MCPcopy Index your code
hub / github.com/pytorch/pytorch / VGGA

Function VGGA

caffe2/experiments/python/convnet_benchmarks.py:268–380  ·  view source on GitHub ↗
(order)

Source from the content-addressed store, hash-verified

266
267
268def VGGA(order):
269 model = cnn.CNNModelHelper(order, name='vgg-a',
270 use_cudnn=True, cudnn_exhaustive_search=True)
271 conv1 = model.Conv(
272 "data",
273 "conv1",
274 3,
275 64,
276 3,
277 ('XavierFill', {}),
278 ('ConstantFill', {}),
279 pad=1
280 )
281 relu1 = model.Relu(conv1, "conv1")
282 pool1 = model.MaxPool(relu1, "pool1", kernel=2, stride=2)
283 conv2 = model.Conv(
284 pool1,
285 "conv2",
286 64,
287 128,
288 3,
289 ('XavierFill', {}),
290 ('ConstantFill', {}),
291 pad=1
292 )
293 relu2 = model.Relu(conv2, "conv2")
294 pool2 = model.MaxPool(relu2, "pool2", kernel=2, stride=2)
295 conv3 = model.Conv(
296 pool2,
297 "conv3",
298 128,
299 256,
300 3,
301 ('XavierFill', {}),
302 ('ConstantFill', {}),
303 pad=1
304 )
305 relu3 = model.Relu(conv3, "conv3")
306 conv4 = model.Conv(
307 relu3,
308 "conv4",
309 256,
310 256,
311 3,
312 ('XavierFill', {}),
313 ('ConstantFill', {}),
314 pad=1
315 )
316 relu4 = model.Relu(conv4, "conv4")
317 pool4 = model.MaxPool(relu4, "pool4", kernel=2, stride=2)
318 conv5 = model.Conv(
319 pool4,
320 "conv5",
321 256,
322 512,
323 3,
324 ('XavierFill', {}),
325 ('ConstantFill', {}),

Callers

nothing calls this directly

Calls 5

ConvMethod · 0.95
ReluMethod · 0.95
MaxPoolMethod · 0.95
FCMethod · 0.95
SoftmaxMethod · 0.95

Tested by

no test coverage detected

Used in the wild real call sites across dependent graphs

searching dependent graphs…