Week 7 · Distinguish probability from classification criteria

1. Run the data-loading cell. Print the shape, dtype, and first five rows of x_raw and x separately. Label the outputs before and after preprocessing.
2. In the same cell, print the shape, dtype, and first five values of labels and y. Find where Labels are converted to the form required by Loss.
3. Explain why you must reuse the training data's mean and scale to predict new samples, relating this to the scale of each Feature fed to the model.

1. Call train_classifier with Learning Rates $\eta\in\{0.001,0.01,0.1\}$, each for 1000 epochs. Keep the Seed and model initialization inside the function.
2. Plot the three Loss curves together and tabulate the final Training Loss and Training Accuracy for each Learning Rate.
3. Explain differences in the rate of Loss decrease by pointing to sections of the curves.

1. Choose one model trained in the previous activity and compute Probability once. Record the chosen Learning Rate. Apply Thresholds $\theta\in\{0.3,0.5,0.7\}$ separately to the same Probability.
2. Record Threshold, Accuracy, and Positive prediction count in a three-row table.
3. Choose up to 3 samples whose decisions change with the Threshold. Tabulate the sample index, Probability, target, and decisions at the three Thresholds. If no samples change decisions, state that.

Include code, actual outputs and plots, and your own interpretation. Submit through KLAS; check the semester schedule and latest announcements for the deadline.

Filename: Name_StudentID_Week7.ipynb.

When the BCE Target shape differs

1. After training the model, add a Code Cell at the bottom of the notebook and run the following code.

   with torch.no_grad():
       debug_logits = model(x)
   print("Logits:", debug_logits.shape)
   for debug_target in [y.squeeze(1), y]:
       print("Target:", debug_target.shape)
       try:
           print("Loss:", F.binary_cross_entropy_with_logits(
               debug_logits, debug_target).item())
       except (RuntimeError, ValueError) as error:
           print(type(error).__name__, str(error))

2. In a Text Cell, tabulate the Logits shape, each Target shape, and execution success. Explain which axis squeeze(1) removed and how Logits and Target shapes must match for this Loss.