"""
Example code of a simple bidirectional LSTM on the MNIST dataset.
Note that using RNNs on image data is not the best idea, but it is a 
good example to show how to use RNNs that still generalizes to other tasks.

Programmed by Aladdin Persson <aladdin.persson at hotmail dot com>
*    2020-05-09 Initial coding
*    2022-12-16 Updated with more detailed comments, docstrings to functions, and checked code still functions as intended.

"""


# Imports
import torch
import torch.nn as nn  # All neural network modules, nn.Linear, nn.Conv2d, BatchNorm, Loss functions
import torch.optim as optim  # For all Optimization algorithms, SGD, Adam, etc.
import torch.nn.functional as F  # All functions that don't have any parameters
from torch.utils.data import (
    DataLoader,
)  # Gives easier dataset managment and creates mini batches
import torchvision.datasets as datasets  # Has standard datasets we can import in a nice way
import torchvision.transforms as transforms  # Transformations we can perform on our dataset
from tqdm import tqdm  # progress bar

# Set device
device = "cuda" if torch.cuda.is_available() else "cpu"

# Hyperparameters
input_size = 28
sequence_length = 28
num_layers = 2
hidden_size = 256
num_classes = 10
learning_rate = 3e-4
batch_size = 64
num_epochs = 2

# Create a bidirectional LSTM
class BRNN(nn.Module):
    def __init__(self, input_size, hidden_size, num_layers, num_classes):
        super(BRNN, self).__init__()
        self.hidden_size = hidden_size
        self.num_layers = num_layers
        self.lstm = nn.LSTM(
            input_size, hidden_size, num_layers, batch_first=True, bidirectional=True
        )
        self.fc = nn.Linear(hidden_size * 2, num_classes)

    def forward(self, x):
        h0 = torch.zeros(self.num_layers * 2, x.size(0), self.hidden_size).to(device)
        c0 = torch.zeros(self.num_layers * 2, x.size(0), self.hidden_size).to(device)

        out, _ = self.lstm(x)
        out = self.fc(out[:, -1, :])

        return out


# Load Data
train_dataset = datasets.MNIST(
    root="dataset/", train=True, transform=transforms.ToTensor(), download=True
)

test_dataset = datasets.MNIST(
    root="dataset/", train=False, transform=transforms.ToTensor(), download=True
)

train_loader = DataLoader(dataset=train_dataset, batch_size=batch_size, shuffle=True)
test_loader = DataLoader(dataset=test_dataset, batch_size=batch_size, shuffle=True)

# Initialize network
model = BRNN(input_size, hidden_size, num_layers, num_classes).to(device)

# Loss and optimizer
criterion = nn.CrossEntropyLoss()
optimizer = optim.Adam(model.parameters(), lr=learning_rate)

# Train Network
for epoch in range(num_epochs):
    for batch_idx, (data, targets) in enumerate(tqdm(train_loader)):
        # Get data to cuda if possible
        data = data.to(device=device).squeeze(1)
        targets = targets.to(device=device)

        # forward
        scores = model(data)
        loss = criterion(scores, targets)

        # backward
        optimizer.zero_grad()
        loss.backward()

        # gradient descent or adam step
        optimizer.step()


# Check accuracy on training & test to see how good our model
def check_accuracy(loader, model):
    if loader.dataset.train:
        print("Checking accuracy on training data")
    else:
        print("Checking accuracy on test data")

    num_correct = 0
    num_samples = 0
    model.eval()

    with torch.no_grad():
        for x, y in loader:
            x = x.to(device=device).squeeze(1)
            y = y.to(device=device)

            scores = model(x)
            _, predictions = scores.max(1)
            num_correct += (predictions == y).sum()
            num_samples += predictions.size(0)

        print(
            f"Got {num_correct} / {num_samples} with accuracy  \
              {float(num_correct)/float(num_samples)*100:.2f}"
        )

    model.train()


check_accuracy(train_loader, model)
check_accuracy(test_loader, model)