Readability and code quality improvements (#959)

* Consistent dataset naming

* consistent section headers

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

@@ -86,7 +86,8 @@
     "id": "3a84cf35-b37f-4c15-8972-dfafc9fadc1c"
    },
    "source": [
-    "## 6.1 Different categories of finetuning"
+    " \n",
+    "### 6.1 Different categories of finetuning"
    ]
   },
   {
@@ -142,7 +143,8 @@
     "id": "8c7017a2-32aa-4002-a2f3-12aac293ccdf"
    },
    "source": [
-    "## 6.2 Preparing the dataset"
+    " \n",
+    "### 6.2 Preparing the dataset"
    ]
   },
   {
@@ -699,7 +701,8 @@
    "id": "a8d7a0c5-1d5f-458a-b685-3f49520b0094",
    "metadata": {},
    "source": [
-    "## 6.3 Creating data loaders"
+    " \n",
+    "### 6.3 Creating data loaders"
    ]
   },
   {
@@ -1019,7 +1022,8 @@
     "id": "d1c4f61a-5f5d-4b3b-97cf-151b617d1d6c"
    },
    "source": [
-    "## 6.4 Initializing a model with pretrained weights"
+    " \n",
+    "### 6.4 Initializing a model with pretrained weights"
    ]
   },
   {
@@ -1219,7 +1223,8 @@
     "id": "4c9ae440-32f9-412f-96cf-fd52cc3e2522"
    },
    "source": [
-    "## 6.5 Adding a classification head"
+    " \n",
+    "### 6.5 Adding a classification head"
    ]
   },
   {
@@ -1722,7 +1727,8 @@
    "id": "32aa4aef-e1e9-491b-9adf-5aa973e59b8c",
    "metadata": {},
    "source": [
-    "## 6.6 Calculating the classification loss and accuracy"
+    " \n",
+    "### 6.6 Calculating the classification loss and accuracy"
    ]
   },
   {
@@ -2042,7 +2048,8 @@
     "id": "456ae0fd-6261-42b4-ab6a-d24289953083"
    },
    "source": [
-    "## 6.7 Finetuning the model on supervised data"
+    " \n",
+    "### 6.7 Finetuning the model on supervised data"
    ]
   },
   {
@@ -2372,7 +2379,8 @@
    "id": "a74d9ad7-3ec1-450e-8c9f-4fc46d3d5bb0",
    "metadata": {},
    "source": [
-    "## 6.8 Using the LLM as a spam classifier"
+    " \n",
+    "### 6.8 Using the LLM as a spam classifier"
    ]
   },
   {
@@ -2564,6 +2572,7 @@
     "id": "5b70ac71-234f-4eeb-b33d-c62726d50cd4"
    },
    "source": [
+    " \n",
     "## Summary and takeaways"
    ]
   },

ch06/02_bonus_additional-experiments/additional_experiments.py

@@ -130,20 +130,20 @@ def download_and_unzip(url, zip_path, extract_to, new_file_path):
     print(f"File downloaded and saved as {new_file_path}")
 
 
-def random_split(df, train_frac, validation_frac):
+def random_split(df, train_frac, val_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)
+    val_end = train_end + int(len(df) * val_frac)
 
     # Split the DataFrame
     train_df = df[:train_end]
-    validation_df = df[train_end:validation_end]
-    test_df = df[validation_end:]
+    val_df = df[train_end:val_end]
+    test_df = df[val_end:]
 
-    return train_df, validation_df, test_df
+    return train_df, val_df, test_df
 
 
 def create_dataset_csvs(new_file_path):
@@ -157,9 +157,9 @@ def create_dataset_csvs(new_file_path):
     balanced_df["Label"] = balanced_df["Label"].map({"ham": 0, "spam": 1})
 
     # Sample and save csv files
-    train_df, validation_df, test_df = random_split(balanced_df, 0.7, 0.1)
+    train_df, val_df, test_df = random_split(balanced_df, 0.7, 0.1)
     train_df.to_csv("train.csv", index=None)
-    validation_df.to_csv("validation.csv", index=None)
+    val_df.to_csv("validation.csv", index=None)
     test_df.to_csv("test.csv", index=None)
 
 
@@ -611,7 +611,7 @@ if __name__ == "__main__":
     base_path = Path(".")
     file_names = ["train.csv", "validation.csv", "test.csv"]
     all_exist = all((base_path / file_name).exists() for file_name in file_names)
-    
+
     if not all_exist:
         try:
             download_and_unzip(url, zip_path, extract_to, new_file_path)

ch06/03_bonus_imdb-classification/sklearn-baseline.ipynb

@@ -144,6 +144,7 @@
    "id": "fae87bc1-14ca-4f89-8e12-49f77b0ec00d",
    "metadata": {},
    "source": [
+    " \n",
     "## Scikit-learn baseline"
    ]
   },
@@ -269,7 +270,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.10.16"
+   "version": "3.13.5"
   }
  },
  "nbformat": 4,

ch06/03_bonus_imdb-classification/train_bert_hf_spam.py

@@ -79,20 +79,20 @@ def download_and_unzip(url, zip_path, extract_to, new_file_path):
     print(f"File downloaded and saved as {new_file_path}")
 
 
-def random_split(df, train_frac, validation_frac):
+def random_split(df, train_frac, val_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)
+    val_end = train_end + int(len(df) * val_frac)
 
     # Split the DataFrame
     train_df = df[:train_end]
-    validation_df = df[train_end:validation_end]
-    test_df = df[validation_end:]
+    val_df = df[train_end:val_end]
+    test_df = df[val_end:]
 
-    return train_df, validation_df, test_df
+    return train_df, val_df, test_df
 
 
 def create_dataset_csvs(new_file_path):
@@ -106,9 +106,9 @@ def create_dataset_csvs(new_file_path):
     balanced_df["Label"] = balanced_df["Label"].map({"ham": 0, "spam": 1})
 
     # Sample and save csv files
-    train_df, validation_df, test_df = random_split(balanced_df, 0.7, 0.1)
+    train_df, val_df, test_df = random_split(balanced_df, 0.7, 0.1)
     train_df.to_csv("train.csv", index=None)
-    validation_df.to_csv("validation.csv", index=None)
+    val_df.to_csv("validation.csv", index=None)
     test_df.to_csv("test.csv", index=None)