-
Notifications
You must be signed in to change notification settings - Fork 1
/
helpers_layer.py
420 lines (341 loc) · 14.6 KB
/
helpers_layer.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
from torch import Tensor, empty
from typing import Optional, Tuple, Union, List
import helpers_functional as F
class Module(object):
def forward(self, *input):
raise NotImplementedError
def backward(self,*gradwrtoutput):
raise NotImplementedError
def param(self):
return [] # It should have (param, grad) or (param, None)
class Sequential(Module):
def __init__(self, modules):
self.modules = modules
self.input = None
def forward(self, input):
self.input = input
output = self.input
for module in self.modules:
output = module.forward(output)
return output
def backward(self,*gradwrtouput):
gradient = gradwrtouput
for module in reversed(self.modules):
gradient = module.backward(gradient)
self.input = None
return gradient
def param(self):
params = []
for module in self.modules:
params.append(module.param())
return params
class Optimizer(object):
def step(self):
return NotImplementedError
def zero_grad(self):
return NotImplementedError
class SGD(Optimizer):
def __init__(self, params, lr, mu = 0, tau = 0):
self.params = params
self.lr = lr
# parameters in order to add momemtum
self.momemtum = mu
self.dampening = tau
self.state_momemtum = None
def step(self):
for x, grad in self.params: # Make sure that params have (param, grad)
x.add_(-self.lr * grad) #TODO: Make sure it updates params
def zero_grad(self):
for x, grad in self.params: # TODO: Make sure that params have (param, grad)
grad = grad.zero_() #TODO: Update these grad
class MSE(Module):
def forward(self, input, target):
self.input = input
self.target = target
return (self.input - self.target).pow(2).mean()
def backward(self):
return 2*(self.input - self.target).div(torch.tensor(self.input.size(0)))
## we divide by the batch size as in Pytorch
class Sigmoid(Module):
def forward(self,input):
self.input = input
self.sigmoid = 1./(1+(-self.input).exp())
return self.sigmoid
def backward(self,*gradwrtouput):
return gradwrtouput*self.sigmoid*(1-self.sigmoid)
class ReLU(Module):
def forward(self, input):
self.input = input
return (self.input>0.)*self.input
def backward(self, *gradwrtouput):
return gradwrtouput*(self.input>=0.)
# TODO: Groups argument & padding mode is not taken care!
class _ConvNd(Module):
# Base Class for Convolution Layers
__constants__ = ['stride', 'padding', 'dilation', 'groups', 'padding_mode', 'in_channels', 'out_channels', 'kernel_size']
__annotations__ = {'bias': Optional[Tensor]}
def _conv_forward(self, input: Tensor, bias: Optional[Tensor]) -> Tensor:
...
_in_channels: int
out_channels: int
kernel_size: Tuple[int, ...]
stride: Tuple[int, ...]
padding: Union[str, Tuple[int, ...]]
dilation: Tuple[int, ...]
transposed: bool
output_padding: Tuple[int, ...]
groups: int
padding_mode: str
weight: Tensor
bias: Optional[Tensor]
def __init__(self, in_channels: int,
out_channels: int, kernel_size: Tuple[int, ...],
stride: Tuple[int, ...], padding: Tuple[int, ...],
dilation: Tuple[int, ...], transposed: bool,
output_padding: Tuple[int, ...],
groups: int, bias: bool, grad_bool: bool, padding_mode: str,
device=None, dtype=None) -> None:
factory_kwargs = {'device': device, 'dtype': dtype}
super(_ConvNd, self).__init__()
if in_channels % groups != 0:
raise ValueError('in_channels must be divisible by groups')
if out_channels % groups != 0:
raise ValueError('out_channels must be divisible by groups')
valid_padding_strings = {'same', 'valid'}
if isinstance(padding, str):
if padding not in valid_padding_strings:
raise ValueError("Invalid padding string {!r}, should be one of {}".format(padding, valid_padding_strings))
if padding == 'same' and any(s != 1 for s in stride):
raise ValueError("padding='same' not supported for strided convolutions")
valid_padding_modes = {'zeros', 'reflect', 'replicate', 'circular'}
if padding_mode not in valid_padding_modes:
raise ValueError("padding_mode must be one of {}, but got padding_mode={}".format(valid_padding_modes, padding_mode))
self.in_channels = in_channels
self.out_channels = out_channels
self.kernel_size = kernel_size
self.stride = stride
self.padding = padding
self.dilation = dilation
self.transposed = transposed
self.output_padding = output_padding
self.groups = groups
self.padding_mode = padding_mode
self.grad_bool = grad_bool
# TODO: Replace Parameter class
if transposed:
self.weight = empty(
(in_channels, out_channels // groups, *kernel_size), **factory_kwargs
)
if self.grad_bool:
self.grad_weight = empty(
(in_channels, out_channels // groups, *kernel_size), **factory_kwargs
).fill_(0.0)
else:
self.grad_weight = None
else:
self.weight = empty(
(out_channels, in_channels // groups, *kernel_size), **factory_kwargs
)
if self.grad_bool:
self.grad_weight = empty(
(out_channels, in_channels // groups, *kernel_size), **factory_kwargs
)
else:
self.grad_weight = None
if bias:
self.bias = empty(out_channels, **factory_kwargs)
if self.grad_bool:
self.grad_bias = empty(out_channels, **factory_kwargs)
else:
self.grad_bias = None
else:
self.bias = None
self.reset_params()
def reset_params(self) -> None:
# TODO: Reset weight and bias. Sample from U(-sqrt(k), sqrt(k))
if isinstance(self.kernel_size, int):
prod_kernel_size = self.kernel_size**2
else:
prod_kernel_size = self.kernel_size[0]*self.kernel_size[1]
sqrt_k = 1/(self.weight.size(1) * prod_kernel_size)**(0.5)
self.weight.uniform_(-sqrt_k, sqrt_k)
if self.bias is not None:
self.bias.uniform_(-sqrt_k, sqrt_k)
if self.grad_bool:
# TODO: DO SOMETHING MEANINGFUL
pass
def reset_grad(self) -> None:
# TODO: DO SOMETHING MEANINGFUL
self.grad_weight = None
self.grad_bias = None
# TODO: F should have pad,
def _pair(x):
if isinstance(x, int):
return (x, x)
return x
class Conv2d(_ConvNd):
# Convolution 2d layer Class
def __init__(
self,
in_channels: int,
out_channels: int,
kernel_size,
stride = 1,
padding = 0,
dilation = 1,
groups: int = 1,
bias: bool = True,
grad_bool: bool = True,
padding_mode: str = 'zeros',
device = None,
dtype = None,
) -> None:
# Change Parameters Accordingly
factory_kwargs = {'device': device, 'dtype': dtype}
kernel_size_ = _pair(kernel_size)
stride_ = _pair(stride)
padding_ = padding if isinstance(padding, str) else _pair(padding)
dilation_ = _pair(dilation)
super(Conv2d, self).__init__(
in_channels, out_channels, kernel_size_, stride_, padding_, dilation_, False,
_pair(0), groups, bias, grad_bool, padding_mode, **factory_kwargs
)
self.input = None
def _conv_forward(self, input: Tensor, weight: Tensor, bias: Optional[Tensor]):
if self.padding_mode != 'zeros':
# TODO: Handle non zeros padding. Replace REVERSED_PADDING
return F.conv2d(F.pad(input, REVERSED_PADDING, mode=self.padding_mode),
weight, bias, self.stride,
_pair(0), self.dilation, self.groups)
return F.conv2d(input, weight, bias, self.stride,
self.padding, self.dilation, self.groups)
def forward(self, input: Tensor) -> Tensor:
self.input = input
return self._conv_forward(input, self.weight, self.bias)
def __call__(self, input: Tensor) -> Tensor:
return self.forward(input)
def backward(self, grad_output):
grad_input = None
if self.input == None:
raise ValueError("Forward is not implemented, so backward cannot be implemented.")
grad_input = F.conv2d_grad_input(input_shape=self.input.shape, weight=self.weight, grad_output=grad_output, stride=self.stride, padding=self.padding, dilation=self.dilation, groups=self.groups)
self.grad_weight = F.conv2d_grad_weight(self.input, self.weight.shape, grad_output, self.stride, self.padding, self.dilation, self.groups)
if self.bias is not None:
self.grad_bias = grad_output.sum((0, 2, 3)).squeeze(0)
return grad_input
# TODO: Check if it's correct or not, It should work with SGD
def param(self):
params = []
params.append(self.weight)
params.append(self.grad_weight)
params.append(self.bias)
params.append(self.grad_bias)
return params
def update_params(self, params):
self.weight = params[0]
self.grad_weight = params[1]
self.bias = params[2]
self.grad_bias = params[3]
class _ConvTransposeNd(_ConvNd):
# Base Class for Transpose Convolution Layers
def __init__(self, in_channels: int,
out_channels: int, kernel_size: Tuple[int,...],
stride: Tuple[int,...], padding: Tuple[int,...],
dilation: Tuple[int,...], transposed: bool,
output_padding: Tuple[int,...],
groups: int, bias: bool, grad_bool: bool, padding_mode: str,
device=None, dtype=None) -> None:
if padding_mode != 'zeros':
raise ValueError('Only "zeros" padding mode is supported for {}'.format(self.__class__.__name__))
factory_kwargs = {'device': device, 'dtype': dtype}
super(_ConvTransposeNd, self).__init__(
in_channels, out_channels, kernel_size, stride,
padding, dilation, transposed, output_padding,
groups, bias, grad_bool, padding_mode, **factory_kwargs)
def _output_padding(self, input: Tensor, output_size: Optional[List[int]],
stride: List[int], padding: List[int], kernel_size: List[int],
dilation: Optional[List[int]] = None) -> List[int]:
if output_size is None:
if isinstance(self.output_padding, int):
ret = self.output_padding
else:
ret = self.output_padding[0]
else:
k = input.dim() - 2
if len(output_size) == k + 2:
output_size = output_size[2:]
if len(output_size) != k:
raise ValueError("output_size must have {} or {} elements (got {})"
.format(k, k + 2, len(output_size)))
min_sizes = []
max_sizes = []
for d in range(k):
dim_size = ((input.size(d + 2) - 1) * stride[d] -
2 * padding[d] +
(dilation[d] if (dilation is not None) else 1) *
(kernel_size[d] - 1) + 1)
min_sizes.append(dim_size)
max_sizes.append(min_sizes[d] + stride[d] - 1)
for i in range(len(output_size)):
size = output_size[i]
min_size = min_sizes[i]
max_size = max_sizes[i]
if size < min_size or size > max_size:
raise ValueError(
"requested an output size of {}, but valid sizes range from {} to {} (for an input of {})"
.format(output_size, min_sizes, max_sizes, input.size()[2:])
)
res = []
for d in range(k):
res.append(output_size[d] - min_sizes[d])
ret = res
return ret
class ConvTranspose2d(_ConvTransposeNd):
def __init__(
self,
in_channels: int,
out_channels: int,
kernel_size,
stride = 1,
padding = 0,
output_padding = 0,
groups = 1,
bias: bool = True,
dilation: int = 1,
padding_mode: str = 'zeros',
grad_bool: bool = True,
device = None,
dtype = None,
) -> None:
factory_kwargs = {'device': device, 'dtype': dtype}
kernel_size = _pair(kernel_size)
stride_ = _pair(stride)
padding = _pair(padding)
dilation = _pair(dilation)
output_padding = _pair(output_padding)
super(ConvTranspose2d, self).__init__(
in_channels, out_channels, kernel_size, stride_, padding,
dilation, True, output_padding, groups, bias, grad_bool,
padding_mode, **factory_kwargs)
def forward(self, input: Tensor, output_size: Optional[List[int]] = None) -> Tensor:
if self.padding_mode != 'zeros':
raise ValueError('Only "zeros" padding mode is supported for ConvTranspose2d')
assert isinstance(self.padding, tuple)
output_padding = self._output_padding(
input, output_size, self.stride, self.padding, self.kernel_size,
self.dilation)
return F.conv_transpose2d(
input, self.weight, self.bias, stride=self.stride, padding=self.padding,
output_padding=output_padding, groups=self.groups, dilation=self.dilation)
def __call__(self, input: Tensor) -> Tensor:
return self.forward(input)
"""
# Check if it's correct or not
def param(self):
params = []
params.append(self.weight)
params.append(self.grad_weight)
params.append(self.bias)
params.append(self.grad_bias)
return params
"""