = self.final_norm(x)
-        logits = self.out_head(x)
-        return logits
-
-
-def generate_text_simple(model, idx, max_new_tokens, context_size):
-    # idx is (B, T) array of indices in the current context
-    for _ in range(max_new_tokens):
-
-        # Crop current context if it exceeds the supported context size
-        # E.g., if LLM supports only 5 tokens, and the context size is 10
-        # then only the last 5 tokens are used as context
-        idx_cond = idx[:, -context_size:]
-
-        # Get the predictions
-        with torch.no_grad():
-            logits = model(idx_cond)
-
-        # Focus only on the last time step
-        # (batch, n_token, vocab_size) becomes (batch, vocab_size)
-        logits = logits[:, -1, :]
-
-        # Get the idx of the vocab entry with the highest logits value
-        idx_next = torch.argmax(logits, dim=-1, keepdim=True)  # (batch, 1)
-
-        # Append sampled index to the running sequence
-        idx = torch.cat((idx, idx_next), dim=1)  # (batch, n_tokens+1)
-
-    return idx
-
-
-if __name__ == "__main__":
-
-    GPT_CONFIG_124M = {
-        "vocab_size": 50257,     # Vocabulary size
-        "context_length": 1024,  # Context length
-        "emb_dim": 768,          # Embedding dimension
-        "n_heads": 12,           # Number of attention heads
-        "n_layers": 12,          # Number of layers
-        "drop_rate": 0.1,        # Dropout rate
-        "qkv_bias": False        # Query-Key-Value bias
-    }
-
-    torch.manual_seed(123)
-    model = GPTModel(GPT_CONFIG_124M)
-    model.eval()  # disable dropout
-
-    start_context = "Hello, I am"
-
-    tokenizer = tiktoken.get_encoding("gpt2")
-    encoded = tokenizer.encode(start_context)
-    encoded_tensor = torch.tensor(encoded).unsqueeze(0)
-
-    print(f"\n{50*'='}\n{22*' '}IN\n{50*'='}")
-    print("\nInput text:", start_context)
-    print("Encoded input text:", encoded)
-    print("encoded_tensor.shape:", encoded_tensor.shape)
-
-    out = generate_text_simple(
-        model=model,
-        idx=encoded_tensor,
-        max_new_tokens=10,
-        context_size=GPT_CONFIG_124M["context_length"]
-    )
-    decoded_text = tokenizer.decode(out.squeeze(0).tolist())
-
-    print(f"\n\n{50*'='}\n{22*' '}OUT\n{50*'='}")
-    print("\nOutput:", out)
-    print("Output length:", len(out[0]))
-    print("Output text:", decoded_text)

ch05/06_user_interface/app_orig.py

@@ -7,10 +7,12 @@ import tiktoken
 import torch
 import chainlit
 
-from previous_chapters import (
+# For llms_from_scratch installation instructions, see:
+# https://github.com/rasbt/LLMs-from-scratch/tree/main/pkg
+from llms_from_scratch.ch04 import GPTModel
+from llms_from_scratch.ch05 import (
     download_and_load_gpt2,
     generate,
-    GPTModel,
     load_weights_into_gpt,
     text_to_token_ids,
     token_ids_to_text,

ch05/06_user_interface/app_own.py

@@ -10,13 +10,16 @@ import tiktoken
 import torch
 import chainlit
 
-from previous_chapters import (
+# For llms_from_scratch installation instructions, see:
+# https://github.com/rasbt/LLMs-from-scratch/tree/main/pkg
+from llms_from_scratch.ch04 import GPTModel
+from llms_from_scratch.ch05 import (
     generate,
-    GPTModel,
     text_to_token_ids,
     token_ids_to_text,
 )
 
+
 device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
 
 

ch05/06_user_interface/previous_chapters.py

@@ -1,384 +0,0 @@
-# Copyright (c) Sebastian Raschka under Apache License 2.0 (see LICENSE.txt).
-# Source for "Build a Large Language Model From Scratch"
-#   - https://www.manning.com/books/build-a-large-language-model-from-scratch
-# Code: https://github.com/rasbt/LLMs-from-scratch
-#
-# This file collects all the relevant code that we covered thus far
-# throughout Chapters 2-5.
-
-import json
-import os
-import urllib
-
-import numpy as np
-import tensorflow as tf
-import torch
-import torch.nn as nn
-from tqdm import tqdm
-
-
-#####################################
-# Chapter 3
-#####################################
-class MultiHeadAttention(nn.Module):
-    def __init__(self, d_in, d_out, context_length, dropout, num_heads, qkv_bias=False):
-        super().__init__()
-        assert d_out % num_heads == 0, "d_out must be divisible by n_heads"
-
-        self.d_out = d_out
-        self.num_heads = num_heads
-        self.head_dim = d_out // num_heads  # Reduce the projection dim to match desired output dim
-
-        self.W_query = nn.Linear(d_in, d_out, bias=qkv_bias)
-        self.W_key = nn.Linear(d_in, d_out, bias=qkv_bias)
-        self.W_value = nn.Linear(d_in, d_out, bias=qkv_bias)
-        self.out_proj = nn.Linear(d_out, d_out)  # Linear layer to combine head outputs
-        self.dropout = nn.Dropout(dropout)
-        self.register_buffer('mask', torch.triu(torch.ones(context_length, context_length), diagonal=1))
-
-    def forward(self, x):
-        b, num_tokens, d_in = x.shape
-
-        keys = self.W_key(x)  # Shape: (b, num_tokens, d_out)
-        queries = self.W_query(x)
-        values = self.W_value(x)
-
-        # We implicitly split the matrix by adding a `num_heads` dimension
-        # Unroll last dim: (b, num_tokens, d_out) -> (b, num_tokens, num_heads, head_dim)
-        keys = keys.view(b, num_tokens, self.num_heads, self.head_dim)
-        values = values.view(b, num_tokens, self.num_heads, self.head_dim)
-        queries = queries.view(b, num_tokens, self.num_heads, self.head_dim)
-
-        # Transpose: (b, num_tokens, num_heads, head_dim) -> (b, num_heads, num_tokens, head_dim)
-        keys = keys.transpose(1, 2)
-        queries = queries.transpose(1, 2)
-        values = values.transpose(1, 2)
-
-        # Compute scaled dot-product attention (aka self-attention) with a causal mask
-        attn_scores = queries @ keys.transpose(2, 3)  # Dot product for each head
-
-        # Original mask truncated to the number of tokens and converted to boolean
-        mask_bool = self.mask.bool()[:num_tokens, :num_tokens]
-
-        # Use the mask to fill attention scores
-        attn_scores.masked_fill_(mask_bool, -torch.inf)
-
-        attn_weights = torch.softmax(attn_scores / keys.shape[-1]**0.5, dim=-1)
-        attn_weights = self.dropout(attn_weights)
-
-        # Shape: (b, num_tokens, num_heads, head_dim)
-        context_vec = (attn_weights @ values).transpose(1, 2)
-
-        # Combine heads, where self.d_out = self.num_heads * self.head_dim
-        context_vec = context_vec.reshape(b, num_tokens, self.d_out)
-        context_vec = self.out_proj(context_vec)  # optional projection
-
-        return context_vec
-
-
-#####################################
-# Chapter 4
-#####################################
-class LayerNorm(nn.Module):
-    def __init__(self, emb_dim):
-        super().__init__()
-        self.eps = 1e-5
-        self.scale = nn.Parameter(torch.ones(emb_dim))
-        self.shift = nn.Parameter(torch.zeros(emb_dim))
-
-    def forward(self, x):
-        mean = x.mean(dim=-1, keepdim=True)
-        var = x.var(dim=-1, keepdim=True, unbiased=False)
-        norm_x = (x - mean) / torch.sqrt(var + self.eps)
-        return self.scale * norm_x + self.shift
-
-
-class GELU(nn.Module):
-    def __init__(self):
-        super().__init__()
-
-    def forward(self, x):
-        return 0.5 * x * (1 + torch.tanh(
-            torch.sqrt(torch.tensor(2.0 / torch.pi)) *
-            (x + 0.044715 * torch.pow(x, 3))
-        ))
-
-
-class FeedForward(nn.Module):
-    def __init__(self, cfg):
-        super().__init__()
-        self.layers = nn.Sequential(
-            nn.Linear(cfg["emb_dim"], 4 * cfg["emb_dim"]),
-            GELU(),
-            nn.Linear(4 * cfg["emb_dim"], cfg["emb_dim"]),
-        )
-
-    def forward(self, x):
-        return self.layers(x)
-
-
-class TransformerBlock(nn.Module):
-    def __init__(self, cfg):
-        super().__init__()
-        self.att = MultiHeadAttention(
-            d_in=cfg["emb_dim"],
-            d_out=cfg["emb_dim"],
-            context_length=cfg["context_length"],
-            num_heads=cfg["n_heads"],
-            dropout=cfg["drop_rate"],
-            qkv_bias=cfg["qkv_bias"])
-        self.ff = FeedForward(cfg)
-        self.norm1 = LayerNorm(cfg["emb_dim"])
-        self.norm2 = LayerNorm(cfg["emb_dim"])
-        self.drop_shortcut = nn.Dropout(cfg["drop_rate"])
-
-    def forward(self, x):
-        # Shortcut connection for attention block
-        shortcut = x
-        x = self.norm1(x)
-        x = self.att(x)   # Shape [batch_size, num_tokens, emb_size]
-        x = self.drop_shortcut(x)
-        x = x + shortcut  # Add the original input back
-
-        # Shortcut connection for feed-forward block
-        shortcut = x
-        x = self.norm2(x)
-        x = self.ff(x)
-        x = self.drop_shortcut(x)
-        x = x + shortcut  # Add the original input back
-
-        return x
-
-
-class GPTModel(nn.Module):
-    def __init__(self, cfg):
-        super().__init__()
-        self.tok_emb = nn.Embedding(cfg["vocab_size"], cfg["emb_dim"])
-        self.pos_emb = nn.Embedding(cfg["context_length"], cfg["emb_dim"])
-        self.drop_emb = nn.Dropout(cfg["drop_rate"])
-
-        self.trf_blocks = nn.Sequential(
-            *[TransformerBlock(cfg) for _ in range(cfg["n_layers"])])
-
-        self.final_norm = LayerNorm(cfg["emb_dim"])
-        self.out_head = nn.Linear(cfg["emb_dim"], cfg["vocab_size"], bias=False)
-
-    def forward(self, in_idx):
-        batch_size, seq_len = in_idx.shape
-        tok_embeds = self.tok_emb(in_idx)
-        pos_embeds = self.pos_emb(torch.arange(seq_len, device=in_idx.device))
-        x = tok_embeds + pos_embeds  # Shape [batch_size, num_tokens, emb_size]
-        x = self.drop_emb(x)
-        x = self.trf_blocks(x)
-        x = self.final_norm(x)
-        logits = self.out_head(x)
-        return logits
-
-
-#####################################
-# Chapter 5
-#####################################
-def text_to_token_ids(text, tokenizer):
-    encoded = tokenizer.encode(text)
-    encoded_tensor = torch.tensor(encoded).unsqueeze(0)  # add batch dimension
-    return encoded_tensor
-
-
-def token_ids_to_text(token_ids, tokenizer):
-    flat = token_ids.squeeze(0)  # remove batch dimension
-    return tokenizer.decode(flat.tolist())
-
-
-def download_and_load_gpt2(model_size, models_dir):
-    # Validate model size
-    allowed_sizes = ("124M", "355M", "774M", "1558M")
-    if model_size not in allowed_sizes:
-        raise ValueError(f"Model size not in {allowed_sizes}")
-
-    # Define paths
-    model_dir = os.path.join(models_dir, model_size)
-    base_url = "https://openaipublic.blob.core.windows.net/gpt-2/models"
-    filenames = [
-        "checkpoint", "encoder.json", "hparams.json",
-        "model.ckpt.data-00000-of-00001", "model.ckpt.index",
-        "model.ckpt.meta", "vocab.bpe"
-    ]
-
-    # Download files
-    os.makedirs(model_dir, exist_ok=True)
-    for filename in filenames:
-        file_url = os.path.join(base_url, model_size, filename)
-        file_path = os.path.join(model_dir, filename)
-        download_file(file_url, file_path)
-
-    # Load settings and params
-    tf_ckpt_path = tf.train.latest_checkpoint(model_dir)
-    settings = json.load(open(os.path.join(model_dir, "hparams.json")))
-    params = load_gpt2_params_from_tf_ckpt(tf_ckpt_path, settings)
-
-    return settings, params
-
-
-def download_file(url, destination):
-    # Send a GET request to download the file
-    with urllib.request.urlopen(url) as response:
-        # Get the total file size from headers, defaulting to 0 if not present
-        file_size = int(response.headers.get("Content-Length", 0))
-
-        # Check if file exists and has the same size
-        if os.path.exists(destination):
-            file_size_local = os.path.getsize(destination)
-            if file_size == file_size_local:
-                print(f"File already exists and is up-to-date: {destination}")
-                return
-
-        # Define the block size for reading the file
-        block_size = 1024  # 1 Kilobyte
-
-        # Initialize the progress bar with total file size
-        progress_bar_description = os.path.basename(url)  # Extract filename from URL
-        with tqdm(total=file_size, unit="iB", unit_scale=True, desc=progress_bar_description) as progress_bar:
-            # Open the destination file in binary write mode
-            with open(destination, "wb") as file:
-                # Read the file in chunks and write to destination
-                while True:
-                    chunk = response.read(block_size)
-                    if not chunk:
-                        break
-                    file.write(chunk)
-                    progress_bar.update(len(chunk))  # Update progress bar
-
-
-def load_gpt2_params_from_tf_ckpt(ckpt_path, settings):
-    # Initialize parameters dictionary with empty blocks for each layer
-    params = {"blocks": [{} for _ in range(settings["n_layer"])]}
-
-    # Iterate over each variable in the checkpoint
-    for name, _ in tf.train.list_variables(ckpt_path):
-        # Load the variable and remove singleton dimensions
-        variable_array = np.squeeze(tf.train.load_variable(ckpt_path, name))
-
-        # Process the variable name to extract relevant parts
-        variable_name_parts = name.split("/")[1:]  # Skip the 'model/' prefix
-
-        # Identify the target dictionary for the variable
-        target_dict = params
-        if variable_name_parts[0].startswith("h"):
-            layer_number = int(variable_name_parts[0][1:])
-            target_dict = params["blocks"][layer_number]
-
-        # Recursively access or create nested dictionaries
-        for key in variable_name_parts[1:-1]:
-            target_dict = target_dict.setdefault(key, {})
-
-        # Assign the variable array to the last key
-        last_key = variable_name_parts[-1]
-        target_dict[last_key] = variable_array
-
-    return params
-
-
-def assign(left, right):
-    if left.shape != right.shape:
-        raise ValueError(f"Shape mismatch. Left: {left.shape}, Right: {right.shape}")
-    return torch.nn.Parameter(torch.tensor(right))
-
-
-def load_weights_into_gpt(gpt, params):
-    gpt.pos_emb.weight = assign(gpt.pos_emb.weight, params['wpe'])
-    gpt.tok_emb.weight = assign(gpt.tok_emb.weight, params['wte'])
-
-    for b in range(len(params["blocks"])):
-        q_w, k_w, v_w = np.split(
-            (params["blocks"][b]["attn"]["c_attn"])["w"], 3, axis=-1)
-        gpt.trf_blocks[b].att.W_query.weight = assign(
-            gpt.trf_blocks[b].att.W_query.weight, q_w.T)
-        gpt.trf_blocks[b].att.W_key.weight = assign(
-            gpt.trf_blocks[b].att.W_key.weight, k_w.T)
-        gpt.trf_blocks[b].att.W_value.weight = assign(
-            gpt.trf_blocks[b].att.W_value.weight, v_w.T)
-
-        q_b, k_b, v_b = np.split(
-            (params["blocks"][b]["attn"]["c_attn"])["b"], 3, axis=-1)
-        gpt.trf_blocks[b].att.W_query.bias = assign(
-            gpt.trf_blocks[b].att.W_query.bias, q_b)
-        gpt.trf_blocks[b].att.W_key.bias = assign(
-            gpt.trf_blocks[b].att.W_key.bias, k_b)
-        gpt.trf_blocks[b].att.W_value.bias = assign(
-            gpt.trf_blocks[b].att.W_value.bias, v_b)
-
-        gpt.trf_blocks[b].att.out_proj.weight = assign(
-            gpt.trf_blocks[b].att.out_proj.weight,
-            params["blocks"][b]["attn"]["c_proj"]["w"].T)
-        gpt.trf_blocks[b].att.out_proj.bias = assign(
-            gpt.trf_blocks[b].att.out_proj.bias,
-            params["blocks"][b]["attn"]["c_proj"]["b"])
-
-        gpt.trf_blocks[b].ff.layers[0].weight = assign(
-            gpt.trf_blocks[b].ff.layers[0].weight,
-            params["blocks"][b]["mlp"]["c_fc"]["w"].T)
-        gpt.trf_blocks[b].ff.layers[0].bias = assign(
-            gpt.trf_blocks[b].ff.layers[0].bias,
-            params["blocks"][b]["mlp"]["c_fc"]["b"])
-        gpt.trf_blocks[b].ff.layers[2].weight = assign(
-            gpt.trf_blocks[b].ff.layers[2].weight,
-            params["blocks"][b]["mlp"]["c_proj"]["w"].T)
-        gpt.trf_blocks[b].ff.layers[2].bias = assign(
-            gpt.trf_blocks[b].ff.layers[2].bias,
-            params["blocks"][b]["mlp"]["c_proj"]["b"])
-
-        gpt.trf_blocks[b].norm1.scale = assign(
-            gpt.trf_blocks[b].norm1.scale,
-            params["blocks"][b]["ln_1"]["g"])
-        gpt.trf_blocks[b].norm1.shift = assign(
-            gpt.trf_blocks[b].norm1.shift,
-            params["blocks"][b]["ln_1"]["b"])
-        gpt.trf_blocks[b].norm2.scale = assign(
-            gpt.trf_blocks[b].norm2.scale,
-            params["blocks"][b]["ln_2"]["g"])
-        gpt.trf_blocks[b].norm2.shift = assign(
-            gpt.trf_blocks[b].norm2.shift,
-            params["blocks"][b]["ln_2"]["b"])
-
-    gpt.final_norm.scale = assign(gpt.final_norm.scale, params["g"])
-    gpt.final_norm.shift = assign(gpt.final_norm.shift, params["b"])
-    gpt.out_head.weight = assign(gpt.out_head.weight, params["wte"])
-
-
-def generate(model, idx, max_new_tokens, context_size, temperature=0.0, top_k=None, eos_id=None):
-
-    # For-loop is the same as before: Get logits, and only focus on last time step
-    for _ in range(max_new_tokens):
-        idx_cond = idx[:, -context_size:]
-        with torch.no_grad():
-            logits = model(idx_cond)
-        logits = logits[:, -1, :]
-
-        # New: Filter logits with top_k sampling
-        if top_k is not None:
-            # Keep only top_k values
-            top_logits, _ = torch.topk(logits, top_k)
-            min_val = top_logits[:, -1]
-            logits = torch.where(logits < min_val, torch.tensor(float('-inf')).to(logits.device), logits)
-
-        # New: Apply temperature scaling
-        if temperature > 0.0:
-            logits = logits / temperature
-
-            # Apply softmax to get probabilities
-            probs = torch.softmax(logits, dim=-1)  # (batch_size, context_len)
-
-            # Sample from the distribution
-            idx_next = torch.multinomial(probs, num_samples=1)  # (batch_size, 1)
-
-        # Otherwise same as before: get idx of the vocab entry with the highest logits value
-        else:
-            idx_next = torch.argmax(logits, dim=-1, keepdim=True)  # (batch_size, 1)
-
-        if idx_next == eos_id:  # Stop generating early if end-of-sequence token is encountered and eos_id is specified
-            break
-
-        # Same as before: append sampled index to the running sequence
-        idx = torch.cat((idx, idx_next), dim=1)  # (batch_size, num_tokens+1)
-
-    return idx

ch05/07_gpt_to_llama/converting-gpt-to-llama2.ipynb

@@ -1264,6 +1264,12 @@
    ],
    "source": [
     "from previous_chapters import generate, text_to_token_ids, token_ids_to_text\n",
+    "# If the `previous_chapters.py` file is not available locally,\n",
+    "# you can import it from the `llms-from-scratch` PyPI package.\n",
+    "# For details, see: https://github.com/rasbt/LLMs-from-scratch/tree/main/pkg\n",
+    "# E.g.,\n",
+    "# from llms_from_scratch.ch05 import generate, text_to_token_ids, token_ids_to_text\n",
+    "\n",
     "\n",
     "\n",
     "torch.manual_seed(123)\n",
@@ -1691,7 +1697,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.11.4"
+   "version": "3.10.16"
   },
   "widgets": {
    "application/vnd.jupyter.widget-state+json": {

ch05/07_gpt_to_llama/converting-llama2-to-llama3.ipynb

@@ -1324,6 +1324,11 @@
    ],
    "source": [
     "from previous_chapters import generate, text_to_token_ids, token_ids_to_text\n",
+    "# If the `previous_chapters.py` file is not available locally,\n",
+    "# you can import it from the `llms-from-scratch` PyPI package.\n",
+    "# For details, see: https://github.com/rasbt/LLMs-from-scratch/tree/main/pkg\n",
+    "# E.g.,\n",
+    "# from llms_from_scratch.ch05 import generate, text_to_token_ids, token_ids_to_text\n",
     "\n",
     "\n",
     "torch.manual_seed(123)\n",

ch05/08_memory_efficient_weight_loading/memory-efficient-state-dict.ipynb

@@ -161,6 +161,12 @@
    "outputs": [],
    "source": [
     "from previous_chapters import GPTModel\n",
+    "# If the `previous_chapters.py` file is not available locally,\n",
+    "# you can import it from the `llms-from-scratch` PyPI package.\n",
+    "# For details, see: https://github.com/rasbt/LLMs-from-scratch/tree/main/pkg\n",
+    "# E.g.,\n",
+    "# from llms_from_scratch.ch04 import GPTModel\n",
+    "\n",
     "\n",
     "\n",
     "BASE_CONFIG = {\n",
@@ -921,7 +927,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.11.4"
+   "version": "3.10.16"
   }
  },
  "nbformat": 4,

ch05/09_extending-tokenizers/extend-tiktoken.ipynb

@@ -301,8 +301,9 @@
     }
    ],
    "source": [
-    "# Relative import from the gpt_download.py contained in this folder\n",
+    "from llms_from_scratch.ch05 import download_and_load_gpt2\n",
-    "from gpt_download import download_and_load_gpt2\n",
+    "# For llms_from_scratch installation instructions, see:\n",
+    "# https://github.com/rasbt/LLMs-from-scratch/tree/main/pkg\n",
     "\n",
     "settings, params = download_and_load_gpt2(model_size=\"124M\", models_dir=\"gpt2\")"
    ]
@@ -314,8 +315,9 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "# Relative import from the gpt_download.py contained in this folder\n",
+    "from llms_from_scratch.ch04 import GPTModel\n",
-    "from previous_chapters import GPTModel\n",
+    "# For llms_from_scratch installation instructions, see:\n",
+    "# https://github.com/rasbt/LLMs-from-scratch/tree/main/pkg\n",
     "\n",
     "GPT_CONFIG_124M = {\n",
     "    \"vocab_size\": 50257,   # Vocabulary size\n",
@@ -763,7 +765,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.11.4"
+   "version": "3.10.16"
   }
  },
  "nbformat": 4,

ch05/09_extending-tokenizers/gpt_download.py

@@ -1,157 +0,0 @@
-# Copyright (c) Sebastian Raschka under Apache License 2.0 (see LICENSE.txt).
-# Source for "Build a Large Language Model From Scratch"
-#   - https://www.manning.com/books/build-a-large-language-model-from-scratch
-# Code: https://github.com/rasbt/LLMs-from-scratch
-
-
-import os
-import urllib.request
-
-# import requests
-import json
-import numpy as np
-import tensorflow as tf
-from tqdm import tqdm
-
-
-def download_and_load_gpt2(model_size, models_dir):
-    # Validate model size
-    allowed_sizes = ("124M", "355M", "774M", "1558M")
-    if model_size not in allowed_sizes:
-        raise ValueError(f"Model size not in {allowed_sizes}")
-
-    # Define paths
-    model_dir = os.path.join(models_dir, model_size)
-    base_url = "https://openaipublic.blob.core.windows.net/gpt-2/models"
-    backup_base_url = "https://f001.backblazeb2.com/file/LLMs-from-scratch/gpt2"
-    filenames = [
-        "checkpoint", "encoder.json", "hparams.json",
-        "model.ckpt.data-00000-of-00001", "model.ckpt.index",
-        "model.ckpt.meta", "vocab.bpe"
-    ]
-
-    # Download files
-    os.makedirs(model_dir, exist_ok=True)
-    for filename in filenames:
-        file_url = os.path.join(base_url, model_size, filename)
-        backup_url = os.path.join(backup_base_url, model_size, filename)
-        file_path = os.path.join(model_dir, filename)
-        download_file(file_url, file_path, backup_url)
-
-    # Load settings and params
-    tf_ckpt_path = tf.train.latest_checkpoint(model_dir)
-    settings = json.load(open(os.path.join(model_dir, "hparams.json"), "r", encoding="utf-8"))
-    params = load_gpt2_params_from_tf_ckpt(tf_ckpt_path, settings)
-
-    return settings, params
-
-
-def download_file(url, destination, backup_url=None):
-    def _attempt_download(download_url):
-        with urllib.request.urlopen(download_url) as response:
-            # Get the total file size from headers, defaulting to 0 if not present
-            file_size = int(response.headers.get("Content-Length", 0))
-
-            # Check if file exists and has the same size
-            if os.path.exists(destination):
-                file_size_local = os.path.getsize(destination)
-                if file_size == file_size_local:
-                    print(f"File already exists and is up-to-date: {destination}")
-                    return True  # Indicate success without re-downloading
-
-            block_size = 1024  # 1 Kilobyte
-
-            # Initialize the progress bar with total file size
-            progress_bar_description = os.path.basename(download_url)
-            with tqdm(total=file_size, unit="iB", unit_scale=True, desc=progress_bar_description) as progress_bar:
-                with open(destination, "wb") as file:
-                    while True:
-                        chunk = response.read(block_size)
-                        if not chunk:
-                            break
-                        file.write(chunk)
-                        progress_bar.update(len(chunk))
-            return True
-
-    try:
-        if _attempt_download(url):
-            return
-    except (urllib.error.HTTPError, urllib.error.URLError):
-        if backup_url is not None:
-            print(f"Primary URL ({url}) failed. Attempting backup URL: {backup_url}")
-            try:
-                if _attempt_download(backup_url):
-                    return
-            except urllib.error.HTTPError:
-                pass
-
-        # If we reach here, both attempts have failed
-        error_message = (
-            f"Failed to download from both primary URL ({url})"
-            f"{' and backup URL (' + backup_url + ')' if backup_url else ''}."
-            "\nCheck your internet connection or the file availability.\n"
-            "For help, visit: https://github.com/rasbt/LLMs-from-scratch/discussions/273"
-        )
-        print(error_message)
-    except Exception as e:
-        print(f"An unexpected error occurred: {e}")
-
-
-# Alternative way using `requests`
-"""
-def download_file(url, destination):
-    # Send a GET request to download the file in streaming mode
-    response = requests.get(url, stream=True)
-
-    # Get the total file size from headers, defaulting to 0 if not present
-    file_size = int(response.headers.get("content-length", 0))
-
-    # Check if file exists and has the same size
-    if os.path.exists(destination):
-        file_size_local = os.path.getsize(destination)
-        if file_size == file_size_local:
-            print(f"File already exists and is up-to-date: {destination}")
-            return
-
-    # Define the block size for reading the file
-    block_size = 1024  # 1 Kilobyte
-
-    # Initialize the progress bar with total file size
-    progress_bar_description = url.split("/")[-1]  # Extract filename from URL
-    with tqdm(total=file_size, unit="iB", unit_scale=True, desc=progress_bar_description) as progress_bar:
-        # Open the destination file in binary write mode
-        with open(destination, "wb") as file:
-            # Iterate over the file data in chunks
-            for chunk in response.iter_content(block_size):
-                progress_bar.update(len(chunk))  # Update progress bar
-                file.write(chunk)  # Write the chunk to the file
-"""
-
-
-def load_gpt2_params_from_tf_ckpt(ckpt_path, settings):
-    # Initialize parameters dictionary with empty blocks for each layer
-    params = {"blocks": [{} for _ in range(settings["n_layer"])]}
-
-    # Iterate over each variable in the checkpoint
-    for name, _ in tf.train.list_variables(ckpt_path):
-        # Load the variable and remove singleton dimensions
-        variable_array = np.squeeze(tf.train.load_variable(ckpt_path, name))
-
-        # Process the variable name to extract relevant parts
-        variable_name_parts = name.split("/")[1:]  # Skip the 'model/' prefix
-
-        # Identify the target dictionary for the variable
-        target_dict = params
-        if variable_name_parts[0].startswith("h"):
-            layer_number = int(variable_name_parts[0][1:])
-            target_dict = params["blocks"][layer_number]
-
-        # Recursively access or create nested dictionaries
-        for key in variable_name_parts[1:-1]:
-            target_dict = target_dict.setdefault(key, {})
-
-        # Assign the variable array to the last key
-        last_key = variable_name_parts[-1]
-        target_dict[last_key] = variable_array
-
-    return params

ch05/09_extending-tokenizers/previous_chapters.py

@@ -1,279 +0,0 @@
-# Copyright (c) Sebastian Raschka under Apache License 2.0 (see LICENSE.txt).
-# Source for "Build a Large Language Model From Scratch"
-#   - https://www.manning.com/books/build-a-large-language-model-from-scratch
-# Code: https://github.com/rasbt/LLMs-from-scratch
-#
-# This file collects all the relevant code that we covered thus far
-# throughout Chapters 2-4.
-# This file can be run as a standalone script.
-
-import tiktoken
-import torch
-import torch.nn as nn
-from torch.utils.data import Dataset, DataLoader
-
-#####################################
-# Chapter 2
-#####################################
-
-
-class GPTDatasetV1(Dataset):
-    def __init__(self, txt, tokenizer, max_length, stride):
-        self.input_ids = []
-        self.target_ids = []
-
-        # Tokenize the entire text
-        token_ids = tokenizer.encode(txt, allowed_special={"<|endoftext|>"})
-
-        # Use a sliding window to chunk the book into overlapping sequences of max_length
-        for i in range(0, len(token_ids) - max_length, stride):
-            input_chunk = token_ids[i:i + max_length]
-            target_chunk = token_ids[i + 1: i + max_length + 1]
-            self.input_ids.append(torch.tensor(input_chunk))
-            self.target_ids.append(torch.tensor(target_chunk))
-
-    def __len__(self):
-        return len(self.input_ids)
-
-    def __getitem__(self, idx):
-        return self.input_ids[idx], self.target_ids[idx]
-
-
-def create_dataloader_v1(txt, batch_size=4, max_length=256,
-                         stride=128, shuffle=True, drop_last=True, num_workers=0):
-    # Initialize the tokenizer
-    tokenizer = tiktoken.get_encoding("gpt2")
-
-    # Create dataset
-    dataset = GPTDatasetV1(txt, tokenizer, max_length, stride)
-
-    # Create dataloader
-    dataloader = DataLoader(
-        dataset, batch_size=batch_size, shuffle=shuffle, drop_last=drop_last, num_workers=num_workers)
-
-    return dataloader
-
-
-#####################################
-# Chapter 3
-#####################################
-class MultiHeadAttention(nn.Module):
-    def __init__(self, d_in, d_out, context_length, dropout, num_heads, qkv_bias=False):
-        super().__init__()
-        assert d_out % num_heads == 0, "d_out must be divisible by n_heads"
-
-        self.d_out = d_out
-        self.num_heads = num_heads
-        self.head_dim = d_out // num_heads  # Reduce the projection dim to match desired output dim
-
-        self.W_query = nn.Linear(d_in, d_out, bias=qkv_bias)
-        self.W_key = nn.Linear(d_in, d_out, bias=qkv_bias)
-        self.W_value = nn.Linear(d_in, d_out, bias=qkv_bias)
-        self.out_proj = nn.Linear(d_out, d_out)  # Linear layer to combine head outputs
-        self.dropout = nn.Dropout(dropout)
-        self.register_buffer('mask', torch.triu(torch.ones(context_length, context_length), diagonal=1))
-
-    def forward(self, x):
-        b, num_tokens, d_in = x.shape
-
-        keys = self.W_key(x)  # Shape: (b, num_tokens, d_out)
-        queries = self.W_query(x)
-        values = self.W_value(x)
-
-        # We implicitly split the matrix by adding a `num_heads` dimension
-        # Unroll last dim: (b, num_tokens, d_out) -> (b, num_tokens, num_heads, head_dim)
-        keys = keys.view(b, num_tokens, self.num_heads, self.head_dim)
-        values = values.view(b, num_tokens, self.num_heads, self.head_dim)
-        queries = queries.view(b, num_tokens, self.num_heads, self.head_dim)
-
-        # Transpose: (b, num_tokens, num_heads, head_dim) -> (b, num_heads, num_tokens, head_dim)
-        keys = keys.transpose(1, 2)
-        queries = queries.transpose(1, 2)
-        values = values.transpose(1, 2)
-
-        # Compute scaled dot-product attention (aka self-attention) with a causal mask
-        attn_scores = queries @ keys.transpose(2, 3)  # Dot product for each head
-
-        # Original mask truncated to the number of tokens and converted to boolean
-        mask_bool = self.mask.bool()[:num_tokens, :num_tokens]
-
-        # Use the mask to fill attention scores
-        attn_scores.masked_fill_(mask_bool, -torch.inf)
-
-        attn_weights = torch.softmax(attn_scores / keys.shape[-1]**0.5, dim=-1)
-        attn_weights = self.dropout(attn_weights)
-
-        # Shape: (b, num_tokens, num_heads, head_dim)
-        context_vec = (attn_weights @ values).transpose(1, 2)
-
-        # Combine heads, where self.d_out = self.num_heads * self.head_dim
-        context_vec = context_vec.reshape(b, num_tokens, self.d_out)
-        context_vec = self.out_proj(context_vec)  # optional projection
-
-        return context_vec
-
-
-#####################################
-# Chapter 4
-#####################################
-class LayerNorm(nn.Module):
-    def __init__(self, emb_dim):
-        super().__init__()
-        self.eps = 1e-5
-        self.scale = nn.Parameter(torch.ones(emb_dim))
-        self.shift = nn.Parameter(torch.zeros(emb_dim))
-
-    def forward(self, x):
-        mean = x.mean(dim=-1, keepdim=True)
-        var = x.var(dim=-1, keepdim=True, unbiased=False)
-        norm_x = (x - mean) / torch.sqrt(var + self.eps)
-        return self.scale * norm_x + self.shift
-
-
-class GELU(nn.Module):
-    def __init__(self):
-        super().__init__()
-
-    def forward(self, x):
-        return 0.5 * x * (1 + torch.tanh(
-            torch.sqrt(torch.tensor(2.0 / torch.pi)) *
-            (x + 0.044715 * torch.pow(x, 3))
-        ))
-
-
-class FeedForward(nn.Module):
-    def __init__(self, cfg):
-        super().__init__()
-        self.layers = nn.Sequential(
-            nn.Linear(cfg["emb_dim"], 4 * cfg["emb_dim"]),
-            GELU(),
-            nn.Linear(4 * cfg["emb_dim"], cfg["emb_dim"]),
-        )
-
-    def forward(self, x):
-        return self.layers(x)
-
-
-class TransformerBlock(nn.Module):
-    def __init__(self, cfg):
-        super().__init__()
-        self.att = MultiHeadAttention(
-            d_in=cfg["emb_dim"],
-            d_out=cfg["emb_dim"],
-            context_length=cfg["context_length"],
-            num_heads=cfg["n_heads"],
-            dropout=cfg["drop_rate"],
-            qkv_bias=cfg["qkv_bias"])
-        self.ff = FeedForward(cfg)
-        self.norm1 = LayerNorm(cfg["emb_dim"])
-        self.norm2 = LayerNorm(cfg["emb_dim"])
-        self.drop_shortcut = nn.Dropout(cfg["drop_rate"])
-
-    def forward(self, x):
-        # Shortcut connection for attention block
-        shortcut = x
-        x = self.norm1(x)
-        x = self.att(x)   # Shape [batch_size, num_tokens, emb_size]
-        x = self.drop_shortcut(x)
-        x = x + shortcut  # Add the original input back
-
-        # Shortcut connection for feed-forward block
-        shortcut = x
-        x = self.norm2(x)
-        x = self.ff(x)
-        x = self.drop_shortcut(x)
-        x = x + shortcut  # Add the original input back
-
-        return x
-
-
-class GPTModel(nn.Module):
-    def __init__(self, cfg):
-        super().__init__()
-        self.tok_emb = nn.Embedding(cfg["vocab_size"], cfg["emb_dim"])
-        self.pos_emb = nn.Embedding(cfg["context_length"], cfg["emb_dim"])
-        self.drop_emb = nn.Dropout(cfg["drop_rate"])
-
-        self.trf_blocks = nn.Sequential(
-            *[TransformerBlock(cfg) for _ in range(cfg["n_layers"])])
-
-        self.final_norm = LayerNorm(cfg["emb_dim"])
-        self.out_head = nn.Linear(cfg["emb_dim"], cfg["vocab_size"], bias=False)
-
-    def forward(self, in_idx):
-        batch_size, seq_len = in_idx.shape
-        tok_embeds = self.tok_emb(in_idx)
-        pos_embeds = self.pos_emb(torch.arange(seq_len, device=in_idx.device))
-        x = tok_embeds + pos_embeds  # Shape [batch_size, num_tokens, emb_size]
-        x = self.drop_emb(x)
-        x = self.trf_blocks(x)
-        x = self.final_norm(x)
-        logits = self.out_head(x)
-        return logits
-
-
-def generate_text_simple(model, idx, max_new_tokens, context_size):
-    # idx is (B, T) array of indices in the current context
-    for _ in range(max_new_tokens):
-
-        # Crop current context if it exceeds the supported context size
-        # E.g., if LLM supports only 5 tokens, and the context size is 10
-        # then only the last 5 tokens are used as context
-        idx_cond = idx[:, -context_size:]
-
-        # Get the predictions
-        with torch.no_grad():
-            logits = model(idx_cond)
-
-        # Focus only on the last time step
-        # (batch, n_token, vocab_size) becomes (batch, vocab_size)
-        logits = logits[:, -1, :]
-
-        # Get the idx of the vocab entry with the highest logits value
-        idx_next = torch.argmax(logits, dim=-1, keepdim=True)  # (batch, 1)
-
-        # Append sampled index to the running sequence
-        idx = torch.cat((idx, idx_next), dim=1)  # (batch, n_tokens+1)
-
-    return idx
-
-
-if __name__ == "__main__":
-
-    GPT_CONFIG_124M = {
-        "vocab_size": 50257,     # Vocabulary size
-        "context_length": 1024,  # Context length
-        "emb_dim": 768,          # Embedding dimension
-        "n_heads": 12,           # Number of attention heads
-        "n_layers": 12,          # Number of layers
-        "drop_rate": 0.1,        # Dropout rate
-        "qkv_bias": False        # Query-Key-Value bias
-    }
-
-    torch.manual_seed(123)
-    model = GPTModel(GPT_CONFIG_124M)
-    model.eval()  # disable dropout
-
-    start_context = "Hello, I am"
-
-    tokenizer = tiktoken.get_encoding("gpt2")
-    encoded = tokenizer.encode(start_context)
-    encoded_tensor = torch.tensor(encoded).unsqueeze(0)
-
-    print(f"\n{50*'='}\n{22*' '}IN\n{50*'='}")
-    print("\nInput text:", start_context)
-    print("Encoded input text:", encoded)
-    print("encoded_tensor.shape:", encoded_tensor.shape)
-
-    out = generate_text_simple(
-        model=model,
-        idx=encoded_tensor,
-        max_new_tokens=10,
-        context_size=GPT_CONFIG_124M["context_length"]
-    )
-    decoded_text = tokenizer.decode(out.squeeze(0).tolist())
-
-    print(f"\n\n{50*'='}\n{22*' '}OUT\n{50*'='}")
-    print("\nOutput:", out)
-    print("Output length:", len(out[0]))
-    print("Output text:", decoded_text)

ch06/01_main-chapter-code/ch06.ipynb

@@ -1035,6 +1035,12 @@
    "source": [
     "from gpt_download import download_and_load_gpt2\n",
     "from previous_chapters import GPTModel, load_weights_into_gpt\n",
+    "# If the `previous_chapters.py` file is not available locally,\n",
+    "# you can import it from the `llms-from-scratch` PyPI package.\n",
+    "# For details, see: https://github.com/rasbt/LLMs-from-scratch/tree/main/pkg\n",
+    "# E.g.,\n",
+    "# from llms_from_scratch.ch04 import GPTModel\n",
+    "# from llms_from_scratch.ch05 import download_and_load_gpt2, load_weights_into_gpt\n",
     "\n",
     "model_size = CHOOSE_MODEL.split(\" \")[-1].lstrip(\"(\").rstrip(\")\")\n",
     "settings, params = download_and_load_gpt2(model_size=model_size, models_dir=\"gpt2\")\n",
@@ -1075,6 +1081,13 @@
     "    token_ids_to_text\n",
     ")\n",
     "\n",
+    "# Alternatively:\n",
+    "# from llms_from_scratch.ch05 import (\n",
+    "#    generate_text_simple,\n",
+    "#    text_to_token_ids,\n",
+    "#    token_ids_to_text\n",
+    "# )\n",
+    "\n",
     "\n",
     "text_1 = \"Every effort moves you\"\n",
     "\n",

ch06/02_bonus_additional-experiments/README.md

@@ -55,7 +55,7 @@ You can use the following code to reproduce the experiments:
 - Row 16: `python additional_experiments.py --trainable_token_pos "flexible"`
 - Row 17: `python additional_experiments.py --disable_causal_mask`
 - Row 18: `python additional_experiments.py --ignore_index 50256`
-- Row 19: `python additional_experiments.py --average embeddings`
+- Row 19: `python additional_experiments.py --average_embeddings`
 
 I've kept the LLM and dataset small on purpose, so you can run the training on a regular laptop like a MacBook Air M3 in about 15 minutes (for the default setting) in case you don't have access to a GPU.
 

ch06/02_bonus_additional-experiments/additional_experiments.py

@@ -21,6 +21,14 @@ from gpt_download import download_and_load_gpt2
 from previous_chapters import GPTModel, load_weights_into_gpt
 
 
+# If the `previous_chapters.py` file is not available locally,
+# you can import it from the `llms-from-scratch` PyPI package.
+# For details, see: https://github.com/rasbt/LLMs-from-scratch/tree/main/pkg
+# E.g.,
+# from llms_from_scratch.ch04 import GPTModel
+# from llms_from_scratch.ch05 import download_and_load_gpt2, load_weights_into_gpt
+
+
 class LoRALayer(torch.nn.Module):
     def __init__(self, in_dim, out_dim, rank, alpha):
         super().__init__()

ch06/04_user_interface/app.py

@@ -10,10 +10,11 @@ import tiktoken
 import torch
 import chainlit
 
-from previous_chapters import (
+# For llms_from_scratch installation instructions, see:
-    classify_review,
+# https://github.com/rasbt/LLMs-from-scratch/tree/main/pkg
-    GPTModel
+from llms_from_scratch.ch04 import GPTModel
-)
+from llms_from_scratch.ch06 import classify_review
+
 
 device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
 

ch06/04_user_interface/previous_chapters.py

@@ -1,371 +0,0 @@
-# Copyright (c) Sebastian Raschka under Apache License 2.0 (see LICENSE.txt).
-# Source for "Build a Large Language Model From Scratch"
-#   - https://www.manning.com/books/build-a-large-language-model-from-scratch
-# Code: https://github.com/rasbt/LLMs-from-scratch
-#
-# This file collects all the relevant code that we covered thus far
-# throughout Chapters 2-5.
-
-import json
-import os
-import urllib
-
-import numpy as np
-import tensorflow as tf
-import torch
-import torch.nn as nn
-from tqdm import tqdm
-
-
-#####################################
-# Chapter 3
-#####################################
-class MultiHeadAttention(nn.Module):
-    def __init__(self, d_in, d_out, context_length, dropout, num_heads, qkv_bias=False):
-        super().__init__()
-        assert d_out % num_heads == 0, "d_out must be divisible by n_heads"
-
-        self.d_out = d_out
-        self.num_heads = num_heads
-        self.head_dim = d_out // num_heads  # Reduce the projection dim to match desired output dim
-
-        self.W_query = nn.Linear(d_in, d_out, bias=qkv_bias)
-        self.W_key = nn.Linear(d_in, d_out, bias=qkv_bias)
-        self.W_value = nn.Linear(d_in, d_out, bias=qkv_bias)
-        self.out_proj = nn.Linear(d_out, d_out)  # Linear layer to combine head outputs
-        self.dropout = nn.Dropout(dropout)
-        self.register_buffer('mask', torch.triu(torch.ones(context_length, context_length), diagonal=1))
-
-    def forward(self, x):
-        b, num_tokens, d_in = x.shape
-
-        keys = self.W_key(x)  # Shape: (b, num_tokens, d_out)
-        queries = self.W_query(x)
-        values = self.W_value(x)
-
-        # We implicitly split the matrix by adding a `num_heads` dimension
-        # Unroll last dim: (b, num_tokens, d_out) -> (b, num_tokens, num_heads, head_dim)
-        keys = keys.view(b, num_tokens, self.num_heads, self.head_dim)
-        values = values.view(b, num_tokens, self.num_heads, self.head_dim)
-        queries = queries.view(b, num_tokens, self.num_heads, self.head_dim)
-
-        # Transpose: (b, num_tokens, num_heads, head_dim) -> (b, num_heads, num_tokens, head_dim)
-        keys = keys.transpose(1, 2)
-        queries = queries.transpose(1, 2)
-        values = values.transpose(1, 2)
-
-        # Compute scaled dot-product attention (aka self-attention) with a causal mask
-        attn_scores = queries @ keys.transpose(2, 3)  # Dot product for each head
-
-        # Original mask truncated to the number of tokens and converted to boolean
-        mask_bool = self.mask.bool()[:num_tokens, :num_tokens]
-
-        # Use the mask to fill attention scores
-        attn_scores.masked_fill_(mask_bool, -torch.inf)
-
-        attn_weights = torch.softmax(attn_scores / keys.shape[-1]**0.5, dim=-1)
-        attn_weights = self.dropout(attn_weights)
-
-        # Shape: (b, num_tokens, num_heads, head_dim)
-        context_vec = (attn_weights @ values).transpose(1, 2)
-
-        # Combine heads, where self.d_out = self.num_heads * self.head_dim
-        context_vec = context_vec.reshape(b, num_tokens, self.d_out)
-        context_vec = self.out_proj(context_vec)  # optional projection
-
-        return context_vec
-
-
-#####################################
-# Chapter 4
-#####################################
-class LayerNorm(nn.Module):
-    def __init__(self, emb_dim):
-        super().__init__()
-        self.eps = 1e-5
-        self.scale = nn.Parameter(torch.ones(emb_dim))
-        self.shift = nn.Parameter(torch.zeros(emb_dim))
-
-    def forward(self, x):
-        mean = x.mean(dim=-1, keepdim=True)
-        var = x.var(dim=-1, keepdim=True, unbiased=False)
-        norm_x = (x - mean) / torch.sqrt(var + self.eps)
-        return self.scale * norm_x + self.shift
-
-
-class GELU(nn.Module):
-    def __init__(self):
-        super().__init__()
-
-    def forward(self, x):
-        return 0.5 * x * (1 + torch.tanh(
-            torch.sqrt(torch.tensor(2.0 / torch.pi)) *
-            (x + 0.044715 * torch.pow(x, 3))
-        ))
-
-
-class FeedForward(nn.Module):
-    def __init__(self, cfg):
-        super().__init__()
-        self.layers = nn.Sequential(
-            nn.Linear(cfg["emb_dim"], 4 * cfg["emb_dim"]),
-            GELU(),
-            nn.Linear(4 * cfg["emb_dim"], cfg["emb_dim"]),
-        )
-
-    def forward(self, x):
-        return self.layers(x)
-
-
-class TransformerBlock(nn.Module):
-    def __init__(self, cfg):
-        super().__init__()
-        self.att = MultiHeadAttention(
-            d_in=cfg["emb_dim"],
-            d_out=cfg["emb_dim"],
-            context_length=cfg["context_length"],
-            num_heads=cfg["n_heads"],
-            dropout=cfg["drop_rate"],
-            qkv_bias=cfg["qkv_bias"])
-        self.ff = FeedForward(cfg)
-        self.norm1 = LayerNorm(cfg["emb_dim"])
-        self.norm2 = LayerNorm(cfg["emb_dim"])
-        self.drop_shortcut = nn.Dropout(cfg["drop_rate"])
-
-    def forward(self, x):
-        # Shortcut connection for attention block
-        shortcut = x
-        x = self.norm1(x)
-        x = self.att(x)   # Shape [batch_size, num_tokens, emb_size]
-        x = self.drop_shortcut(x)
-        x = x + shortcut  # Add the original input back
-
-        # Shortcut connection for feed-forward block
-        shortcut = x
-        x = self.norm2(x)
-        x = self.ff(x)
-        x = self.drop_shortcut(x)
-        x = x + shortcut  # Add the original input back
-
-        return x
-
-
-class GPTModel(nn.Module):
-    def __init__(self, cfg):
-        super().__init__()
-        self.tok_emb = nn.Embedding(cfg["vocab_size"], cfg["emb_dim"])
-        self.pos_emb = nn.Embedding(cfg["context_length"], cfg["emb_dim"])
-        self.drop_emb = nn.Dropout(cfg["drop_rate"])
-
-        self.trf_blocks = nn.Sequential(
-            *[TransformerBlock(cfg) for _ in range(cfg["n_layers"])])
-
-        self.final_norm = LayerNorm(cfg["emb_dim"])
-        self.out_head = nn.Linear(cfg["emb_dim"], cfg["vocab_size"], bias=False)
-
-    def forward(self, in_idx):
-        batch_size, seq_len = in_idx.shape
-        tok_embeds = self.tok_emb(in_idx)
-        pos_embeds = self.pos_emb(torch.arange(seq_len, device=in_idx.device))
-        x = tok_embeds + pos_embeds  # Shape [batch_size, num_tokens, emb_size]
-        x = self.drop_emb(x)
-        x = self.trf_blocks(x)
-        x = self.final_norm(x)
-        logits = self.out_head(x)
-        return logits
-
-
-#####################################
-# Chapter 5
-#####################################
-def text_to_token_ids(text, tokenizer):
-    encoded = tokenizer.encode(text)
-    encoded_tensor = torch.tensor(encoded).unsqueeze(0)  # add batch dimension
-    return encoded_tensor
-
-
-def token_ids_to_text(token_ids, tokenizer):
-    flat = token_ids.squeeze(0)  # remove batch dimension
-    return tokenizer.decode(flat.tolist())
-
-
-def download_and_load_gpt2(model_size, models_dir):
-    # Validate model size
-    allowed_sizes = ("124M", "355M", "774M", "1558M")
-    if model_size not in allowed_sizes:
-        raise ValueError(f"Model size not in {allowed_sizes}")
-
-    # Define paths
-    model_dir = os.path.join(models_dir, model_size)
-    base_url = "https://openaipublic.blob.core.windows.net/gpt-2/models"
-    filenames = [
-        "checkpoint", "encoder.json", "hparams.json",
-        "model.ckpt.data-00000-of-00001", "model.ckpt.index",
-        "model.ckpt.meta", "vocab.bpe"
-    ]
-
-    # Download files
-    os.makedirs(model_dir, exist_ok=True)
-    for filename in filenames:
-        file_url = os.path.join(base_url, model_size, filename)
-        file_path = os.path.join(model_dir, filename)
-        download_file(file_url, file_path)
-
-    # Load settings and params
-    tf_ckpt_path = tf.train.latest_checkpoint(model_dir)
-    settings = json.load(open(os.path.join(model_dir, "hparams.json")))
-    params = load_gpt2_params_from_tf_ckpt(tf_ckpt_path, settings)
-
-    return settings, params
-
-
-def download_file(url, destination):
-    # Send a GET request to download the file
-    with urllib.request.urlopen(url) as response:
-        # Get the total file size from headers, defaulting to 0 if not present
-        file_size = int(response.headers.get("Content-Length", 0))
-
-        # Check if file exists and has the same size
-        if os.path.exists(destination):
-            file_size_local = os.path.getsize(destination)
-            if file_size == file_size_local:
-                print(f"File already exists and is up-to-date: {destination}")
-                return
-
-        # Define the block size for reading the file
-        block_size = 1024  # 1 Kilobyte
-
-        # Initialize the progress bar with total file size
-        progress_bar_description = os.path.basename(url)  # Extract filename from URL
-        with tqdm(total=file_size, unit="iB", unit_scale=True, desc=progress_bar_description) as progress_bar:
-            # Open the destination file in binary write mode
-            with open(destination, "wb") as file:
-                # Read the file in chunks and write to destination
-                while True:
-                    chunk = response.read(block_size)
-                    if not chunk:
-                        break
-                    file.write(chunk)
-                    progress_bar.update(len(chunk))  # Update progress bar
-
-
-def load_gpt2_params_from_tf_ckpt(ckpt_path, settings):
-    # Initialize parameters dictionary with empty blocks for each layer
-    params = {"blocks": [{} for _ in range(settings["n_layer"])]}
-
-    # Iterate over each variable in the checkpoint
-    for name, _ in tf.train.list_variables(ckpt_path):
-        # Load the variable and remove singleton dimensions
-        variable_array = np.squeeze(tf.train.load_variable(ckpt_path, name))
-
-        # Process the variable name to extract relevant parts
-        variable_name_parts = name.split("/")[1:]  # Skip the 'model/' prefix
-
-        # Identify the target dictionary for the variable
-        target_dict = params
-        if variable_name_parts[0].startswith("h"):
-            layer_number = int(variable_name_parts[0][1:])
-            target_dict = params["blocks"][layer_number]
-
-        # Recursively access or create nested dictionaries
-        for key in variable_name_parts[1:-1]:
-            target_dict = target_dict.setdefault(key, {})
-
-        # Assign the variable array to the last key
-        last_key = variable_name_parts[-1]
-        target_dict[last_key] = variable_array
-
-    return params
-
-
-def assign(left, right):
-    if left.shape != right.shape:
-        raise ValueError(f"Shape mismatch. Left: {left.shape}, Right: {right.shape}")
-    return torch.nn.Parameter(torch.tensor(right))
-
-
-def load_weights_into_gpt(gpt, params):
-    gpt.pos_emb.weight = assign(gpt.pos_emb.weight, params['wpe'])
-    gpt.tok_emb.weight = assign(gpt.tok_emb.weight, params['wte'])
-
-    for b in range(len(params["blocks"])):
-        q_w, k_w, v_w = np.split(
-            (params["blocks"][b]["attn"]["c_attn"])["w"], 3, axis=-1)
-        gpt.trf_blocks[b].att.W_query.weight = assign(
-            gpt.trf_blocks[b].att.W_query.weight, q_w.T)
-        gpt.trf_blocks[b].att.W_key.weight = assign(
-            gpt.trf_blocks[b].att.W_key.weight, k_w.T)
-        gpt.trf_blocks[b].att.W_value.weight = assign(
-            gpt.trf_blocks[b].att.W_value.weight, v_w.T)
-
-        q_b, k_b, v_b = np.split(
-            (params["blocks"][b]["attn"]["c_attn"])["b"], 3, axis=-1)
-        gpt.trf_blocks[b].att.W_query.bias = assign(
-            gpt.trf_blocks[b].att.W_query.bias, q_b)
-        gpt.trf_blocks[b].att.W_key.bias = assign(
-            gpt.trf_blocks[b].att.W_key.bias, k_b)
-        gpt.trf_blocks[b].att.W_value.bias = assign(
-            gpt.trf_blocks[b].att.W_value.bias, v_b)
-
-        gpt.trf_blocks[b].att.out_proj.weight = assign(
-            gpt.trf_blocks[b].att.out_proj.weight,
-            params["blocks"][b]["attn"]["c_proj"]["w"].T)
-        gpt.trf_blocks[b].att.out_proj.bias = assign(
-            gpt.trf_blocks[b].att.out_proj.bias,
-            params["blocks"][b]["attn"]["c_proj"]["b"])
-
-        gpt.trf_blocks[b].ff.layers[0].weight = assign(
-            gpt.trf_blocks[b].ff.layers[0].weight,
-            params["blocks"][b]["mlp"]["c_fc"]["w"].T)
-        gpt.trf_blocks[b].ff.layers[0].bias = assign(
-            gpt.trf_blocks[b].ff.layers[0].bias,
-            params["blocks"][b]["mlp"]["c_fc"]["b"])
-        gpt.trf_blocks[b].ff.layers[2].weight = assign(
-            gpt.trf_blocks[b].ff.layers[2].weight,
-            params["blocks"][b]["mlp"]["c_proj"]["w"].T)
-        gpt.trf_blocks[b].ff.layers[2].bias = assign(
-            gpt.trf_blocks[b].ff.layers[2].bias,
-            params["blocks"][b]["mlp"]["c_proj"]["b"])
-
-        gpt.trf_blocks[b].norm1.scale = assign(
-            gpt.trf_blocks[b].norm1.scale,
-            params["blocks"][b]["ln_1"]["g"])
-        gpt.trf_blocks[b].norm1.shift = assign(
-            gpt.trf_blocks[b].norm1.shift,
-            params["blocks"][b]["ln_1"]["b"])
-        gpt.trf_blocks[b].norm2.scale = assign(
-            gpt.trf_blocks[b].norm2.scale,
-            params["blocks"][b]["ln_2"]["g"])
-        gpt.trf_blocks[b].norm2.shift = assign(
-            gpt.trf_blocks[b].norm2.shift,
-            params["blocks"][b]["ln_2"]["b"])
-
-    gpt.final_norm.scale = assign(gpt.final_norm.scale, params["g"])
-    gpt.final_norm.shift = assign(gpt.final_norm.shift, params["b"])
-    gpt.out_head.weight = assign(gpt.out_head.weight, params["wte"])
-
-
-#####################################
-# Chapter 6
-#####################################
-def classify_review(text, model, tokenizer, device, max_length=None, pad_token_id=50256):
-    model.eval()
-
-    # Prepare inputs to the model
-    input_ids = tokenizer.encode(text)
-    supported_context_length = model.pos_emb.weight.shape[0]
-
-    # Truncate sequences if they too long
-    input_ids = input_ids[:min(max_length, supported_context_length)]
-
-    # Pad sequences to the longest sequence
-    input_ids += [pad_token_id] * (max_length - len(input_ids))
-    input_tensor = torch.tensor(input_ids, device=device).unsqueeze(0)  # add batch dimension
-
-    # Model inference
-    with torch.no_grad():
-        logits = model(input_tensor.to(device))[:, -1, :]  # Logits of the last output token
-    predicted_label = torch.argmax(logits, dim=-1).item()
-
-    # Return the classified result
-    return "spam" if predicted_label == 1 else "not spam"

ch07/01_main-chapter-code/ch07.ipynb

@@ -1541,6 +1541,12 @@
    "source": [
     "from gpt_download import download_and_load_gpt2\n",
     "from previous_chapters import GPTModel, load_weights_into_gpt\n",
+    "# If the `previous_chapters.py` file is not available locally,\n",
+    "# you can import it from the `llms-from-scratch` PyPI package.\n",
+    "# For details, see: https://github.com/rasbt/LLMs-from-scratch/tree/main/pkg\n",
+    "# E.g.,\n",
+    "# from llms_from_scratch.ch04 import GPTModel\n",
+    "# from llms_from_scratch.ch05 import download_and_load_gpt2, load_weights_into_gpt\n",
     "\n",
     "\n",
     "BASE_CONFIG = {\n",
@@ -1626,6 +1632,13 @@
     "    text_to_token_ids,\n",
     "    token_ids_to_text\n",
     ")\n",
+    "# Alternatively:\n",
+    "# from llms_from_scratch.ch05 import (\n",
+    "#    generate_text_simple,\n",
+    "#    text_to_token_ids,\n",
+    "#    token_ids_to_text\n",
+    "# )\n",
+    "\n",
     "\n",
     "token_ids = generate(\n",
     "    model=model,\n",
@@ -1727,7 +1740,12 @@
     "from previous_chapters import (\n",
     "    calc_loss_loader,\n",
     "    train_model_simple\n",
-    ")"
+    ")\n",
+    "# Alternatively:\n",
+    "# from llms_from_scratch.ch05 import (\n",
+    "#    calc_loss_loader,\n",
+    "#    train_model_simple,\n",
+    "# )\n"
    ]
   },
   {
@@ -1939,6 +1957,8 @@
    ],
    "source": [
     "from previous_chapters import plot_losses\n",
+    "# Alternatively:\n",
+    "# from llms_from_scratch.ch05 import plot_losses\n",
     "\n",
     "epochs_tensor = torch.linspace(0, num_epochs, len(train_losses))\n",
     "plot_losses(epochs_tensor, tokens_seen, train_losses, val_losses)"

ch07/04_preference-tuning-with-dpo/dpo-from-scratch.ipynb

@@ -1613,6 +1613,11 @@
    "outputs": [],
    "source": [
     "from previous_chapters import GPTModel\n",
+    "# If the `previous_chapters.py` file is not available locally,\n",
+    "# you can import it from the `llms-from-scratch` PyPI package.\n",
+    "# For details, see: https://github.com/rasbt/LLMs-from-scratch/tree/main/pkg\n",
+    "# E.g.,\n",
+    "# from llms_from_scratch.ch04 import GPTModel\n",
     "\n",
     "\n",
     "BASE_CONFIG = {\n",
@@ -1715,6 +1720,12 @@
     "    text_to_token_ids,\n",
     "    token_ids_to_text\n",
     ")\n",
+    "# Alternatively:\n",
+    "# from llms_from_scratch.ch05 (\n",
+    "#     generate,\n",
+    "#     text_to_token_ids,\n",
+    "#     token_ids_to_text\n",
+    "# )\n",
     "\n",
     "torch.manual_seed(123)\n",
     "\n",
@@ -2382,6 +2393,16 @@
     " - we also track the rewards and reward margins, which are commonly used in RLHF and DPO contexts to track the training progress\n"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "id": "820d4904-f819-4d62-bfb4-85cf28863683",
+   "metadata": {
+    "id": "820d4904-f819-4d62-bfb4-85cf28863683"
+   },
+   "source": [
+    "- Before we start the training, let's print the initial losses and rewards:"
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": 47,
@@ -2392,6 +2413,8 @@
    "outputs": [],
    "source": [
     "from previous_chapters import generate_and_print_sample\n",
+    "# Alternatively:\n",
+    "# from llms_from_scratch.ch04 import generate_text_simple\n",
     "\n",
     "\n",
     "def train_model_dpo_simple(\n",
@@ -2471,16 +2494,6 @@
     "    return tracking"
    ]
   },
-  {
-   "cell_type": "markdown",
-   "id": "820d4904-f819-4d62-bfb4-85cf28863683",
-   "metadata": {
-    "id": "820d4904-f819-4d62-bfb4-85cf28863683"
-   },
-   "source": [
-    "- Before we start the training, let's print the initial losses and rewards:"
-   ]
-  },
   {
    "cell_type": "code",
    "execution_count": 48,
@@ -2776,6 +2789,8 @@
    ],
    "source": [
     "from previous_chapters import plot_losses\n",
+    "# Alternatively:\n",
+    "# from llms_from_scratch.ch05 import plot_losses\n",
     "\n",
     "\n",
     "epochs_tensor = torch.linspace(0, num_epochs, len(tracking[\"train_losses\"]))\n",
@@ -3112,7 +3127,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.11.4"
+   "version": "3.10.16"
   }
  },
  "nbformat": 4,

ch07/06_user_interface/app.py

@@ -10,9 +10,12 @@ import tiktoken
 import torch
 import chainlit
 
-from previous_chapters import (
+
+# For llms_from_scratch installation instructions, see:
+# https://github.com/rasbt/LLMs-from-scratch/tree/main/pkg
+from llms_from_scratch.ch04 import GPTModel
+from llms_from_scratch.ch05 import (
     generate,
-    GPTModel,
     text_to_token_ids,
     token_ids_to_text,
 )

ch07/06_user_interface/previous_chapters.py

@@ -1,384 +0,0 @@
-# Copyright (c) Sebastian Raschka under Apache License 2.0 (see LICENSE.txt).
-# Source for "Build a Large Language Model From Scratch"
-#   - https://www.manning.com/books/build-a-large-language-model-from-scratch
-# Code: https://github.com/rasbt/LLMs-from-scratch
-#
-# This file collects all the relevant code that we covered thus far
-# throughout Chapters 2-5.
-
-import json
-import os
-import urllib
-
-import numpy as np
-import tensorflow as tf
-import torch
-import torch.nn as nn
-from tqdm import tqdm
-
-
-#####################################
-# Chapter 3
-#####################################
-class MultiHeadAttention(nn.Module):
-    def __init__(self, d_in, d_out, context_length, dropout, num_heads, qkv_bias=False):
-        super().__init__()
-        assert d_out % num_heads == 0, "d_out must be divisible by n_heads"
-
-        self.d_out = d_out
-        self.num_heads = num_heads
-        self.head_dim = d_out // num_heads  # Reduce the projection dim to match desired output dim
-
-        self.W_query = nn.Linear(d_in, d_out, bias=qkv_bias)
-        self.W_key = nn.Linear(d_in, d_out, bias=qkv_bias)
-        self.W_value = nn.Linear(d_in, d_out, bias=qkv_bias)
-        self.out_proj = nn.Linear(d_out, d_out)  # Linear layer to combine head outputs
-        self.dropout = nn.Dropout(dropout)
-        self.register_buffer('mask', torch.triu(torch.ones(context_length, context_length), diagonal=1))
-
-    def forward(self, x):
-        b, num_tokens, d_in = x.shape
-
-        keys = self.W_key(x)  # Shape: (b, num_tokens, d_out)
-        queries = self.W_query(x)
-        values = self.W_value(x)
-
-        # We implicitly split the matrix by adding a `num_heads` dimension
-        # Unroll last dim: (b, num_tokens, d_out) -> (b, num_tokens, num_heads, head_dim)
-        keys = keys.view(b, num_tokens, self.num_heads, self.head_dim)
-        values = values.view(b, num_tokens, self.num_heads, self.head_dim)
-        queries = queries.view(b, num_tokens, self.num_heads, self.head_dim)
-
-        # Transpose: (b, num_tokens, num_heads, head_dim) -> (b, num_heads, num_tokens, head_dim)
-        keys = keys.transpose(1, 2)
-        queries = queries.transpose(1, 2)
-        values = values.transpose(1, 2)
-
-        # Compute scaled dot-product attention (aka self-attention) with a causal mask
-        attn_scores = queries @ keys.transpose(2, 3)  # Dot product for each head
-
-        # Original mask truncated to the number of tokens and converted to boolean
-        mask_bool = self.mask.bool()[:num_tokens, :num_tokens]
-
-        # Use the mask to fill attention scores
-        attn_scores.masked_fill_(mask_bool, -torch.inf)
-
-        attn_weights = torch.softmax(attn_scores / keys.shape[-1]**0.5, dim=-1)
-        attn_weights = self.dropout(attn_weights)
-
-        # Shape: (b, num_tokens, num_heads, head_dim)
-        context_vec = (attn_weights @ values).transpose(1, 2)
-
-        # Combine heads, where self.d_out = self.num_heads * self.head_dim
-        context_vec = context_vec.reshape(b, num_tokens, self.d_out)
-        context_vec = self.out_proj(context_vec)  # optional projection
-
-        return context_vec
-
-
-#####################################
-# Chapter 4
-#####################################
-class LayerNorm(nn.Module):
-    def __init__(self, emb_dim):
-        super().__init__()
-        self.eps = 1e-5
-        self.scale = nn.Parameter(torch.ones(emb_dim))
-        self.shift = nn.Parameter(torch.zeros(emb_dim))
-
-    def forward(self, x):
-        mean = x.mean(dim=-1, keepdim=True)
-        var = x.var(dim=-1, keepdim=True, unbiased=False)
-        norm_x = (x - mean) / torch.sqrt(var + self.eps)
-        return self.scale * norm_x + self.shift
-
-
-class GELU(nn.Module):
-    def __init__(self):
-        super().__init__()
-
-    def forward(self, x):
-        return 0.5 * x * (1 + torch.tanh(
-            torch.sqrt(torch.tensor(2.0 / torch.pi)) *
-            (x + 0.044715 * torch.pow(x, 3))
-        ))
-
-
-class FeedForward(nn.Module):
-    def __init__(self, cfg):
-        super().__init__()
-        self.layers = nn.Sequential(
-            nn.Linear(cfg["emb_dim"], 4 * cfg["emb_dim"]),
-            GELU(),
-            nn.Linear(4 * cfg["emb_dim"], cfg["emb_dim"]),
-        )
-
-    def forward(self, x):
-        return self.layers(x)
-
-
-class TransformerBlock(nn.Module):
-    def __init__(self, cfg):
-        super().__init__()
-        self.att = MultiHeadAttention(
-            d_in=cfg["emb_dim"],
-            d_out=cfg["emb_dim"],
-            context_length=cfg["context_length"],
-            num_heads=cfg["n_heads"],
-            dropout=cfg["drop_rate"],
-            qkv_bias=cfg["qkv_bias"])
-        self.ff = FeedForward(cfg)
-        self.norm1 = LayerNorm(cfg["emb_dim"])
-        self.norm2 = LayerNorm(cfg["emb_dim"])
-        self.drop_shortcut = nn.Dropout(cfg["drop_rate"])
-
-    def forward(self, x):
-        # Shortcut connection for attention block
-        shortcut = x
-        x = self.norm1(x)
-        x = self.att(x)   # Shape [batch_size, num_tokens, emb_size]
-        x = self.drop_shortcut(x)
-        x = x + shortcut  # Add the original input back
-
-        # Shortcut connection for feed-forward block
-        shortcut = x
-        x = self.norm2(x)
-        x = self.ff(x)
-        x = self.drop_shortcut(x)
-        x = x + shortcut  # Add the original input back
-
-        return x
-
-
-class GPTModel(nn.Module):
-    def __init__(self, cfg):
-        super().__init__()
-        self.tok_emb = nn.Embedding(cfg["vocab_size"], cfg["emb_dim"])
-        self.pos_emb = nn.Embedding(cfg["context_length"], cfg["emb_dim"])
-        self.drop_emb = nn.Dropout(cfg["drop_rate"])
-
-        self.trf_blocks = nn.Sequential(
-            *[TransformerBlock(cfg) for _ in range(cfg["n_layers"])])
-
-        self.final_norm = LayerNorm(cfg["emb_dim"])
-        self.out_head = nn.Linear(cfg["emb_dim"], cfg["vocab_size"], bias=False)
-
-    def forward(self, in_idx):
-        batch_size, seq_len = in_idx.shape
-        tok_embeds = self.tok_emb(in_idx)
-        pos_embeds = self.pos_emb(torch.arange(seq_len, device=in_idx.device))
-        x = tok_embeds + pos_embeds  # Shape [batch_size, num_tokens, emb_size]
-        x = self.drop_emb(x)
-        x = self.trf_blocks(x)
-        x = self.final_norm(x)
-        logits = self.out_head(x)
-        return logits
-
-
-#####################################
-# Chapter 5
-#####################################
-def text_to_token_ids(text, tokenizer):
-    encoded = tokenizer.encode(text)
-    encoded_tensor = torch.tensor(encoded).unsqueeze(0)  # add batch dimension
-    return encoded_tensor
-
-
-def token_ids_to_text(token_ids, tokenizer):
-    flat = token_ids.squeeze(0)  # remove batch dimension
-    return tokenizer.decode(flat.tolist())
-
-
-def download_and_load_gpt2(model_size, models_dir):
-    # Validate model size
-    allowed_sizes = ("124M", "355M", "774M", "1558M")
-    if model_size not in allowed_sizes:
-        raise ValueError(f"Model size not in {allowed_sizes}")
-
-    # Define paths
-    model_dir = os.path.join(models_dir, model_size)
-    base_url = "https://openaipublic.blob.core.windows.net/gpt-2/models"
-    filenames = [
-        "checkpoint", "encoder.json", "hparams.json",
-        "model.ckpt.data-00000-of-00001", "model.ckpt.index",
-        "model.ckpt.meta", "vocab.bpe"
-    ]
-
-    # Download files
-    os.makedirs(model_dir, exist_ok=True)
-    for filename in filenames:
-        file_url = os.path.join(base_url, model_size, filename)
-        file_path = os.path.join(model_dir, filename)
-        download_file(file_url, file_path)
-
-    # Load settings and params
-    tf_ckpt_path = tf.train.latest_checkpoint(model_dir)
-    settings = json.load(open(os.path.join(model_dir, "hparams.json")))
-    params = load_gpt2_params_from_tf_ckpt(tf_ckpt_path, settings)
-
-    return settings, params
-
-
-def download_file(url, destination):
-    # Send a GET request to download the file
-    with urllib.request.urlopen(url) as response:
-        # Get the total file size from headers, defaulting to 0 if not present
-        file_size = int(response.headers.get("Content-Length", 0))
-
-        # Check if file exists and has the same size
-        if os.path.exists(destination):
-            file_size_local = os.path.getsize(destination)
-            if file_size == file_size_local:
-                print(f"File already exists and is up-to-date: {destination}")
-                return
-
-        # Define the block size for reading the file
-        block_size = 1024  # 1 Kilobyte
-
-        # Initialize the progress bar with total file size
-        progress_bar_description = os.path.basename(url)  # Extract filename from URL
-        with tqdm(total=file_size, unit="iB", unit_scale=True, desc=progress_bar_description) as progress_bar:
-            # Open the destination file in binary write mode
-            with open(destination, "wb") as file:
-                # Read the file in chunks and write to destination
-                while True:
-                    chunk = response.read(block_size)
-                    if not chunk:
-                        break
-                    file.write(chunk)
-                    progress_bar.update(len(chunk))  # Update progress bar
-
-
-def load_gpt2_params_from_tf_ckpt(ckpt_path, settings):
-    # Initialize parameters dictionary with empty blocks for each layer
-    params = {"blocks": [{} for _ in range(settings["n_layer"])]}
-
-    # Iterate over each variable in the checkpoint
-    for name, _ in tf.train.list_variables(ckpt_path):
-        # Load the variable and remove singleton dimensions
-        variable_array = np.squeeze(tf.train.load_variable(ckpt_path, name))
-
-        # Process the variable name to extract relevant parts
-        variable_name_parts = name.split("/")[1:]  # Skip the 'model/' prefix
-
-        # Identify the target dictionary for the variable
-        target_dict = params
-        if variable_name_parts[0].startswith("h"):
-            layer_number = int(variable_name_parts[0][1:])
-            target_dict = params["blocks"][layer_number]
-
-        # Recursively access or create nested dictionaries
-        for key in variable_name_parts[1:-1]:
-            target_dict = target_dict.setdefault(key, {})
-
-        # Assign the variable array to the last key
-        last_key = variable_name_parts[-1]
-        target_dict[last_key] = variable_array
-
-    return params
-
-
-def assign(left, right):
-    if left.shape != right.shape:
-        raise ValueError(f"Shape mismatch. Left: {left.shape}, Right: {right.shape}")
-    return torch.nn.Parameter(torch.tensor(right))
-
-
-def load_weights_into_gpt(gpt, params):
-    gpt.pos_emb.weight = assign(gpt.pos_emb.weight, params['wpe'])
-    gpt.tok_emb.weight = assign(gpt.tok_emb.weight, params['wte'])
-
-    for b in range(len(params["blocks"])):
-        q_w, k_w, v_w = np.split(
-            (params["blocks"][b]["attn"]["c_attn"])["w"], 3, axis=-1)
-        gpt.trf_blocks[b].att.W_query.weight = assign(
-            gpt.trf_blocks[b].att.W_query.weight, q_w.T)
-        gpt.trf_blocks[b].att.W_key.weight = assign(
-            gpt.trf_blocks[b].att.W_key.weight, k_w.T)
-        gpt.trf_blocks[b].att.W_value.weight = assign(
-            gpt.trf_blocks[b].att.W_value.weight, v_w.T)
-
-        q_b, k_b, v_b = np.split(
-            (params["blocks"][b]["attn"]["c_attn"])["b"], 3, axis=-1)
-        gpt.trf_blocks[b].att.W_query.bias = assign(
-            gpt.trf_blocks[b].att.W_query.bias, q_b)
-        gpt.trf_blocks[b].att.W_key.bias = assign(
-            gpt.trf_blocks[b].att.W_key.bias, k_b)
-        gpt.trf_blocks[b].att.W_value.bias = assign(
-            gpt.trf_blocks[b].att.W_value.bias, v_b)
-
-        gpt.trf_blocks[b].att.out_proj.weight = assign(
-            gpt.trf_blocks[b].att.out_proj.weight,
-            params["blocks"][b]["attn"]["c_proj"]["w"].T)
-        gpt.trf_blocks[b].att.out_proj.bias = assign(
-            gpt.trf_blocks[b].att.out_proj.bias,
-            params["blocks"][b]["attn"]["c_proj"]["b"])
-
-        gpt.trf_blocks[b].ff.layers[0].weight = assign(
-            gpt.trf_blocks[b].ff.layers[0].weight,
-            params["blocks"][b]["mlp"]["c_fc"]["w"].T)
-        gpt.trf_blocks[b].ff.layers[0].bias = assign(
-            gpt.trf_blocks[b].ff.layers[0].bias,
-            params["blocks"][b]["mlp"]["c_fc"]["b"])
-        gpt.trf_blocks[b].ff.layers[2].weight = assign(
-            gpt.trf_blocks[b].ff.layers[2].weight,
-            params["blocks"][b]["mlp"]["c_proj"]["w"].T)
-        gpt.trf_blocks[b].ff.layers[2].bias = assign(
-            gpt.trf_blocks[b].ff.layers[2].bias,
-            params["blocks"][b]["mlp"]["c_proj"]["b"])
-
-        gpt.trf_blocks[b].norm1.scale = assign(
-            gpt.trf_blocks[b].norm1.scale,
-            params["blocks"][b]["ln_1"]["g"])
-        gpt.trf_blocks[b].norm1.shift = assign(
-            gpt.trf_blocks[b].norm1.shift,
-            params["blocks"][b]["ln_1"]["b"])
-        gpt.trf_blocks[b].norm2.scale = assign(
-            gpt.trf_blocks[b].norm2.scale,
-            params["blocks"][b]["ln_2"]["g"])
-        gpt.trf_blocks[b].norm2.shift = assign(
-            gpt.trf_blocks[b].norm2.shift,
-            params["blocks"][b]["ln_2"]["b"])
-
-    gpt.final_norm.scale = assign(gpt.final_norm.scale, params["g"])
-    gpt.final_norm.shift = assign(gpt.final_norm.shift, params["b"])
-    gpt.out_head.weight = assign(gpt.out_head.weight, params["wte"])
-
-
-def generate(model, idx, max_new_tokens, context_size, temperature=0.0, top_k=None, eos_id=None):
-
-    # For-loop is the same as before: Get logits, and only focus on last time step
-    for _ in range(max_new_tokens):
-        idx_cond = idx[:, -context_size:]
-        with torch.no_grad():
-            logits = model(idx_cond)
-        logits = logits[:, -1, :]
-
-        # New: Filter logits with top_k sampling
-        if top_k is not None:
-            # Keep only top_k values
-            top_logits, _ = torch.topk(logits, top_k)
-            min_val = top_logits[:, -1]
-            logits = torch.where(logits < min_val, torch.tensor(float('-inf')).to(logits.device), logits)
-
-        # New: Apply temperature scaling
-        if temperature > 0.0:
-            logits = logits / temperature
-
-            # Apply softmax to get probabilities
-            probs = torch.softmax(logits, dim=-1)  # (batch_size, context_len)
-
-            # Sample from the distribution
-            idx_next = torch.multinomial(probs, num_samples=1)  # (batch_size, 1)
-
-        # Otherwise same as before: get idx of the vocab entry with the highest logits value
-        else:
-            idx_next = torch.argmax(logits, dim=-1, keepdim=True)  # (batch_size, 1)
-
-        if idx_next == eos_id:  # Stop generating early if end-of-sequence token is encountered and eos_id is specified
-            break
-
-        # Same as before: append sampled index to the running sequence
-        idx = torch.cat((idx, idx_next), dim=1)  # (batch_size, num_tokens+1)
-
-    return idx

pkg/llms_from_scratch/__init__.py

@@ -0,0 +1,4 @@
+# Copyright (c) Sebastian Raschka under Apache License 2.0 (see LICENSE.txt).
+# Source for "Build a Large Language Model From Scratch"
+#   - https://www.manning.com/books/build-a-large-language-model-from-scratch
+# Code: https://github.com/rasbt/LLMs-from-scratch

pkg/llms_from_scratch/appendix_a.py

@@ -0,0 +1,44 @@
+# Copyright (c) Sebastian Raschka under Apache License 2.0 (see LICENSE.txt).
+# Source for "Build a Large Language Model From Scratch"
+#   - https://www.manning.com/books/build-a-large-language-model-from-scratch
+# Code: https://github.com/rasbt/LLMs-from-scratch
+
+import torch
+from torch.utils.data import Dataset
+
+
+class NeuralNetwork(torch.nn.Module):
+    def __init__(self, num_inputs, num_outputs):
+        super().__init__()
+
+        self.layers = torch.nn.Sequential(
+
+            # 1st hidden layer
+            torch.nn.Linear(num_inputs, 30),
+            torch.nn.ReLU(),
+
+            # 2nd hidden layer
+            torch.nn.Linear(30, 20),
+            torch.nn.ReLU(),
+
+            # output layer
+            torch.nn.Linear(20, num_outputs),
+        )
+
+    def forward(self, x):
+        logits = self.layers(x)
+        return logits
+
+
+class ToyDataset(Dataset):
+    def __init__(self, X, y):
+        self.features = X
+        self.labels = y
+
+    def __getitem__(self, index):
+        one_x = self.features[index]
+        one_y = self.labels[index]
+        return one_x, one_y
+
+    def __len__(self):
+        return self.labels.shape[0]

pkg/llms_from_scratch/appendix_d.py

@@ -0,0 +1,94 @@
+# Copyright (c) Sebastian Raschka under Apache License 2.0 (see LICENSE.txt).
+# Source for "Build a Large Language Model From Scratch"
+#   - https://www.manning.com/books/build-a-large-language-model-from-scratch
+# Code: https://github.com/rasbt/LLMs-from-scratch
+
+from .ch05 import calc_loss_batch, evaluate_model, generate_and_print_sample
+
+import math
+import torch
+
+
+def find_highest_gradient(model):
+    max_grad = None
+    for param in model.parameters():
+        if param.grad is not None:
+            grad_values = param.grad.data.flatten()
+            max_grad_param = grad_values.max()
+            if max_grad is None or max_grad_param > max_grad:
+                max_grad = max_grad_param
+    return max_grad
+
+
+def train_model(model, train_loader, val_loader, optimizer, device,
+                n_epochs, eval_freq, eval_iter, start_context, tokenizer,
+                warmup_steps, initial_lr=3e-05, min_lr=1e-6, orig_book_version=False):
+
+    train_losses, val_losses, track_tokens_seen, track_lrs = [], [], [], []
+    tokens_seen, global_step = 0, -1
+
+    # Retrieve the maximum learning rate from the optimizer
+    peak_lr = optimizer.param_groups[0]["lr"]
+
+    # Calculate the total number of iterations in the training process
+    total_training_steps = len(train_loader) * n_epochs
+
+    # Calculate the learning rate increment during the warmup phase
+    lr_increment = (peak_lr - initial_lr) / warmup_steps
+
+    for epoch in range(n_epochs):
+        model.train()
+        for input_batch, target_batch in train_loader:
+            optimizer.zero_grad()
+            global_step += 1
+
+            # Adjust the learning rate based on the current phase (warmup or cosine annealing)
+            if global_step < warmup_steps:
+                # Linear warmup
+                lr = initial_lr + global_step * lr_increment
+            else:
+                # Cosine annealing after warmup
+                progress = ((global_step - warmup_steps) /
+                            (total_training_steps - warmup_steps))
+                lr = min_lr + (peak_lr - min_lr) * 0.5 * (1 + math.cos(math.pi * progress))
+
+            # Apply the calculated learning rate to the optimizer
+            for param_group in optimizer.param_groups:
+                param_group["lr"] = lr
+            track_lrs.append(lr)  # Store the current learning rate
+
+            # Calculate and backpropagate the loss
+            loss = calc_loss_batch(input_batch, target_batch, model, device)
+            loss.backward()
+
+            # Apply gradient clipping after the warmup phase to avoid exploding gradients
+            if orig_book_version:
+                if global_step > warmup_steps:
+                    torch.nn.utils.clip_grad_norm_(model.parameters(), max_norm=1.0)
+            else:
+                if global_step >= warmup_steps:  # the book originally used global_step > warmup_steps, which lead to a skipped clipping step after warmup
+                    torch.nn.utils.clip_grad_norm_(model.parameters(), max_norm=1.0)
+
+            optimizer.step()
+            tokens_seen += input_batch.numel()
+
+            # Periodically evaluate the model on the training and validation sets
+            if global_step % eval_freq == 0:
+                train_loss, val_loss = evaluate_model(
+                    model, train_loader, val_loader,
+                    device, eval_iter
+                )
+                train_losses.append(train_loss)
+                val_losses.append(val_loss)
+                track_tokens_seen.append(tokens_seen)
+                # Print the current losses
+                print(f"Ep {epoch+1} (Iter {global_step:06d}): "
+                      f"Train loss {train_loss:.3f}, "
+                      f"Val loss {val_loss:.3f}")
+
+        # Generate and print a sample from the model to monitor progress
+        generate_and_print_sample(
+            model, tokenizer, device, start_context
+        )
+
+    return train_losses, val_losses, track_tokens_seen, track_lrs

pkg/llms_from_scratch/appendix_e.py

@@ -0,0 +1,42 @@
+# Copyright (c) Sebastian Raschka under Apache License 2.0 (see LICENSE.txt).
+# Source for "Build a Large Language Model From Scratch"
+#   - https://www.manning.com/books/build-a-large-language-model-from-scratch
+# Code: https://github.com/rasbt/LLMs-from-scratch
+
+import torch
+import math
+
+
+class LoRALayer(torch.nn.Module):
+    def __init__(self, in_dim, out_dim, rank, alpha):
+        super().__init__()
+        self.A = torch.nn.Parameter(torch.empty(in_dim, rank))
+        torch.nn.init.kaiming_uniform_(self.A, a=math.sqrt(5))  # similar to standard weight initialization
+        self.B = torch.nn.Parameter(torch.zeros(rank, out_dim))
+        self.alpha = alpha
+
+    def forward(self, x):
+        x = self.alpha * (x @ self.A @ self.B)
+        return x
+
+
+class LinearWithLoRA(torch.nn.Module):
+    def __init__(self, linear, rank, alpha):
+        super().__init__()
+        self.linear = linear
+        self.lora = LoRALayer(
+            linear.in_features, linear.out_features, rank, alpha
+        )
+
+    def forward(self, x):
+        return self.linear(x) + self.lora(x)
+
+
+def replace_linear_with_lora(model, rank, alpha):
+    for name, module in model.named_children():
+        if isinstance(module, torch.nn.Linear):
+            # Replace the Linear layer with LinearWithLoRA
+            setattr(model, name, LinearWithLoRA(module, rank, alpha))
+        else:
+            # Recursively apply the same function to child modules
+            replace_linear_with_lora(module, rank, alpha)

pkg/llms_from_scratch/ch02.py

@@ -0,0 +1,46 @@
+# Copyright (c) Sebastian Raschka under Apache License 2.0 (see LICENSE.txt).
+# Source for "Build a Large Language Model From Scratch"
+#   - https://www.manning.com/books/build-a-large-language-model-from-scratch
+# Code: https://github.com/rasbt/LLMs-from-scratch
+
+import torch
+from torch.utils.data import Dataset, DataLoader
+import tiktoken
+
+
+class GPTDatasetV1(Dataset):
+    def __init__(self, txt, tokenizer, max_length, stride):
+        self.tokenizer = tokenizer
+        self.input_ids = []
+        self.target_ids = []
+
+        # Tokenize the entire text
+        token_ids = tokenizer.encode(txt, allowed_special={"<|endoftext|>"})
+
+        # Use a sliding window to chunk the book into overlapping sequences of max_length
+        for i in range(0, len(token_ids) - max_length, stride):
+            input_chunk = token_ids[i:i + max_length]
+            target_chunk = token_ids[i + 1: i + max_length + 1]
+            self.input_ids.append(torch.tensor(input_chunk))
+            self.target_ids.append(torch.tensor(target_chunk))
+
+    def __len__(self):
+        return len(self.input_ids)
+
+    def __getitem__(self, idx):
+        return self.input_ids[idx], self.target_ids[idx]
+
+
+def create_dataloader_v1(txt, batch_size=4, max_length=256,
+                         stride=128, shuffle=True, drop_last=True, num_workers=0):
+    # Initialize the tokenizer
+    tokenizer = tiktoken.get_encoding("gpt2")
+
+    # Create dataset
+    dataset = GPTDatasetV1(txt, tokenizer, max_length, stride)
+
+    # Create dataloader
+    dataloader = DataLoader(
+        dataset, batch_size=batch_size, shuffle=shuffle, drop_last=drop_last, num_workers=num_workers)
+
+    return dataloader

pkg/llms_from_scratch/ch03.py

@@ -0,0 +1,151 @@
+# Copyright (c) Sebastian Raschka under Apache License 2.0 (see LICENSE.txt).
+# Source for "Build a Large Language Model From Scratch"
+#   - https://www.manning.com/books/build-a-large-language-model-from-scratch
+# Code: https://github.com/rasbt/LLMs-from-scratch
+
+import torch
+import torch.nn as nn
+
+
+class SelfAttention_v1(nn.Module):
+
+    def __init__(self, d_in, d_out):
+        super().__init__()
+        self.W_query = nn.Parameter(torch.rand(d_in, d_out))
+        self.W_key = nn.Parameter(torch.rand(d_in, d_out))
+        self.W_value = nn.Parameter(torch.rand(d_in, d_out))
+
+    def forward(self, x):
+        keys = x @ self.W_key
+        queries = x @ self.W_query
+        values = x @ self.W_value
+
+        attn_scores = queries @ keys.T # omega
+        attn_weights = torch.softmax(
+            attn_scores / keys.shape[-1]**0.5, dim=-1
+        )
+
+        context_vec = attn_weights @ values
+        return context_vec
+
+
+class SelfAttention_v2(nn.Module):
+
+    def __init__(self, d_in, d_out, qkv_bias=False):
+        super().__init__()
+        self.W_query = nn.Linear(d_in, d_out, bias=qkv_bias)
+        self.W_key = nn.Linear(d_in, d_out, bias=qkv_bias)
+        self.W_value = nn.Linear(d_in, d_out, bias=qkv_bias)
+
+    def forward(self, x):
+        keys = self.W_key(x)
+        queries = self.W_query(x)
+        values = self.W_value(x)
+
+        attn_scores = queries @ keys.T
+        attn_weights = torch.softmax(attn_scores / keys.shape[-1]**0.5, dim=-1)
+
+        context_vec = attn_weights @ values
+        return context_vec
+
+
+class CausalAttention(nn.Module):
+
+    def __init__(self, d_in, d_out, context_length,
+                 dropout, qkv_bias=False):
+        super().__init__()
+        self.d_out = d_out
+        self.W_query = nn.Linear(d_in, d_out, bias=qkv_bias)
+        self.W_key = nn.Linear(d_in, d_out, bias=qkv_bias)
+        self.W_value = nn.Linear(d_in, d_out, bias=qkv_bias)
+        self.dropout = nn.Dropout(dropout)  # New
+        self.register_buffer('mask', torch.triu(torch.ones(context_length, context_length), diagonal=1)) # New
+
+    def forward(self, x):
+        b, num_tokens, d_in = x.shape  # New batch dimension b
+        # For inputs where `num_tokens` exceeds `context_length`, this will result in errors
+        # in the mask creation further below.
+        # In practice, this is not a problem since the LLM (chapters 4-7) ensures that inputs
+        # do not exceed `context_length` before reaching this forward method.
+        keys = self.W_key(x)
+        queries = self.W_query(x)
+        values = self.W_value(x)
+
+        attn_scores = queries @ keys.transpose(1, 2)  # Changed transpose
+        attn_scores.masked_fill_(  # New, _ ops are in-place
+            self.mask.bool()[:num_tokens, :num_tokens], -torch.inf)  # `:num_tokens` to account for cases where the number of tokens in the batch is smaller than the supported context_size
+        attn_weights = torch.softmax(
+            attn_scores / keys.shape[-1]**0.5, dim=-1
+        )
+        attn_weights = self.dropout(attn_weights)  # New
+
+        context_vec = attn_weights @ values
+        return context_vec
+
+
+class MultiHeadAttentionWrapper(nn.Module):
+    def __init__(self, d_in, d_out, context_length, dropout, num_heads, qkv_bias=False):
+        super().__init__()
+        self.heads = nn.ModuleList(
+            [CausalAttention(d_in, d_out, context_length, dropout, qkv_bias)
+             for _ in range(num_heads)]
+        )
+
+    def forward(self, x):
+        return torch.cat([head(x) for head in self.heads], dim=-1)
+
+
+class MultiHeadAttention(nn.Module):
+    def __init__(self, d_in, d_out, context_length, dropout, num_heads, qkv_bias=False):
+        super().__init__()
+        assert d_out % num_heads == 0, "d_out must be divisible by n_heads"
+
+        self.d_out = d_out
+        self.num_heads = num_heads
+        self.head_dim = d_out // num_heads  # Reduce the projection dim to match desired output dim
+
+        self.W_query = nn.Linear(d_in, d_out, bias=qkv_bias)
+        self.W_key = nn.Linear(d_in, d_out, bias=qkv_bias)
+        self.W_value = nn.Linear(d_in, d_out, bias=qkv_bias)
+        self.out_proj = nn.Linear(d_out, d_out)  # Linear layer to combine head outputs
+        self.dropout = nn.Dropout(dropout)
+        self.register_buffer('mask', torch.triu(torch.ones(context_length, context_length), diagonal=1))
+
+    def forward(self, x):
+        b, num_tokens, d_in = x.shape
+
+        keys = self.W_key(x)  # Shape: (b, num_tokens, d_out)
+        queries = self.W_query(x)
+        values = self.W_value(x)
+
+        # We implicitly split the matrix by adding a `num_heads` dimension
+        # Unroll last dim: (b, num_tokens, d_out) -> (b, num_tokens, num_heads, head_dim)
+        keys = keys.view(b, num_tokens, self.num_heads, self.head_dim)
+        values = values.view(b, num_tokens, self.num_heads, self.head_dim)
+        queries = queries.view(b, num_tokens, self.num_heads, self.head_dim)
+
+        # Transpose: (b, num_tokens, num_heads, head_dim) -> (b, num_heads, num_tokens, head_dim)
+        keys = keys.transpose(1, 2)
+        queries = queries.transpose(1, 2)
+        values = values.transpose(1, 2)
+
+        # Compute scaled dot-product attention (aka self-attention) with a causal mask
+        attn_scores = queries @ keys.transpose(2, 3)  # Dot product for each head
+
+        # Original mask truncated to the number of tokens and converted to boolean
+        mask_bool = self.mask.bool()[:num_tokens, :num_tokens]
+
+        # Use the mask to fill attention scores
+        attn_scores.masked_fill_(mask_bool, -torch.inf)
+
+        attn_weights = torch.softmax(attn_scores / keys.shape[-1]**0.5, dim=-1)
+        attn_weights = self.dropout(attn_weights)
+
+        # Shape: (b, num_tokens, num_heads, head_dim)
+        context_vec = (attn_weights @ values).transpose(1, 2)
+
+        # Combine heads, where self.d_out = self.num_heads * self.head_dim
+        context_vec = context_vec.reshape(b, num_tokens, self.d_out)
+        context_vec = self.out_proj(context_vec)  # optional projection
+
+        return context_vec

pkg/llms_from_scratch/ch04.py

@@ -2,76 +2,12 @@
 # Source for "Build a Large Language Model From Scratch"
 #   - https://www.manning.com/books/build-a-large-language-model-from-scratch
 # Code: https://github.com/rasbt/LLMs-from-scratch
-#
+
-# This file collects all the relevant code that we covered thus far
+from .ch03 import MultiHeadAttention
-# throughout Chapters 2-4.
-# This file can be run as a standalone script.
 import torch
 import torch.nn as nn
 
 
-#####################################
-# Chapter 3
-#####################################
-class MultiHeadAttention(nn.Module):
-    def __init__(self, d_in, d_out, context_length, dropout, num_heads, qkv_bias=False):
-        super().__init__()
-        assert d_out % num_heads == 0, "d_out must be divisible by n_heads"
-
-        self.d_out = d_out
-        self.num_heads = num_heads
-        self.head_dim = d_out // num_heads  # Reduce the projection dim to match desired output dim
-
-        self.W_query = nn.Linear(d_in, d_out, bias=qkv_bias)
-        self.W_key = nn.Linear(d_in, d_out, bias=qkv_bias)
-        self.W_value = nn.Linear(d_in, d_out, bias=qkv_bias)
-        self.out_proj = nn.Linear(d_out, d_out)  # Linear layer to combine head outputs
-        self.dropout = nn.Dropout(dropout)
-        self.register_buffer('mask', torch.triu(torch.ones(context_length, context_length), diagonal=1))
-
-    def forward(self, x):
-        b, num_tokens, d_in = x.shape
-
-        keys = self.W_key(x)  # Shape: (b, num_tokens, d_out)
-        queries = self.W_query(x)
-        values = self.W_value(x)
-
-        # We implicitly split the matrix by adding a `num_heads` dimension
-        # Unroll last dim: (b, num_tokens, d_out) -> (b, num_tokens, num_heads, head_dim)
-        keys = keys.view(b, num_tokens, self.num_heads, self.head_dim)
-        values = values.view(b, num_tokens, self.num_heads, self.head_dim)
-        queries = queries.view(b, num_tokens, self.num_heads, self.head_dim)
-
-        # Transpose: (b, num_tokens, num_heads, head_dim) -> (b, num_heads, num_tokens, head_dim)
-        keys = keys.transpose(1, 2)
-        queries = queries.transpose(1, 2)
-        values = values.transpose(1, 2)
-
-        # Compute scaled dot-product attention (aka self-attention) with a causal mask
-        attn_scores = queries @ keys.transpose(2, 3)  # Dot product for each head
-
-        # Original mask truncated to the number of tokens and converted to boolean
-        mask_bool = self.mask.bool()[:num_tokens, :num_tokens]
-
-        # Use the mask to fill attention scores
-        attn_scores.masked_fill_(mask_bool, -torch.inf)
-
-        attn_weights = torch.softmax(attn_scores / keys.shape[-1]**0.5, dim=-1)
-        attn_weights = self.dropout(attn_weights)
-
-        # Shape: (b, num_tokens, num_heads, head_dim)
-        context_vec = (attn_weights @ values).transpose(1, 2)
-
-        # Combine heads, where self.d_out = self.num_heads * self.head_dim
-        context_vec = context_vec.reshape(b, num_tokens, self.d_out)
-        context_vec = self.out_proj(context_vec)  # optional projection
-
-        return context_vec
-
-
-#####################################
-# Chapter 4
-#####################################
 class LayerNorm(nn.Module):
     def __init__(self, emb_dim):
         super().__init__()
@@ -123,21 +59,21 @@ class TransformerBlock(nn.Module):
         self.ff = FeedForward(cfg)
         self.norm1 = LayerNorm(cfg["emb_dim"])
         self.norm2 = LayerNorm(cfg["emb_dim"])
-        self.drop_shortcut = nn.Dropout(cfg["drop_rate"])
+        self.drop_resid = nn.Dropout(cfg["drop_rate"])
 
     def forward(self, x):
         # Shortcut connection for attention block
         shortcut = x
         x = self.norm1(x)
         x = self.att(x)   # Shape [batch_size, num_tokens, emb_size]
-        x = self.drop_shortcut(x)
+        x = self.drop_resid(x)
         x = x + shortcut  # Add the original input back
 
         # Shortcut connection for feed-forward block
         shortcut = x
         x = self.norm2(x)
         x = self.ff(x)
-        x = self.drop_shortcut(x)
+        x = self.drop_resid(x)
         x = x + shortcut  # Add the original input back
 
         return x

pkg/llms_from_scratch/ch05.py

@@ -0,0 +1,233 @@
+# Copyright (c) Sebastian Raschka under Apache License 2.0 (see LICENSE.txt).
+# Source for "Build a Large Language Model From Scratch"
+#   - https://www.manning.com/books/build-a-large-language-model-from-scratch
+# Code: https://github.com/rasbt/LLMs-from-scratch
+
+from .ch04 import generate_text_simple
+import numpy as np
+import matplotlib.pyplot as plt
+from matplotlib.ticker import MaxNLocator
+import torch
+
+
+def generate(model, idx, max_new_tokens, context_size, temperature=0.0, top_k=None, eos_id=None):
+
+    # For-loop is the same as before: Get logits, and only focus on last time step
+    for _ in range(max_new_tokens):
+        idx_cond = idx[:, -context_size:]
+        with torch.no_grad():
+            logits = model(idx_cond)
+        logits = logits[:, -1, :]
+
+        # New: Filter logits with top_k sampling
+        if top_k is not None:
+            # Keep only top_k values
+            top_logits, _ = torch.topk(logits, top_k)
+            min_val = top_logits[:, -1]
+            logits = torch.where(logits < min_val, torch.tensor(float('-inf')).to(logits.device), logits)
+
+        # New: Apply temperature scaling
+        if temperature > 0.0:
+            logits = logits / temperature
+
+            # Apply softmax to get probabilities
+            probs = torch.softmax(logits, dim=-1)  # (batch_size, context_len)
+
+            # Sample from the distribution
+            idx_next = torch.multinomial(probs, num_samples=1)  # (batch_size, 1)
+
+        # Otherwise same as before: get idx of the vocab entry with the highest logits value
+        else:
+            idx_next = torch.argmax(logits, dim=-1, keepdim=True)  # (batch_size, 1)
+
+        if idx_next == eos_id:  # Stop generating early if end-of-sequence token is encountered and eos_id is specified
+            break
+
+        # Same as before: append sampled index to the running sequence
+        idx = torch.cat((idx, idx_next), dim=1)  # (batch_size, num_tokens+1)
+
+    return idx
+
+
+def train_model_simple(model, train_loader, val_loader, optimizer, device, num_epochs,
+                       eval_freq, eval_iter, start_context, tokenizer):
+    # Initialize lists to track losses and tokens seen
+    train_losses, val_losses, track_tokens_seen = [], [], []
+    tokens_seen, global_step = 0, -1
+
+    # Main training loop
+    for epoch in range(num_epochs):
+        model.train()  # Set model to training mode
+
+        for input_batch, target_batch in train_loader:
+            optimizer.zero_grad()  # Reset loss gradients from previous batch iteration
+            loss = calc_loss_batch(input_batch, target_batch, model, device)
+            loss.backward()  # Calculate loss gradients
+            optimizer.step()  # Update model weights using loss gradients
+            tokens_seen += input_batch.numel()
+            global_step += 1
+
+            # Optional evaluation step
+            if global_step % eval_freq == 0:
+                train_loss, val_loss = evaluate_model(
+                    model, train_loader, val_loader, device, eval_iter)
+                train_losses.append(train_loss)
+                val_losses.append(val_loss)
+                track_tokens_seen.append(tokens_seen)
+                print(f"Ep {epoch+1} (Step {global_step:06d}): "
+                      f"Train loss {train_loss:.3f}, Val loss {val_loss:.3f}")
+
+        # Print a sample text after each epoch
+        generate_and_print_sample(
+            model, tokenizer, device, start_context
+        )
+
+    return train_losses, val_losses, track_tokens_seen
+
+
+def evaluate_model(model, train_loader, val_loader, device, eval_iter):
+    model.eval()
+    with torch.no_grad():
+        train_loss = calc_loss_loader(train_loader, model, device, num_batches=eval_iter)
+        val_loss = calc_loss_loader(val_loader, model, device, num_batches=eval_iter)
+    model.train()
+    return train_loss, val_loss
+
+
+def generate_and_print_sample(model, tokenizer, device, start_context):
+    model.eval()
+    context_size = model.pos_emb.weight.shape[0]
+    encoded = text_to_token_ids(start_context, tokenizer).to(device)
+    with torch.no_grad():
+        token_ids = generate_text_simple(
+            model=model, idx=encoded,
+            max_new_tokens=50, context_size=context_size
+        )
+        decoded_text = token_ids_to_text(token_ids, tokenizer)
+        print(decoded_text.replace("\n", " "))  # Compact print format
+    model.train()
+
+
+def assign(left, right):
+    if left.shape != right.shape:
+        raise ValueError(f"Shape mismatch. Left: {left.shape}, Right: {right.shape}")
+    return torch.nn.Parameter(torch.tensor(right))
+
+
+def load_weights_into_gpt(gpt, params):
+    gpt.pos_emb.weight = assign(gpt.pos_emb.weight, params['wpe'])
+    gpt.tok_emb.weight = assign(gpt.tok_emb.weight, params['wte'])
+
+    for b in range(len(params["blocks"])):
+        q_w, k_w, v_w = np.split(
+            (params["blocks"][b]["attn"]["c_attn"])["w"], 3, axis=-1)
+        gpt.trf_blocks[b].att.W_query.weight = assign(
+            gpt.trf_blocks[b].att.W_query.weight, q_w.T)
+        gpt.trf_blocks[b].att.W_key.weight = assign(
+            gpt.trf_blocks[b].att.W_key.weight, k_w.T)
+        gpt.trf_blocks[b].att.W_value.weight = assign(
+            gpt.trf_blocks[b].att.W_value.weight, v_w.T)
+
+        q_b, k_b, v_b = np.split(
+            (params["blocks"][b]["attn"]["c_attn"])["b"], 3, axis=-1)
+        gpt.trf_blocks[b].att.W_query.bias = assign(
+            gpt.trf_blocks[b].att.W_query.bias, q_b)
+        gpt.trf_blocks[b].att.W_key.bias = assign(
+            gpt.trf_blocks[b].att.W_key.bias, k_b)
+        gpt.trf_blocks[b].att.W_value.bias = assign(
+            gpt.trf_blocks[b].att.W_value.bias, v_b)
+
+        gpt.trf_blocks[b].att.out_proj.weight = assign(
+            gpt.trf_blocks[b].att.out_proj.weight,
+            params["blocks"][b]["attn"]["c_proj"]["w"].T)
+        gpt.trf_blocks[b].att.out_proj.bias = assign(
+            gpt.trf_blocks[b].att.out_proj.bias,
+            params["blocks"][b]["attn"]["c_proj"]["b"])
+
+        gpt.trf_blocks[b].ff.layers[0].weight = assign(
+            gpt.trf_blocks[b].ff.layers[0].weight,
+            params["blocks"][b]["mlp"]["c_fc"]["w"].T)
+        gpt.trf_blocks[b].ff.layers[0].bias = assign(
+            gpt.trf_blocks[b].ff.layers[0].bias,
+            params["blocks"][b]["mlp"]["c_fc"]["b"])
+        gpt.trf_blocks[b].ff.layers[2].weight = assign(
+            gpt.trf_blocks[b].ff.layers[2].weight,
+            params["blocks"][b]["mlp"]["c_proj"]["w"].T)
+        gpt.trf_blocks[b].ff.layers[2].bias = assign(
+            gpt.trf_blocks[b].ff.layers[2].bias,
+            params["blocks"][b]["mlp"]["c_proj"]["b"])
+
+        gpt.trf_blocks[b].norm1.scale = assign(
+            gpt.trf_blocks[b].norm1.scale,
+            params["blocks"][b]["ln_1"]["g"])
+        gpt.trf_blocks[b].norm1.shift = assign(
+            gpt.trf_blocks[b].norm1.shift,
+            params["blocks"][b]["ln_1"]["b"])
+        gpt.trf_blocks[b].norm2.scale = assign(
+            gpt.trf_blocks[b].norm2.scale,
+            params["blocks"][b]["ln_2"]["g"])
+        gpt.trf_blocks[b].norm2.shift = assign(
+            gpt.trf_blocks[b].norm2.shift,
+            params["blocks"][b]["ln_2"]["b"])
+
+    gpt.final_norm.scale = assign(gpt.final_norm.scale, params["g"])
+    gpt.final_norm.shift = assign(gpt.final_norm.shift, params["b"])
+    gpt.out_head.weight = assign(gpt.out_head.weight, params["wte"])
+
+
+def text_to_token_ids(text, tokenizer):
+    encoded = tokenizer.encode(text, allowed_special={"<|endoftext|>"})
+    encoded_tensor = torch.tensor(encoded).unsqueeze(0)  # add batch dimension
+    return encoded_tensor
+
+
+def token_ids_to_text(token_ids, tokenizer):
+    flat = token_ids.squeeze(0)  # remove batch dimension
+    return tokenizer.decode(flat.tolist())
+
+
+def calc_loss_batch(input_batch, target_batch, model, device):
+    input_batch, target_batch = input_batch.to(device), target_batch.to(device)
+    logits = model(input_batch)
+    loss = torch.nn.functional.cross_entropy(logits.flatten(0, 1), target_batch.flatten())
+    return loss
+
+
+def calc_loss_loader(data_loader, model, device, num_batches=None):
+    total_loss = 0.
+    if len(data_loader) == 0:
+        return float("nan")
+    elif num_batches is None:
+        num_batches = len(data_loader)
+    else:
+        # Reduce the number of batches to match the total number of batches in the data loader
+        # if num_batches exceeds the number of batches in the data loader
+        num_batches = min(num_batches, len(data_loader))
+    for i, (input_batch, target_batch) in enumerate(data_loader):
+        if i < num_batches:
+            loss = calc_loss_batch(input_batch, target_batch, model, device)
+            total_loss += loss.item()
+        else:
+            break
+    return total_loss / num_batches
+
+
+def plot_losses(epochs_seen, tokens_seen, train_losses, val_losses):
+    fig, ax1 = plt.subplots(figsize=(5, 3))
+
+    # Plot training and validation loss against epochs
+    ax1.plot(epochs_seen, train_losses, label="Training loss")
+    ax1.plot(epochs_seen, val_losses, linestyle="-.", label="Validation loss")
+    ax1.set_xlabel("Epochs")
+    ax1.set_ylabel("Loss")
+    ax1.legend(loc="upper right")
+    ax1.xaxis.set_major_locator(MaxNLocator(integer=True))  # only show integer labels on x-axis
+
+    # Create a second x-axis for tokens seen
+    ax2 = ax1.twiny()  # Create a second x-axis that shares the same y-axis
+    ax2.plot(tokens_seen, train_losses, alpha=0)  # Invisible plot for aligning ticks
+    ax2.set_xlabel("Tokens seen")
+
+    fig.tight_layout()  # Adjust layout to make room
+    plt.savefig("loss-plot.pdf")
+    plt.show()

pkg/llms_from_scratch/ch06.py

@@ -0,0 +1,254 @@
+# Copyright (c) Sebastian Raschka under Apache License 2.0 (see LICENSE.txt).
+# Source for "Build a Large Language Model From Scratch"
+#   - https://www.manning.com/books/build-a-large-language-model-from-scratch
+# Code: https://github.com/rasbt/LLMs-from-scratch
+
+
+import urllib.request
+import zipfile
+import os
+from pathlib import Path
+
+import matplotlib.pyplot as plt
+from torch.utils.data import Dataset
+import torch
+import pandas as pd
+
+
+def download_and_unzip_spam_data(url, zip_path, extracted_path, data_file_path):
+    if data_file_path.exists():
+        print(f"{data_file_path} already exists. Skipping download and extraction.")
+        return
+
+    # Downloading the file
+    with urllib.request.urlopen(url) as response:
+        with open(zip_path, "wb") as out_file:
+            out_file.write(response.read())
+
+    # Unzipping the file
+    with zipfile.ZipFile(zip_path, "r") as zip_ref:
+        zip_ref.extractall(extracted_path)
+
+    # Add .tsv file extension
+    original_file_path = Path(extracted_path) / "SMSSpamCollection"
+    os.rename(original_file_path, data_file_path)
+    print(f"File downloaded and saved as {data_file_path}")
+
+
+def create_balanced_dataset(df):
+
+    # Count the instances of "spam"
+    num_spam = df[df["Label"] == "spam"].shape[0]
+
+    # Randomly sample "ham" instances to match the number of "spam" instances
+    ham_subset = df[df["Label"] == "ham"].sample(num_spam, random_state=123)
+
+    # Combine ham "subset" with "spam"
+    balanced_df = pd.concat([ham_subset, df[df["Label"] == "spam"]])
+
+    return balanced_df
+
+
+def random_split(df, train_frac, validation_frac):
+    # Shuffle the entire DataFrame
+    df = df.sample(frac=1, random_state=123).reset_index(drop=True)
+
+    # Calculate split indices
+    train_end = int(len(df) * train_frac)
+    validation_end = train_end + int(len(df) * validation_frac)
+
+    # Split the DataFrame
+    train_df = df[:train_end]
+    validation_df = df[train_end:validation_end]
+    test_df = df[validation_end:]
+
+    return train_df, validation_df, test_df
+
+
+class SpamDataset(Dataset):
+    def __init__(self, csv_file, tokenizer, max_length=None, pad_token_id=50256):
+        self.data = pd.read_csv(csv_file)
+
+        # Pre-tokenize texts
+        self.encoded_texts = [
+            tokenizer.encode(text) for text in self.data["Text"]
+        ]
+
+        if max_length is None:
+            self.max_length = self._longest_encoded_length()
+        else:
+            self.max_length = max_length
+            # Truncate sequences if they are longer than max_length
+            self.encoded_texts = [
+                encoded_text[:self.max_length]
+                for encoded_text in self.encoded_texts
+            ]
+
+        # Pad sequences to the longest sequence
+        self.encoded_texts = [
+            encoded_text + [pad_token_id] * (self.max_length - len(encoded_text))
+            for encoded_text in self.encoded_texts
+        ]
+
+    def __getitem__(self, index):
+        encoded = self.encoded_texts[index]
+        label = self.data.iloc[index]["Label"]
+        return (
+            torch.tensor(encoded, dtype=torch.long),
+            torch.tensor(label, dtype=torch.long)
+        )
+
+    def __len__(self):
+        return len(self.data)
+
+    def _longest_encoded_length(self):
+        max_length = 0
+        for encoded_text in self.encoded_texts:
+            encoded_length = len(encoded_text)
+            if encoded_length > max_length:
+                max_length = encoded_length
+        return max_length
+        # Note: A more pythonic version to implement this method
+        # is the following, which is also used in the next chapter:
+        # return max(len(encoded_text) for encoded_text in self.encoded_texts)
+
+
+def calc_accuracy_loader(data_loader, model, device, num_batches=None):
+    model.eval()
+    correct_predictions, num_examples = 0, 0
+
+    if num_batches is None:
+        num_batches = len(data_loader)
+    else:
+        num_batches = min(num_batches, len(data_loader))
+    for i, (input_batch, target_batch) in enumerate(data_loader):
+        if i < num_batches:
+            input_batch, target_batch = input_batch.to(device), target_batch.to(device)
+
+            with torch.no_grad():
+                logits = model(input_batch)[:, -1, :]  # Logits of last output token
+            predicted_labels = torch.argmax(logits, dim=-1)
+
+            num_examples += predicted_labels.shape[0]
+            correct_predictions += (predicted_labels == target_batch).sum().item()
+        else:
+            break
+    return correct_predictions / num_examples
+
+
+def calc_loss_batch(input_batch, target_batch, model, device):
+    input_batch, target_batch = input_batch.to(device), target_batch.to(device)
+    logits = model(input_batch)[:, -1, :]  # Logits of last output token
+    loss = torch.nn.functional.cross_entropy(logits, target_batch)
+    return loss
+
+
+def calc_loss_loader(data_loader, model, device, num_batches=None):
+    total_loss = 0.
+    if len(data_loader) == 0:
+        return float("nan")
+    elif num_batches is None:
+        num_batches = len(data_loader)
+    else:
+        # Reduce the number of batches to match the total number of batches in the data loader
+        # if num_batches exceeds the number of batches in the data loader
+        num_batches = min(num_batches, len(data_loader))
+    for i, (input_batch, target_batch) in enumerate(data_loader):
+        if i < num_batches:
+            loss = calc_loss_batch(input_batch, target_batch, model, device)
+            total_loss += loss.item()
+        else:
+            break
+    return total_loss / num_batches
+
+
+def evaluate_model(model, train_loader, val_loader, device, eval_iter):
+    model.eval()
+    with torch.no_grad():
+        train_loss = calc_loss_loader(train_loader, model, device, num_batches=eval_iter)
+        val_loss = calc_loss_loader(val_loader, model, device, num_batches=eval_iter)
+    model.train()
+    return train_loss, val_loss
+
+
+def train_classifier_simple(model, train_loader, val_loader, optimizer, device, num_epochs,
+                            eval_freq, eval_iter):
+    # Initialize lists to track losses and examples seen
+    train_losses, val_losses, train_accs, val_accs = [], [], [], []
+    examples_seen, global_step = 0, -1
+
+    # Main training loop
+    for epoch in range(num_epochs):
+        model.train()  # Set model to training mode
+
+        for input_batch, target_batch in train_loader:
+            optimizer.zero_grad()  # Reset loss gradients from previous batch iteration
+            loss = calc_loss_batch(input_batch, target_batch, model, device)
+            loss.backward()  # Calculate loss gradients
+            optimizer.step()  # Update model weights using loss gradients
+            examples_seen += input_batch.shape[0]  # New: track examples instead of tokens
+            global_step += 1
+
+            # Optional evaluation step
+            if global_step % eval_freq == 0:
+                train_loss, val_loss = evaluate_model(
+                    model, train_loader, val_loader, device, eval_iter)
+                train_losses.append(train_loss)
+                val_losses.append(val_loss)
+                print(f"Ep {epoch+1} (Step {global_step:06d}): "
+                      f"Train loss {train_loss:.3f}, Val loss {val_loss:.3f}")
+
+        # Calculate accuracy after each epoch
+        train_accuracy = calc_accuracy_loader(train_loader, model, device, num_batches=eval_iter)
+        val_accuracy = calc_accuracy_loader(val_loader, model, device, num_batches=eval_iter)
+        print(f"Training accuracy: {train_accuracy*100:.2f}% | ", end="")
+        print(f"Validation accuracy: {val_accuracy*100:.2f}%")
+        train_accs.append(train_accuracy)
+        val_accs.append(val_accuracy)
+
+    return train_losses, val_losses, train_accs, val_accs, examples_seen
+
+
+def plot_values(epochs_seen, examples_seen, train_values, val_values, label="loss"):
+    fig, ax1 = plt.subplots(figsize=(5, 3))
+
+    # Plot training and validation loss against epochs
+    ax1.plot(epochs_seen, train_values, label=f"Training {label}")
+    ax1.plot(epochs_seen, val_values, linestyle="-.", label=f"Validation {label}")
+    ax1.set_xlabel("Epochs")
+    ax1.set_ylabel(label.capitalize())
+    ax1.legend()
+
+    # Create a second x-axis for examples seen
+    ax2 = ax1.twiny()  # Create a second x-axis that shares the same y-axis
+    ax2.plot(examples_seen, train_values, alpha=0)  # Invisible plot for aligning ticks
+    ax2.set_xlabel("Examples seen")
+
+    fig.tight_layout()  # Adjust layout to make room
+    plt.savefig(f"{label}-plot.pdf")
+    plt.show()
+
+
+def classify_review(text, model, tokenizer, device, max_length=None, pad_token_id=50256):
+    model.eval()
+
+    # Prepare inputs to the model
+    input_ids = tokenizer.encode(text)
+    supported_context_length = model.pos_emb.weight.shape[0]
+    # Note: In the book, this was originally written as pos_emb.weight.shape[1] by mistake
+    # It didn't break the code but would have caused unnecessary truncation (to 768 instead of 1024)
+
+    # Truncate sequences if they too long
+    input_ids = input_ids[:min(max_length, supported_context_length)]
+
+    # Pad sequences to the longest sequence
+    input_ids += [pad_token_id] * (max_length - len(input_ids))
+    input_tensor = torch.tensor(input_ids, device=device).unsqueeze(0) # add batch dimension
+
+    # Model inference
+    with torch.no_grad():
+        logits = model(input_tensor)[:, -1, :]  # Logits of the last output token
+    predicted_label = torch.argmax(logits, dim=-1).item()
+
+    # Return the classified result
+    return "spam" if predicted_label == 1 else "not spam"

pkg/llms_from_scratch/ch07.py

@@ -0,0 +1,247 @@
+# Copyright (c) Sebastian Raschka under Apache License 2.0 (see LICENSE.txt).
+# Source for "Build a Large Language Model From Scratch"
+#   - https://www.manning.com/books/build-a-large-language-model-from-scratch
+# Code: https://github.com/rasbt/LLMs-from-scratch
+
+import json
+import os
+import psutil
+import urllib
+
+import torch
+import tqdm
+from torch.utils.data import Dataset
+
+
+def download_and_load_file(file_path, url):
+
+    if not os.path.exists(file_path):
+        with urllib.request.urlopen(url) as response:
+            text_data = response.read().decode("utf-8")
+        with open(file_path, "w", encoding="utf-8") as file:
+            file.write(text_data)
+
+    # The book originally contained this unnecessary "else" clause:
+    # else:
+    #     with open(file_path, "r", encoding="utf-8") as file:
+    #         text_data = file.read()
+
+    with open(file_path, "r", encoding="utf-8") as file:
+        data = json.load(file)
+
+    return data
+
+
+def format_input(entry):
+    instruction_text = (
+        f"Below is an instruction that describes a task. "
+        f"Write a response that appropriately completes the request."
+        f"\n\n### Instruction:\n{entry['instruction']}"
+    )
+
+    input_text = f"\n\n### Input:\n{entry['input']}" if entry["input"] else ""
+
+    return instruction_text + input_text
+
+
+class InstructionDataset(Dataset):
+    def __init__(self, data, tokenizer):
+        self.data = data
+
+        # Pre-tokenize texts
+        self.encoded_texts = []
+        for entry in data:
+            instruction_plus_input = format_input(entry)
+            response_text = f"\n\n### Response:\n{entry['output']}"
+            full_text = instruction_plus_input + response_text
+            self.encoded_texts.append(
+                tokenizer.encode(full_text)
+            )
+
+    def __getitem__(self, index):
+        return self.encoded_texts[index]
+
+    def __len__(self):
+        return len(self.data)
+
+
+def custom_collate_draft_1(
+    batch,
+    pad_token_id=50256,
+    device="cpu"
+):
+    # Find the longest sequence in the batch
+    # and increase the max length by +1, which will add one extra
+    # padding token below
+    batch_max_length = max(len(item)+1 for item in batch)
+
+    # Pad and prepare inputs
+    inputs_lst = []
+
+    for item in batch:
+        new_item = item.copy()
+        # Add an <|endoftext|> token
+        new_item += [pad_token_id]
+        # Pad sequences to batch_max_length
+        padded = (
+            new_item + [pad_token_id] *
+            (batch_max_length - len(new_item))
+        )
+        # Via padded[:-1], we remove the extra padded token
+        # that has been added via the +1 setting in batch_max_length
+        # (the extra padding token will be relevant in later codes)
+        inputs = torch.tensor(padded[:-1])
+        inputs_lst.append(inputs)
+
+    # Convert list of inputs to tensor and transfer to target device
+    inputs_tensor = torch.stack(inputs_lst).to(device)
+    return inputs_tensor
+
+
+def custom_collate_draft_2(
+    batch,
+    pad_token_id=50256,
+    device="cpu"
+):
+    # Find the longest sequence in the batch
+    batch_max_length = max(len(item)+1 for item in batch)
+
+    # Pad and prepare inputs
+    inputs_lst, targets_lst = [], []
+
+    for item in batch:
+        new_item = item.copy()
+        # Add an <|endoftext|> token
+        new_item += [pad_token_id]
+        # Pad sequences to max_length
+        padded = (
+            new_item + [pad_token_id] *
+            (batch_max_length - len(new_item))
+        )
+        inputs = torch.tensor(padded[:-1])  # Truncate the last token for inputs
+        targets = torch.tensor(padded[1:])  # Shift +1 to the right for targets
+        inputs_lst.append(inputs)
+        targets_lst.append(targets)
+
+    # Convert list of inputs to tensor and transfer to target device
+    inputs_tensor = torch.stack(inputs_lst).to(device)
+    targets_tensor = torch.stack(targets_lst).to(device)
+    return inputs_tensor, targets_tensor
+
+
+def custom_collate_fn(
+    batch,
+    pad_token_id=50256,
+    ignore_index=-100,
+    allowed_max_length=None,
+    device="cpu"
+):
+    # Find the longest sequence in the batch
+    batch_max_length = max(len(item)+1 for item in batch)
+
+    # Pad and prepare inputs and targets
+    inputs_lst, targets_lst = [], []
+
+    for item in batch:
+        new_item = item.copy()
+        # Add an <|endoftext|> token
+        new_item += [pad_token_id]
+        # Pad sequences to max_length
+        padded = (
+            new_item + [pad_token_id] *
+            (batch_max_length - len(new_item))
+        )
+        inputs = torch.tensor(padded[:-1])  # Truncate the last token for inputs
+        targets = torch.tensor(padded[1:])  # Shift +1 to the right for targets
+
+        # New: Replace all but the first padding tokens in targets by ignore_index
+        mask = targets == pad_token_id
+        indices = torch.nonzero(mask).squeeze()
+        if indices.numel() > 1:
+            targets[indices[1:]] = ignore_index
+
+        # New: Optionally truncate to maximum sequence length
+        if allowed_max_length is not None:
+            inputs = inputs[:allowed_max_length]
+            targets = targets[:allowed_max_length]
+
+        inputs_lst.append(inputs)
+        targets_lst.append(targets)
+
+    # Convert list of inputs and targets to tensors and transfer to target device
+    inputs_tensor = torch.stack(inputs_lst).to(device)
+    targets_tensor = torch.stack(targets_lst).to(device)
+
+    return inputs_tensor, targets_tensor
+
+
+def check_if_running(process_name):
+    running = False
+    for proc in psutil.process_iter(["name"]):
+        if process_name in proc.info["name"]:
+            running = True
+            break
+    return running
+
+
+def query_model(
+    prompt,
+    model="llama3",
+    url="http://localhost:11434/api/chat"
+):
+    # Create the data payload as a dictionary
+    data = {
+        "model": model,
+        "messages": [
+            {"role": "user", "content": prompt}
+        ],
+        "options": {     # Settings below are required for deterministic responses
+            "seed": 123,
+            "temperature": 0,
+            "num_ctx": 2048
+        }
+    }
+
+    # Convert the dictionary to a JSON formatted string and encode it to bytes
+    payload = json.dumps(data).encode("utf-8")
+
+    # Create a request object, setting the method to POST and adding necessary headers
+    request = urllib.request.Request(
+        url,
+        data=payload,
+        method="POST"
+    )
+    request.add_header("Content-Type", "application/json")
+
+    # Send the request and capture the response
+    response_data = ""
+    with urllib.request.urlopen(request) as response:
+        # Read and decode the response
+        while True:
+            line = response.readline().decode("utf-8")
+            if not line:
+                break
+            response_json = json.loads(line)
+            response_data += response_json["message"]["content"]
+
+    return response_data
+
+
+def generate_model_scores(json_data, json_key, model="llama3"):
+    scores = []
+    for entry in tqdm(json_data, desc="Scoring entries"):
+        prompt = (
+            f"Given the input `{format_input(entry)}` "
+            f"and correct output `{entry['output']}`, "
+            f"score the model response `{entry[json_key]}`"
+            f" on a scale from 0 to 100, where 100 is the best score. "
+            f"Respond with the integer number only."
+        )
+        score = query_model(prompt, model)
+        try:
+            scores.append(int(score))
+        except ValueError:
+            print(f"Could not convert score: {score}")
+            continue
+
+    return scores

pkg/llms_from_scratch/tests/test_appendix_a.py

@@ -0,0 +1,70 @@
+# Copyright (c) Sebastian Raschka under Apache License 2.0 (see LICENSE.txt).
+# Source for "Build a Large Language Model From Scratch"
+#   - https://www.manning.com/books/build-a-large-language-model-from-scratch
+# Code: https://github.com/rasbt/LLMs-from-scratch
+
+from llms_from_scratch.appendix_a import NeuralNetwork, ToyDataset
+
+import torch
+import torch.nn.functional as F
+from torch.utils.data import DataLoader
+
+
+def test_dataset():
+
+    X_train = torch.tensor([
+        [-1.2, 3.1],
+        [-0.9, 2.9],
+        [-0.5, 2.6],
+        [2.3, -1.1],
+        [2.7, -1.5]
+    ])
+
+    y_train = torch.tensor([0, 0, 0, 1, 1])
+    train_ds = ToyDataset(X_train, y_train)
+
+    len(train_ds) == 5
+    torch.manual_seed(123)
+
+    train_loader = DataLoader(
+        dataset=train_ds,
+        batch_size=2,
+        shuffle=True,
+        num_workers=0
+    )
+
+    torch.manual_seed(123)
+    model = NeuralNetwork(num_inputs=2, num_outputs=2)
+    optimizer = torch.optim.SGD(model.parameters(), lr=0.5)
+
+    num_epochs = 3
+
+    for epoch in range(num_epochs):
+
+        model.train()
+        for batch_idx, (features, labels) in enumerate(train_loader):
+
+            logits = model(features)
+
+            loss = F.cross_entropy(logits, labels)
+
+            optimizer.zero_grad()
+            loss.backward()
+            optimizer.step()
+
+            print(f"Epoch: {epoch+1:03d}/{num_epochs:03d}"
+                  f" | Batch {batch_idx:03d}/{len(train_loader):03d}"
+                  f" | Train/Val Loss: {loss:.2f}")
+
+        model.eval()
+        with torch.no_grad():
+            outputs = model(X_train)
+
+        expected = torch.tensor([
+            [2.8569, -4.1618],
+            [2.5382, -3.7548],
+            [2.0944, -3.1820],
+            [-1.4814, 1.4816],
+            [-1.7176, 1.7342]
+        ])
+        torch.equal(outputs, expected)

pkg/llms_from_scratch/tests/test_appendix_d.py

@@ -0,0 +1,118 @@
+# Copyright (c) Sebastian Raschka under Apache License 2.0 (see LICENSE.txt).
+# Source for "Build a Large Language Model From Scratch"
+#   - https://www.manning.com/books/build-a-large-language-model-from-scratch
+# Code: https://github.com/rasbt/LLMs-from-scratch
+
+from llms_from_scratch.ch02 import create_dataloader_v1
+from llms_from_scratch.ch04 import GPTModel
+from llms_from_scratch.appendix_d import train_model
+
+import os
+import urllib
+
+import tiktoken
+import torch
+from torch.utils.data import Subset, DataLoader
+
+
+def test_train(tmp_path):
+
+    GPT_CONFIG_124M = {
+        "vocab_size": 50257,    # Vocabulary size
+        "context_length": 256,  # Shortened context length (orig: 1024)
+        "emb_dim": 768,         # Embedding dimension
+        "n_heads": 12,          # Number of attention heads
+        "n_layers": 12,         # Number of layers
+        "drop_rate": 0.1,       # Dropout rate
+        "qkv_bias": False       # Query-key-value bias
+    }
+
+    OTHER_SETTINGS = {
+        "learning_rate": 5e-4,
+        "num_epochs": 2,
+        "batch_size": 1,
+        "weight_decay": 0.1
+    }
+
+    torch.manual_seed(123)
+    device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+    ##############################
+    # Download data if necessary
+    ##############################
+
+    file_path = tmp_path / "the-verdict.txt"
+    url = "https://raw.githubusercontent.com/rasbt/LLMs-from-scratch/main/ch02/01_main-chapter-code/the-verdict.txt"
+
+    if not os.path.exists(file_path):
+        with urllib.request.urlopen(url) as response:
+            text_data = response.read().decode("utf-8")
+        with open(file_path, "w", encoding="utf-8") as file:
+            file.write(text_data)
+    else:
+        with open(file_path, "r", encoding="utf-8") as file:
+            text_data = file.read()
+
+    ##############################
+    # Initialize model
+    ##############################
+
+    model = GPTModel(GPT_CONFIG_124M)
+    model.to(device)  # no assignment model = model.to(device) necessary for nn.Module classes
+
+    ##############################
+    # Set up dataloaders
+    ##############################
+
+    # Train/validation ratio
+    train_ratio = 0.90
+    split_idx = int(train_ratio * len(text_data))
+
+    train_loader = create_dataloader_v1(
+        text_data[:split_idx],
+        batch_size=OTHER_SETTINGS["batch_size"],
+        max_length=GPT_CONFIG_124M["context_length"],
+        stride=GPT_CONFIG_124M["context_length"],
+        drop_last=True,
+        shuffle=True,
+        num_workers=0
+    )
+
+    val_loader = create_dataloader_v1(
+        text_data[split_idx:],
+        batch_size=OTHER_SETTINGS["batch_size"],
+        max_length=GPT_CONFIG_124M["context_length"],
+        stride=GPT_CONFIG_124M["context_length"],
+        drop_last=False,
+        shuffle=False,
+        num_workers=0
+    )
+
+    ##############################
+    # Train model
+    ##############################
+
+    tokenizer = tiktoken.get_encoding("gpt2")
+
+    train_subset = Subset(train_loader.dataset, range(1))
+    one_batch_train_loader = DataLoader(train_subset, batch_size=1)
+    val_subset = Subset(val_loader.dataset, range(1))
+    one_batch_val_loader = DataLoader(val_subset, batch_size=1)
+
+    peak_lr = 0.001  # this was originally set to 5e-4 in the book by mistake
+    optimizer = torch.optim.AdamW(model.parameters(), lr=peak_lr, weight_decay=0.1)  # the book accidentally omitted the lr assignment
+    tokenizer = tiktoken.get_encoding("gpt2")
+
+    n_epochs = 6
+    warmup_steps = 1
+
+    train_losses, val_losses, tokens_seen, lrs = train_model(
+        model, one_batch_train_loader, one_batch_val_loader, optimizer, device, n_epochs=n_epochs,
+        eval_freq=5, eval_iter=1, start_context="Every effort moves you",
+        tokenizer=tokenizer, warmup_steps=warmup_steps,
+        initial_lr=1e-5, min_lr=1e-5
+    )
+
+    assert round(train_losses[0], 1) == 10.9
+    assert round(val_losses[0], 1) == 11.0
+    assert train_losses[-1] < train_losses[0]

pkg/llms_from_scratch/tests/test_appendix_e.py

@@ -0,0 +1,150 @@
+# Copyright (c) Sebastian Raschka under Apache License 2.0 (see LICENSE.txt).
+# Source for "Build a Large Language Model From Scratch"
+#   - https://www.manning.com/books/build-a-large-language-model-from-scratch
+# Code: https://github.com/rasbt/LLMs-from-scratch
+
+
+from llms_from_scratch.ch04 import GPTModel
+from llms_from_scratch.ch06 import (
+    download_and_unzip_spam_data, create_balanced_dataset,
+    random_split, SpamDataset, train_classifier_simple
+)
+from llms_from_scratch.appendix_e import replace_linear_with_lora
+
+from pathlib import Path
+import urllib
+
+import pandas as pd
+import tiktoken
+import torch
+from torch.utils.data import DataLoader, Subset
+
+
+def test_train_classifier_lora(tmp_path):
+
+    ########################################
+    # Download and prepare dataset
+    ########################################
+
+    url = "https://archive.ics.uci.edu/static/public/228/sms+spam+collection.zip"
+    zip_path = tmp_path / "sms_spam_collection.zip"
+    extracted_path = tmp_path / "sms_spam_collection"
+    data_file_path = Path(extracted_path) / "SMSSpamCollection.tsv"
+
+    try:
+        download_and_unzip_spam_data(
+            url, zip_path, extracted_path, data_file_path
+        )
+    except (urllib.error.HTTPError, urllib.error.URLError, TimeoutError) as e:
+        print(f"Primary URL failed: {e}. Trying backup URL...")
+        backup_url = "https://f001.backblazeb2.com/file/LLMs-from-scratch/sms%2Bspam%2Bcollection.zip"
+        download_and_unzip_spam_data(
+            backup_url, zip_path, extracted_path, data_file_path
+        )
+
+    df = pd.read_csv(data_file_path, sep="\t", header=None, names=["Label", "Text"])
+    balanced_df = create_balanced_dataset(df)
+    balanced_df["Label"] = balanced_df["Label"].map({"ham": 0, "spam": 1})
+
+    train_df, validation_df, test_df = random_split(balanced_df, 0.7, 0.1)
+    train_df.to_csv(tmp_path / "train.csv", index=None)
+    validation_df.to_csv(tmp_path / "validation.csv", index=None)
+    test_df.to_csv(tmp_path / "test.csv", index=None)
+
+    ########################################
+    # Create data loaders
+    ########################################
+    tokenizer = tiktoken.get_encoding("gpt2")
+
+    train_dataset = SpamDataset(
+        csv_file=tmp_path / "train.csv",
+        max_length=None,
+        tokenizer=tokenizer
+    )
+
+    val_dataset = SpamDataset(
+        csv_file=tmp_path / "validation.csv",
+        max_length=train_dataset.max_length,
+        tokenizer=tokenizer
+    )
+
+    num_workers = 0
+    batch_size = 8
+
+    torch.manual_seed(123)
+
+    train_loader = DataLoader(
+        dataset=train_dataset,
+        batch_size=batch_size,
+        shuffle=True,
+        num_workers=num_workers,
+        drop_last=True,
+    )
+
+    val_loader = DataLoader(
+        dataset=val_dataset,
+        batch_size=batch_size,
+        num_workers=num_workers,
+        drop_last=False,
+    )
+
+    ########################################
+    # Load pretrained model
+    ########################################
+
+    # Small GPT model for testing purposes
+    BASE_CONFIG = {
+        "vocab_size": 50257,
+        "context_length": 120,
+        "drop_rate": 0.0,
+        "qkv_bias": False,
+        "emb_dim": 12,
+        "n_layers": 1,
+        "n_heads": 2
+    }
+    model = GPTModel(BASE_CONFIG)
+    model.eval()
+    device = "cpu"
+
+    ########################################
+    # Modify and pretrained model
+    ########################################
+
+    for param in model.parameters():
+        param.requires_grad = False
+
+    torch.manual_seed(123)
+
+    num_classes = 2
+    model.out_head = torch.nn.Linear(in_features=BASE_CONFIG["emb_dim"], out_features=num_classes)
+    replace_linear_with_lora(model, rank=16, alpha=16)
+    model.to(device)
+
+    for param in model.trf_blocks[-1].parameters():
+        param.requires_grad = True
+
+    for param in model.final_norm.parameters():
+        param.requires_grad = True
+
+    ########################################
+    # Finetune modified model
+    ########################################
+
+    torch.manual_seed(123)
+
+    optimizer = torch.optim.AdamW(model.parameters(), lr=5e-5, weight_decay=0.1)
+
+    train_subset = Subset(train_loader.dataset, range(5))
+    batch_train_loader = DataLoader(train_subset, batch_size=5)
+    val_subset = Subset(val_loader.dataset, range(5))
+    batch_val_loader = DataLoader(val_subset, batch_size=5)
+
+    num_epochs = 6
+    train_losses, val_losses, train_accs, val_accs, examples_seen = train_classifier_simple(
+        model, batch_train_loader, batch_val_loader, optimizer, device,
+        num_epochs=num_epochs, eval_freq=1, eval_iter=1,
+    )
+
+    assert round(train_losses[0], 1) == 0.8
+    assert round(val_losses[0], 1) == 0.8
+    assert train_losses[-1] < train_losses[0]

pkg/llms_from_scratch/tests/test_ch02.py

@@ -0,0 +1,54 @@
+# Copyright (c) Sebastian Raschka under Apache License 2.0 (see LICENSE.txt).
+# Source for "Build a Large Language Model From Scratch"
+#   - https://www.manning.com/books/build-a-large-language-model-from-scratch
+# Code: https://github.com/rasbt/LLMs-from-scratch
+
+from llms_from_scratch.ch02 import create_dataloader_v1
+
+import os
+import urllib.request
+
+import pytest
+import torch
+
+
+@pytest.mark.parametrize("file_name", ["the-verdict.txt"])
+def test_dataloader(tmp_path, file_name):
+
+    if not os.path.exists("the-verdict.txt"):
+        url = ("https://raw.githubusercontent.com/rasbt/"
+               "LLMs-from-scratch/main/ch02/01_main-chapter-code/"
+               "the-verdict.txt")
+        file_path = "the-verdict.txt"
+        urllib.request.urlretrieve(url, file_path)
+
+    with open("the-verdict.txt", "r", encoding="utf-8") as f:
+        raw_text = f.read()
+
+    vocab_size = 50257
+    output_dim = 256
+    context_length = 1024
+
+    token_embedding_layer = torch.nn.Embedding(vocab_size, output_dim)
+    pos_embedding_layer = torch.nn.Embedding(context_length, output_dim)
+
+    batch_size = 8
+    max_length = 4
+    dataloader = create_dataloader_v1(
+        raw_text,
+        batch_size=batch_size,
+        max_length=max_length,
+        stride=max_length
+    )
+
+    for batch in dataloader:
+        x, y = batch
+
+        token_embeddings = token_embedding_layer(x)
+        pos_embeddings = pos_embedding_layer(torch.arange(max_length))
+
+        input_embeddings = token_embeddings + pos_embeddings
+
+        break
+
+    input_embeddings.shape == torch.Size([8, 4, 256])

pkg/llms_from_scratch/tests/test_ch03.py

@@ -0,0 +1,22 @@
+# Copyright (c) Sebastian Raschka under Apache License 2.0 (see LICENSE.txt).
+# Source for "Build a Large Language Model From Scratch"
+#   - https://www.manning.com/books/build-a-large-language-model-from-scratch
+# Code: https://github.com/rasbt/LLMs-from-scratch
+
+
+from llms_from_scratch.ch03 import MultiHeadAttention
+import torch
+
+
+def test_mha():
+
+    context_length = 100
+    d_in = 256
+    d_out = 16
+
+    mha = MultiHeadAttention(d_in, d_out, context_length, 0.0, num_heads=2)
+
+    batch = torch.rand(8, 6, d_in)
+    context_vecs = mha(batch)
+
+    context_vecs.shape == torch.Size([8, 6, d_out])

pkg/llms_from_scratch/tests/test_ch04.py

@@ -0,0 +1,50 @@
+# Copyright (c) Sebastian Raschka under Apache License 2.0 (see LICENSE.txt).
+# Source for "Build a Large Language Model From Scratch"
+#   - https://www.manning.com/books/build-a-large-language-model-from-scratch
+# Code: https://github.com/rasbt/LLMs-from-scratch
+
+from llms_from_scratch.ch04 import GPTModel
+from llms_from_scratch.ch04 import generate_text_simple
+
+import torch
+import tiktoken
+
+
+def test_GPTModel():
+    GPT_CONFIG_124M = {
+        "vocab_size": 50257,     # Vocabulary size
+        "context_length": 1024,  # Context length
+        "emb_dim": 768,          # Embedding dimension
+        "n_heads": 12,           # Number of attention heads
+        "n_layers": 12,          # Number of layers
+        "drop_rate": 0.1,        # Dropout rate
+        "qkv_bias": False        # Query-Key-Value bias
+    }
+
+    torch.manual_seed(123)
+    model = GPTModel(GPT_CONFIG_124M)
+    model.eval()  # disable dropout
+
+    start_context = "Hello, I am"
+
+    tokenizer = tiktoken.get_encoding("gpt2")
+    encoded = tokenizer.encode(start_context)
+    encoded_tensor = torch.tensor(encoded).unsqueeze(0)
+
+    print(f"\n{50*'='}\n{22*' '}IN\n{50*'='}")
+    print("\nInput text:", start_context)
+    print("Encoded input text:", encoded)
+    print("encoded_tensor.shape:", encoded_tensor.shape)
+
+    out = generate_text_simple(
+        model=model,
+        idx=encoded_tensor,
+        max_new_tokens=10,
+        context_size=GPT_CONFIG_124M["context_length"]
+    )
+
+    expect = torch.tensor([
+        [15496,   11,   314,   716, 27018, 24086, 47843, 30961, 42348,  7267,
+         49706, 43231, 47062, 34657]
+    ])
+    torch.equal(expect, out)

pkg/llms_from_scratch/tests/test_ch05.py

@@ -0,0 +1,115 @@
+# Copyright (c) Sebastian Raschka under Apache License 2.0 (see LICENSE.txt).
+# Source for "Build a Large Language Model From Scratch"
+#   - https://www.manning.com/books/build-a-large-language-model-from-scratch
+# Code: https://github.com/rasbt/LLMs-from-scratch
+
+from llms_from_scratch.ch02 import create_dataloader_v1
+from llms_from_scratch.ch04 import GPTModel
+from llms_from_scratch.ch05 import train_model_simple
+
+import os
+import urllib
+
+import pytest
+import tiktoken
+import torch
+from torch.utils.data import Subset, DataLoader
+
+
+@pytest.mark.parametrize("file_name", ["the-verdict.txt"])
+def test_train_simple(tmp_path, file_name):
+
+    GPT_CONFIG_124M = {
+        "vocab_size": 50257,    # Vocabulary size
+        "context_length": 256,  # Shortened context length (orig: 1024)
+        "emb_dim": 768,         # Embedding dimension
+        "n_heads": 12,          # Number of attention heads
+        "n_layers": 12,         # Number of layers
+        "drop_rate": 0.1,       # Dropout rate
+        "qkv_bias": False       # Query-key-value bias
+    }
+
+    OTHER_SETTINGS = {
+        "learning_rate": 5e-4,
+        "num_epochs": 2,
+        "batch_size": 1,
+        "weight_decay": 0.1
+    }
+
+    torch.manual_seed(123)
+    device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+    ##############################
+    # Download data if necessary
+    ##############################
+
+    file_path = tmp_path / "the-verdict.txt"
+    url = "https://raw.githubusercontent.com/rasbt/LLMs-from-scratch/main/ch02/01_main-chapter-code/the-verdict.txt"
+
+    if not os.path.exists(file_path):
+        with urllib.request.urlopen(url) as response:
+            text_data = response.read().decode('utf-8')
+        with open(file_path, "w", encoding="utf-8") as file:
+            file.write(text_data)
+    else:
+        with open(file_path, "r", encoding="utf-8") as file:
+            text_data = file.read()
+
+    ##############################
+    # Initialize model
+    ##############################
+
+    model = GPTModel(GPT_CONFIG_124M)
+    model.to(device)  # no assignment model = model.to(device) necessary for nn.Module classes
+    optimizer = torch.optim.AdamW(
+        model.parameters(), lr=OTHER_SETTINGS["learning_rate"], weight_decay=OTHER_SETTINGS["weight_decay"]
+    )
+
+    ##############################
+    # Set up dataloaders
+    ##############################
+
+    # Train/validation ratio
+    train_ratio = 0.90
+    split_idx = int(train_ratio * len(text_data))
+
+    train_loader = create_dataloader_v1(
+        text_data[:split_idx],
+        batch_size=OTHER_SETTINGS["batch_size"],
+        max_length=GPT_CONFIG_124M["context_length"],
+        stride=GPT_CONFIG_124M["context_length"],
+        drop_last=True,
+        shuffle=True,
+        num_workers=0
+    )
+
+    val_loader = create_dataloader_v1(
+        text_data[split_idx:],
+        batch_size=OTHER_SETTINGS["batch_size"],
+        max_length=GPT_CONFIG_124M["context_length"],
+        stride=GPT_CONFIG_124M["context_length"],
+        drop_last=False,
+        shuffle=False,
+        num_workers=0
+    )
+
+    ##############################
+    # Train model
+    ##############################
+
+    tokenizer = tiktoken.get_encoding("gpt2")
+
+    train_subset = Subset(train_loader.dataset, range(1))
+    one_batch_train_loader = DataLoader(train_subset, batch_size=1)
+    val_subset = Subset(val_loader.dataset, range(1))
+    one_batch_val_loader = DataLoader(val_subset, batch_size=1)
+
+    train_losses, val_losses, tokens_seen = train_model_simple(
+        model, one_batch_train_loader, one_batch_val_loader, optimizer, device,
+        num_epochs=OTHER_SETTINGS["num_epochs"], eval_freq=1, eval_iter=1,
+        start_context="Every effort moves you", tokenizer=tokenizer
+    )
+
+    assert round(train_losses[0], 1) == 7.6
+    assert round(val_losses[0], 1) == 10.1
+    assert train_losses[-1] < train_losses[0]

pkg/llms_from_scratch/tests/test_ch06.py

@@ -0,0 +1,148 @@
+# Copyright (c) Sebastian Raschka under Apache License 2.0 (see LICENSE.txt).
+# Source for "Build a Large Language Model From Scratch"
+#   - https://www.manning.com/books/build-a-large-language-model-from-scratch
+# Code: https://github.com/rasbt/LLMs-from-scratch
+
+
+from llms_from_scratch.ch04 import GPTModel
+from llms_from_scratch.ch06 import (
+    download_and_unzip_spam_data, create_balanced_dataset,
+    random_split, SpamDataset, train_classifier_simple
+)
+
+from pathlib import Path
+import urllib
+
+import pandas as pd
+import tiktoken
+import torch
+from torch.utils.data import DataLoader, Subset
+
+
+def test_train_classifier(tmp_path):
+
+    ########################################
+    # Download and prepare dataset
+    ########################################
+
+    url = "https://archive.ics.uci.edu/static/public/228/sms+spam+collection.zip"
+    zip_path = tmp_path / "sms_spam_collection.zip"
+    extracted_path = tmp_path / "sms_spam_collection"
+    data_file_path = Path(extracted_path) / "SMSSpamCollection.tsv"
+
+    try:
+        download_and_unzip_spam_data(
+            url, zip_path, extracted_path, data_file_path
+        )
+    except (urllib.error.HTTPError, urllib.error.URLError, TimeoutError) as e:
+        print(f"Primary URL failed: {e}. Trying backup URL...")
+        backup_url = "https://f001.backblazeb2.com/file/LLMs-from-scratch/sms%2Bspam%2Bcollection.zip"
+        download_and_unzip_spam_data(
+            backup_url, zip_path, extracted_path, data_file_path
+        )
+
+    df = pd.read_csv(data_file_path, sep="\t", header=None, names=["Label", "Text"])
+    balanced_df = create_balanced_dataset(df)
+    balanced_df["Label"] = balanced_df["Label"].map({"ham": 0, "spam": 1})
+
+    train_df, validation_df, test_df = random_split(balanced_df, 0.7, 0.1)
+    train_df.to_csv(tmp_path / "train.csv", index=None)
+    validation_df.to_csv(tmp_path / "validation.csv", index=None)
+    test_df.to_csv(tmp_path / "test.csv", index=None)
+
+    ########################################
+    # Create data loaders
+    ########################################
+    tokenizer = tiktoken.get_encoding("gpt2")
+
+    train_dataset = SpamDataset(
+        csv_file=tmp_path / "train.csv",
+        max_length=None,
+        tokenizer=tokenizer
+    )
+
+    val_dataset = SpamDataset(
+        csv_file=tmp_path / "validation.csv",
+        max_length=train_dataset.max_length,
+        tokenizer=tokenizer
+    )
+
+    num_workers = 0
+    batch_size = 8
+
+    torch.manual_seed(123)
+
+    train_loader = DataLoader(
+        dataset=train_dataset,
+        batch_size=batch_size,
+        shuffle=True,
+        num_workers=num_workers,
+        drop_last=True,
+    )
+
+    val_loader = DataLoader(
+        dataset=val_dataset,
+        batch_size=batch_size,
+        num_workers=num_workers,
+        drop_last=False,
+    )
+
+    ########################################
+    # Load pretrained model
+    ########################################
+
+    # Small GPT model for testing purposes
+    BASE_CONFIG = {
+        "vocab_size": 50257,
+        "context_length": 120,
+        "drop_rate": 0.0,
+        "qkv_bias": False,
+        "emb_dim": 12,
+        "n_layers": 1,
+        "n_heads": 2
+    }
+    model = GPTModel(BASE_CONFIG)
+    model.eval()
+    device = "cpu"
+
+    ########################################
+    # Modify and pretrained model
+    ########################################
+
+    for param in model.parameters():
+        param.requires_grad = False
+
+    torch.manual_seed(123)
+
+    num_classes = 2
+    model.out_head = torch.nn.Linear(in_features=BASE_CONFIG["emb_dim"], out_features=num_classes)
+    model.to(device)
+
+    for param in model.trf_blocks[-1].parameters():
+        param.requires_grad = True
+
+    for param in model.final_norm.parameters():
+        param.requires_grad = True
+
+    ########################################
+    # Finetune modified model
+    ########################################
+
+    torch.manual_seed(123)
+
+    optimizer = torch.optim.AdamW(model.parameters(), lr=5e-5, weight_decay=0.0)
+
+    train_subset = Subset(train_loader.dataset, range(5))
+    batch_train_loader = DataLoader(train_subset, batch_size=5)
+    val_subset = Subset(val_loader.dataset, range(5))
+    batch_val_loader = DataLoader(val_subset, batch_size=5)
+
+    num_epochs = 5
+    train_losses, val_losses, train_accs, val_accs, examples_seen = train_classifier_simple(
+        model, batch_train_loader, batch_val_loader, optimizer, device,
+        num_epochs=num_epochs, eval_freq=1, eval_iter=1,
+    )
+
+    assert round(train_losses[0], 1) == 0.8
+    assert round(val_losses[0], 1) == 0.8
+    assert train_losses[-1] < train_losses[0]

pkg/llms_from_scratch/tests/test_ch07.py

@@ -0,0 +1,108 @@
+# Copyright (c) Sebastian Raschka under Apache License 2.0 (see LICENSE.txt).
+# Source for "Build a Large Language Model From Scratch"
+#   - https://www.manning.com/books/build-a-large-language-model-from-scratch
+# Code: https://github.com/rasbt/LLMs-from-scratch
+
+from llms_from_scratch.ch04 import GPTModel
+from llms_from_scratch.ch05 import train_model_simple
+from llms_from_scratch.ch07 import (
+    download_and_load_file, InstructionDataset, format_input, custom_collate_fn
+)
+
+from functools import partial
+
+import torch
+from torch.utils.data import DataLoader
+import tiktoken
+
+
+def test_instruction_finetune(tmp_path):
+
+    #######################################
+    # Download and prepare dataset
+    #######################################
+    file_path = tmp_path / "instruction-data.json"
+    url = "https://raw.githubusercontent.com/rasbt/LLMs-from-scratch/main/ch07/01_main-chapter-code/instruction-data.json"
+    data = download_and_load_file(file_path, url)
+
+    train_portion = int(len(data) * 0.85)  # 85% for training
+    test_portion = int(len(data) * 0.1)    # 10% for testing
+
+    train_data = data[:train_portion]
+    test_data = data[train_portion:train_portion + test_portion]
+    val_data = data[train_portion + test_portion:]
+
+    # Use very small subset for testing purposes
+    train_data = train_data[:15]
+    val_data = val_data[:15]
+    test_data = test_data[:15]
+
+    tokenizer = tiktoken.get_encoding("gpt2")
+    device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+    customized_collate_fn = partial(custom_collate_fn, device=device, allowed_max_length=100)
+
+    num_workers = 0
+    batch_size = 8
+
+    torch.manual_seed(123)
+
+    train_dataset = InstructionDataset(train_data, tokenizer)
+    train_loader = DataLoader(
+        train_dataset,
+        batch_size=batch_size,
+        collate_fn=customized_collate_fn,
+        shuffle=True,
+        drop_last=True,
+        num_workers=num_workers
+    )
+
+    val_dataset = InstructionDataset(val_data, tokenizer)
+    val_loader = DataLoader(
+        val_dataset,
+        batch_size=batch_size,
+        collate_fn=customized_collate_fn,
+        shuffle=False,
+        drop_last=False,
+        num_workers=num_workers
+    )
+
+    #######################################
+    # Load pretrained model
+    #######################################
+
+    # Small GPT model for testing purposes
+    BASE_CONFIG = {
+        "vocab_size": 50257,
+        "context_length": 120,
+        "drop_rate": 0.0,
+        "qkv_bias": False,
+        "emb_dim": 12,
+        "n_layers": 1,
+        "n_heads": 2
+    }
+    model = GPTModel(BASE_CONFIG)
+    model.eval()
+    device = "cpu"
+    CHOOSE_MODEL = "Small test model"
+
+    print("Loaded model:", CHOOSE_MODEL)
+    print(50*"-")
+
+    #######################################
+    # Finetuning the model
+    #######################################
+
+    num_epochs = 10
+    optimizer = torch.optim.AdamW(model.parameters(), lr=5e-5, weight_decay=0.1)
+
+    torch.manual_seed(123)
+    train_losses, val_losses, tokens_seen = train_model_simple(
+        model, train_loader, val_loader, optimizer, device,
+        num_epochs=num_epochs, eval_freq=5, eval_iter=5,
+        start_context=format_input(val_data[0]), tokenizer=tokenizer
+    )
+
+    assert round(train_losses[0], 1) == 10.9
+    assert round(val_losses[0], 1) == 10.9
+    assert train_losses[-1] < train_losses[0]

pyproject.toml

@@ -1,6 +1,10 @@
+[build-system]
+requires = ["setuptools>=61.0", "wheel"]
+build-backend = "setuptools.build_meta"
+
 [project]
 name = "llms-from-scratch"
-version = "0.1.0"
+version = "1.0.0"
 description = "Implement a ChatGPT-like LLM in PyTorch from scratch, step by step"
 readme = "README.md"
 requires-python = ">=3.10"
@@ -14,17 +18,17 @@ dependencies = [
     "numpy>=1.26,<2.1",
     "pandas>=2.2.1",
     "pip>=25.0.1",
+    "pytest>=8.3.5",
 ]
 
-[tool.setuptools.packages]
-find = {}
-
 [tool.uv.sources]
 llms-from-scratch = { workspace = true }
 
 [dependency-groups]
 dev = [
+    "build>=1.2.2.post1",
     "llms-from-scratch",
+    "twine>=6.1.0",
 ]
 
 [tool.ruff]
@@ -37,3 +41,12 @@ ignore = [
     "C406", "E226", "E402", "E702", "E703",
     "E722", "E731", "E741"
 ]
+
+
+# `llms_from_scratch` PyPI package
+
+[tool.setuptools]
+package-dir = {"" = "pkg"}
+
+[tool.setuptools.packages.find]
+where = ["pkg"]