1. Did you check for GPU availability and set the device accordingly?

✅ Yes
❌ No

Explanation:
Checking torch.cuda.is_available() ensures you use GPU acceleration for faster training when available.

---

2. Did you define data transformations such as normalization or augmentation?

✅ Yes
❌ No

Explanation:
Transforms standardize inputs and help the model generalize better, especially normalization.

---

3. Did you define a custom CNN class inheriting from nn.Module?

✅ Yes
❌ No

Explanation:
Subclassing nn.Module establishes your model architecture and registers trainable parameters.

---

4. Did you implement a forward method?

✅ Yes
❌ No

Explanation:
The forward() method defines how data flows through the network and enables autograd for backpropagation.

---

5. Did you initialize the loss function and optimizer?

✅ Yes
❌ No

Explanation:
Loss (e.g., CrossEntropyLoss) and optimizer (e.g., Adam) are essential for learning and updating parameters.

---

6. Did you iterate the training loop over epochs and batches?

✅ Yes
❌ No

Explanation:
The training loop performs forward pass → loss → backward pass → optimizer step.

---

7. Did you evaluate model performance during the validation loop without gradient updates?

❌ No
✅ Yes

Explanation:
Validation must use torch.no_grad() to disable gradient tracking and compute metrics efficiently.

---

8. Did the code compute accuracy metrics on the validation set?

✅ Yes
❌ No

Explanation:
Accuracy helps quantify how well the classifier predicts on unseen data.

---

9. Did the notebook plot training and validation loss curves?

✅ Yes
❌ No

Explanation:
Loss curves show training behavior and help detect underfitting or overfitting.

---

10. Did the code save the trained model’s state dictionary?

✅ Yes
❌ No

Explanation:
Saving model.state_dict() allows reloading the model for inference or future fine-tuning.

---

11. Did you complete all tasks and download the notebook?

✅ Yes
❌ No

Explanation:
Downloading the notebook is essential for submission and final evaluation.

---

🧾 Summary Table

---