Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly
Sign up
Appearance settings

Commit 1039e00

Browse files
Add GitHub action and nn.SiLU
1 parent c1ee2d1 commit 1039e00

File tree

3 files changed

+30
-19
lines changed

3 files changed

+30
-19
lines changed

‎.github/workflows/main.yml‎

Lines changed: 5 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,5 @@
1+
- name: Publish a Python distribution to PyPI
2+
uses: pypa/gh-action-pypi-publish@release/v1
3+
with:
4+
user: __token__
5+
password: ${{ secrets.PYPI_API_TOKEN }}

‎efficientnet_pytorch/model.py‎

Lines changed: 8 additions & 7 deletions
Original file line numberDiff line numberDiff line change
@@ -50,7 +50,7 @@ class MBConvBlock(nn.Module):
5050
def __init__(self, block_args, global_params, image_size=None):
5151
super().__init__()
5252
self._block_args = block_args
53-
self._bn_mom = 1 - global_params.batch_norm_momentum # pytorch's difference from tensorflow
53+
self._bn_mom = 1 - global_params.batch_norm_momentum # pytorch's difference from tensorflow
5454
self._bn_eps = global_params.batch_norm_epsilon
5555
self.has_se = (self._block_args.se_ratio is not None) and (0 < self._block_args.se_ratio <= 1)
5656
self.id_skip = block_args.id_skip # whether to use skip connection and drop connect
@@ -196,7 +196,7 @@ def __init__(self, blocks_args=None, global_params=None):
196196
# The first block needs to take care of stride and filter size increase.
197197
self._blocks.append(MBConvBlock(block_args, self._global_params, image_size=image_size))
198198
image_size = calculate_output_image_size(image_size, block_args.stride)
199-
if block_args.num_repeat > 1: # modify block_args to keep same output size
199+
if block_args.num_repeat > 1: # modify block_args to keep same output size
200200
block_args = block_args._replace(input_filters=block_args.output_filters, stride=1)
201201
for _ in range(block_args.num_repeat - 1):
202202
self._blocks.append(MBConvBlock(block_args, self._global_params, image_size=image_size))
@@ -261,15 +261,15 @@ def extract_endpoints(self, inputs):
261261
for idx, block in enumerate(self._blocks):
262262
drop_connect_rate = self._global_params.drop_connect_rate
263263
if drop_connect_rate:
264-
drop_connect_rate *= float(idx) / len(self._blocks) # scale drop connect_rate
264+
drop_connect_rate *= float(idx) / len(self._blocks) # scale drop connect_rate
265265
x = block(x, drop_connect_rate=drop_connect_rate)
266266
if prev_x.size(2) > x.size(2):
267-
endpoints['reduction_{}'.format(len(endpoints)+1)] = prev_x
267+
endpoints['reduction_{}'.format(len(endpoints)+1)] = prev_x
268268
prev_x = x
269269

270270
# Head
271271
x = self._swish(self._bn1(self._conv_head(x)))
272-
endpoints['reduction_{}'.format(len(endpoints)+1)] = x
272+
endpoints['reduction_{}'.format(len(endpoints)+1)] = x
273273

274274
return endpoints
275275

@@ -290,7 +290,7 @@ def extract_features(self, inputs):
290290
for idx, block in enumerate(self._blocks):
291291
drop_connect_rate = self._global_params.drop_connect_rate
292292
if drop_connect_rate:
293-
drop_connect_rate *= float(idx) / len(self._blocks) # scale drop connect_rate
293+
drop_connect_rate *= float(idx) / len(self._blocks) # scale drop connect_rate
294294
x = block(x, drop_connect_rate=drop_connect_rate)
295295

296296
# Head
@@ -373,7 +373,8 @@ def from_pretrained(cls, model_name, weights_path=None, advprop=False,
373373
A pretrained efficientnet model.
374374
"""
375375
model = cls.from_name(model_name, num_classes=num_classes, **override_params)
376-
load_pretrained_weights(model, model_name, weights_path=weights_path, load_fc=(num_classes == 1000), advprop=advprop)
376+
load_pretrained_weights(model, model_name, weights_path=weights_path,
377+
load_fc=(num_classes == 1000), advprop=advprop)
377378
model._change_in_channels(in_channels)
378379
return model
379380

‎efficientnet_pytorch/utils.py‎

Lines changed: 17 additions & 12 deletions
Original file line numberDiff line numberDiff line change
@@ -17,7 +17,7 @@
1717

1818

1919
################################################################################
20-
### Help functions for model architecture
20+
# Help functions for model architecture
2121
################################################################################
2222

2323
# GlobalParams and BlockArgs: Two namedtuples
@@ -50,11 +50,14 @@
5050
GlobalParams.__new__.__defaults__ = (None,) * len(GlobalParams._fields)
5151
BlockArgs.__new__.__defaults__ = (None,) * len(BlockArgs._fields)
5252

53-
54-
# An ordinary implementation of Swish function
55-
class Swish(nn.Module):
56-
def forward(self, x):
57-
return x * torch.sigmoid(x)
53+
# Swish activation function
54+
if hasattr(nn, 'SiLU'):
55+
Swish = nn.SiLU
56+
else:
57+
# For compatibility with old PyTorch versions
58+
class Swish(nn.Module):
59+
def forward(self, x):
60+
return x * torch.sigmoid(x)
5861

5962

6063
# A memory-efficient implementation of Swish function
@@ -97,10 +100,10 @@ def round_filters(filters, global_params):
97100
divisor = global_params.depth_divisor
98101
min_depth = global_params.min_depth
99102
filters *= multiplier
100-
min_depth = min_depth or divisor # pay attention to this line when using min_depth
103+
min_depth = min_depth or divisor # pay attention to this line when using min_depth
101104
# follow the formula transferred from official TensorFlow implementation
102105
new_filters = max(min_depth, int(filters + divisor / 2) // divisor * divisor)
103-
if new_filters < 0.9 * filters: # prevent rounding by more than 10%
106+
if new_filters < 0.9 * filters: # prevent rounding by more than 10%
104107
new_filters += divisor
105108
return int(new_filters)
106109

@@ -234,7 +237,7 @@ def forward(self, x):
234237
ih, iw = x.size()[-2:]
235238
kh, kw = self.weight.size()[-2:]
236239
sh, sw = self.stride
237-
oh, ow = math.ceil(ih / sh), math.ceil(iw / sw) # change the output size according to stride ! ! !
240+
oh, ow = math.ceil(ih / sh), math.ceil(iw / sw) # change the output size according to stride ! ! !
238241
pad_h = max((oh - 1) * self.stride[0] + (kh - 1) * self.dilation[0] + 1 - ih, 0)
239242
pad_w = max((ow - 1) * self.stride[1] + (kw - 1) * self.dilation[1] + 1 - iw, 0)
240243
if pad_h > 0 or pad_w > 0:
@@ -312,6 +315,7 @@ def forward(self, x):
312315
return F.max_pool2d(x, self.kernel_size, self.stride, self.padding,
313316
self.dilation, self.ceil_mode, self.return_indices)
314317

318+
315319
class MaxPool2dStaticSamePadding(nn.MaxPool2d):
316320
"""2D MaxPooling like TensorFlow's 'SAME' mode, with the given input image size.
317321
The padding mudule is calculated in construction function, then used in forward.
@@ -344,7 +348,7 @@ def forward(self, x):
344348

345349

346350
################################################################################
347-
### Helper functions for loading model params
351+
# Helper functions for loading model params
348352
################################################################################
349353

350354
# BlockDecoder: A Class for encoding and decoding BlockArgs
@@ -577,7 +581,7 @@ def get_model_params(model_name, override_params):
577581
# TODO: add the petrained weights url map of 'efficientnet-l2'
578582

579583

580-
def load_pretrained_weights(model, model_name, weights_path=None, load_fc=True, advprop=False):
584+
def load_pretrained_weights(model, model_name, weights_path=None, load_fc=True, advprop=False, verbose=True):
581585
"""Loads pretrained weights from weights path or download using url.
582586
583587
Args:
@@ -608,4 +612,5 @@ def load_pretrained_weights(model, model_name, weights_path=None, load_fc=True,
608612
['_fc.weight', '_fc.bias']), 'Missing keys when loading pretrained weights: {}'.format(ret.missing_keys)
609613
assert not ret.unexpected_keys, 'Missing keys when loading pretrained weights: {}'.format(ret.unexpected_keys)
610614

611-
print('Loaded pretrained weights for {}'.format(model_name))
615+
if verbose:
616+
print('Loaded pretrained weights for {}'.format(model_name))

0 commit comments

Comments
(0)

AltStyle によって変換されたページ (->オリジナル) /