While training on own images, getting a Invalid operation is performed in: LinearFunction (Forward) error

[unography]

While training on my own images, I'm getting this error,

chainer.utils.type_check.InvalidType:
Invalid operation is performed in: LinearFunction (Forward)

Expect: prod(in_types[0].shape[1:]) == in_types[1].shape[1]
Actual: 259656 != 27648

[tjtorres]

Can you give me a bit more info as to how this error came about? Are you using the command line tools or one of the model classes? Have you properly set the image sizes?

[cyrilreboul]

Hi there, I am jumping in as I have the same problem regardless of the model used. I am using the code from the notebook. My own images (64x64=4096) are loaded without trouble (dimensions are set). Here is the error (I am on OSX10.11):

Traceback (most recent call last):
File "test.py", line 12, in
vg.fit(x_all, save_freq=2, pic_freq=2, n_epochs=10, model_path = m_path, img_path=im_path)
File "/opt/local/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/fauxtograph/fauxtograph.py", line 754, in fit
disc_samp, disc_batch = self._forward(x_batch)
File "/opt/local/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/fauxtograph/fauxtograph.py", line 592, in forward
disc_batch = self.disc(batch, dropout_ratio=self.dropout_ratio)[0]
File "/opt/local/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/fauxtograph/vaegan.py", line 439, in call
batch = F.relu(getattr(self, 'linear%i' % i)(batch))
File "/opt/local/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/chainer/links/connection/linear.py", line 65, in call
return linear.linear(x, self.W, self.b)
File "/opt/local/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/chainer/functions/connection/linear.py", line 81, in linear
return LinearFunction()(x, W, b)
File "/opt/local/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/chainer/function.py", line 115, in call
self._check_data_type_forward(in_data)
File "/opt/local/Library/Frameworks/Python.framework/Versions/2.7/lib/python2.7/site-packages/chainer/function.py", line 190, in _check_data_type_forward
raise type_check.InvalidType(e.expect, e.actual, msg=msg)
chainer.utils.type_check.InvalidType:
Invalid operation is performed in: LinearFunction (Forward)

Expect: prod(in_types[0].shape[1:]) == in_types[1].shape[1]
Actual: 4096 != 12288

I would greatly appreciate if you could help!
Cheers.

[RobGeada]

I had the same issue, resolved by using the --shape flag to specify image size for the train command, ie:
fauxtograph train --kl_ratio 0.005 --shape 64 64 ./images ./models/modelname

[ArghyaPal]

Hi, @tjtorres and @RobGeada
have got the similar problem.. My Image size is 300X300 and Gray Scale

`fauxtograph train --kl_ratio 0.005 --shape 300 300 /home/dl-box/Arghya/arghya/lasttry/images/ ./models/VAE
100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 49/49 [00:00<00:00, 1781.81it/s]
Image Files Loaded!
epoch: 1
0%| | 0/1 [00:00<?, ?it/s]
Traceback (most recent call last):
File "/usr/local/bin/fauxtograph", line 11, in
sys.exit(fauxtograph())
File "/usr/local/lib/python2.7/dist-packages/click/core.py", line 722, in call
return self.main(*args, **kwargs)
File "/usr/local/lib/python2.7/dist-packages/click/core.py", line 697, in main
rv = self.invoke(ctx)
File "/usr/local/lib/python2.7/dist-packages/click/core.py", line 1066, in invoke
return process_result(sub_ctx.command.invoke(sub_ctx))
File "/usr/local/lib/python2.7/dist-packages/click/core.py", line 895, in invoke
return ctx.invoke(self.callback, **ctx.params)
File "/usr/local/lib/python2.7/dist-packages/click/core.py", line 535, in invoke
return callback(*args, **kwargs)
File "/usr/local/lib/python2.7/dist-packages/fauxtograph/fauxto.py", line 123, in train
vae.fit(x_all, batch_size=batch, n_epochs=epoch)
File "/usr/local/lib/python2.7/dist-packages/fauxtograph/fauxtograph.py", line 317, in fit
out, kl_loss, rec_loss = self.model.forward(x_batch)
File "/usr/local/lib/python2.7/dist-packages/fauxtograph/vaegan.py", line 526, in forward
out, means, ln_vars = self.encode(batch, test=test)
File "/usr/local/lib/python2.7/dist-packages/fauxtograph/vaegan.py", line 509, in encode
x = self.enc(data, test=test)
File "/usr/local/lib/python2.7/dist-packages/fauxtograph/vaegan.py", line 152, in call
batch = F.relu(getattr(self, 'linear%i' % i)(batch))
File "/usr/local/lib/python2.7/dist-packages/chainer/links/connection/linear.py", line 65, in call
return linear.linear(x, self.W, self.b)
File "/usr/local/lib/python2.7/dist-packages/chainer/functions/connection/linear.py", line 81, in linear
return LinearFunction()(x, W, b)
File "/usr/local/lib/python2.7/dist-packages/chainer/function.py", line 102, in call
self._check_data_type_forward(in_data)
File "/usr/local/lib/python2.7/dist-packages/chainer/function.py", line 144, in _check_data_type_forward
raise type_check.InvalidType(e.expect, e.actual, msg=msg)
chainer.utils.type_check.InvalidType:
Invalid operation is performed in: LinearFunction (Forward)

Expect: prod(in_types[0].shape[1:]) == in_types[1].shape[1]
Actual: 90000 != 270000

`

[miracodezu]

I'm having the same issue. I was able to pass the colored images with --shape, but after I converted to gray scale, I get this error again. Even after converting them to 125x150 just to be sure.