mirror of
https://github.com/MLSysBook/TinyTorch.git
synced 2026-03-11 21:23:33 -05:00
da6e4374e0
This commit includes: - Exported tinytorch package files from nbdev (autograd, losses, optimizers, training, etc.) - Updated activations.py and layers.py with __call__ methods - New module exports: attention, spatial, tokenization, transformer, etc. - Removed old _modidx.py file - Cleanup of duplicate milestone directories These are the generated package files that correspond to the source modules we've been developing. Students will import from these when using TinyTorch.
86 lines
2.7 KiB
Python
Generated
86 lines
2.7 KiB
Python
Generated
# AUTOGENERATED! DO NOT EDIT! File to edit: ../../modules/source/18_compression/compression_dev.ipynb.
|
|
|
|
# %% auto 0
|
|
__all__ = ['Tensor', 'Linear', 'Sequential']
|
|
|
|
# %% ../../modules/source/18_compression/compression_dev.ipynb 1
|
|
import numpy as np
|
|
import copy
|
|
from typing import List, Dict, Any, Tuple, Optional
|
|
import time
|
|
|
|
# Import from previous modules
|
|
# Note: In the full package, these would be imports like:
|
|
# from tinytorch.core.tensor import Tensor
|
|
# from tinytorch.core.layers import Linear
|
|
# For development, we'll create minimal implementations
|
|
|
|
class Tensor:
|
|
"""Minimal Tensor class for compression development - imports from Module 01 in practice."""
|
|
def __init__(self, data, requires_grad=False):
|
|
self.data = np.array(data)
|
|
self.shape = self.data.shape
|
|
self.size = self.data.size
|
|
self.requires_grad = requires_grad
|
|
self.grad = None
|
|
|
|
def __add__(self, other):
|
|
if isinstance(other, Tensor):
|
|
return Tensor(self.data + other.data)
|
|
return Tensor(self.data + other)
|
|
|
|
def __mul__(self, other):
|
|
if isinstance(other, Tensor):
|
|
return Tensor(self.data * other.data)
|
|
return Tensor(self.data * other)
|
|
|
|
def matmul(self, other):
|
|
return Tensor(np.dot(self.data, other.data))
|
|
|
|
def abs(self):
|
|
return Tensor(np.abs(self.data))
|
|
|
|
def sum(self, axis=None):
|
|
return Tensor(self.data.sum(axis=axis))
|
|
|
|
def __repr__(self):
|
|
return f"Tensor(shape={self.shape})"
|
|
|
|
class Linear:
|
|
"""Minimal Linear layer for compression development - imports from Module 03 in practice."""
|
|
def __init__(self, in_features, out_features, bias=True):
|
|
self.in_features = in_features
|
|
self.out_features = out_features
|
|
# Initialize with He initialization
|
|
self.weight = Tensor(np.random.randn(in_features, out_features) * np.sqrt(2.0 / in_features))
|
|
self.bias = Tensor(np.zeros(out_features)) if bias else None
|
|
|
|
def forward(self, x):
|
|
output = x.matmul(self.weight)
|
|
if self.bias is not None:
|
|
output = output + self.bias
|
|
return output
|
|
|
|
def parameters(self):
|
|
params = [self.weight]
|
|
if self.bias is not None:
|
|
params.append(self.bias)
|
|
return params
|
|
|
|
class Sequential:
|
|
"""Minimal Sequential container for model compression."""
|
|
def __init__(self, *layers):
|
|
self.layers = list(layers)
|
|
|
|
def forward(self, x):
|
|
for layer in self.layers:
|
|
x = layer.forward(x)
|
|
return x
|
|
|
|
def parameters(self):
|
|
params = []
|
|
for layer in self.layers:
|
|
if hasattr(layer, 'parameters'):
|
|
params.extend(layer.parameters())
|
|
return params
|