Extract and Inspect State Dict

Problem Statement

Extract a model's state dictionary and inspect its contents.

Background

model.state_dict() returns an OrderedDict mapping parameter names to tensors. This is the standard way to save and inspect model weights.

Your Task

The starter code creates a model with named layers. Extract the model's state dictionary to inspect its parameter names and shapes.

The inspection code (listing keys, shapes) is pre-filled.

Output Format

Returns a dictionary with "keys", "num_params", "fc1_weight_shape", and "fc2_bias_shape".

The function state_dict_test() starts by seeding with torch.manual_seed(42) to ensure reproducibility.
It then creates an nn.Sequential model with two named modules: "fc1" (nn.Linear(4, 8)) and "fc2" (nn.Linear(8, 2)), resulting in four parameter tensors: fc1.weight, fc1.bias, fc2.weight, and fc2.bias.
The state dictionary is extracted using model.state_dict(), which returns an OrderedDict containing these four parameter tensors, leading to the list of keys: ['fc1.weight', 'fc1.bias', 'fc2.weight', 'fc2.bias'].
The shapes of fc1.weight and fc2.bias are determined by the nn.Linear layers: fc1.weight has shape $[8, 4]$ (8 outputs, 4 inputs) and fc2.bias has shape $[2]$ (2 outputs), resulting in the output dictionary with the specified values.

Problem Statement

Extract a model's state dictionary and inspect its contents.

Background

model.state_dict() returns an OrderedDict mapping parameter names to tensors. This is the standard way to save and inspect model weights.

Your Task

The starter code creates a model with named layers. Extract the model's state dictionary to inspect its parameter names and shapes.

The inspection code (listing keys, shapes) is pre-filled.

Output Format

Returns a dictionary with "keys", "num_params", "fc1_weight_shape", and "fc2_bias_shape".

The function state_dict_test() starts by seeding with torch.manual_seed(42) to ensure reproducibility.
It then creates an nn.Sequential model with two named modules: "fc1" (nn.Linear(4, 8)) and "fc2" (nn.Linear(8, 2)), resulting in four parameter tensors: fc1.weight, fc1.bias, fc2.weight, and fc2.bias.
The state dictionary is extracted using model.state_dict(), which returns an OrderedDict containing these four parameter tensors, leading to the list of keys: ['fc1.weight', 'fc1.bias', 'fc2.weight', 'fc2.bias'].
The shapes of fc1.weight and fc2.bias are determined by the nn.Linear layers: fc1.weight has shape $[8, 4]$ (8 outputs, 4 inputs) and fc2.bias has shape $[2]$ (2 outputs), resulting in the output dictionary with the specified values.

Problem Statement