Mohammad Haghighat comments

Results 7 comments of


                                            Mohammad Haghighat

The expanded size of the tensor must match the existing size at non-singleton dimension 0

I made a temporary fix skipping that batch. Add the following before setting the model input in [train.py line #97](https://github.com/aitorzip/PyTorch-CycleGAN/blob/67da8f9e2b69bd68763451803c7700aaccc92f18/train#L97): ``` # Skip the final batch when the total number...

SOLUTION: IndexError: invalid index of a 0-dim tensor. Use tensor.item() to convert a 0-dim tensor to a Python number

The above change triggered an error in the log method plotting the losses at the end of the epoch: `TypeError: Object of type 'Tensor' is not JSON serializable` The following...

Cannot load image, STB Reason: expected marker

I faced the same issue. Is the `COCO_train2014_000000167126.jpg` the only image that causes this error or are there others down the road? If this is the only one, we can...

Updated the dependency list

I didn't want to build is with GnuTLS or anything. I just followed the installation instruction here on a clean Ubuntu and it gave this error when running the `./configure`:...

Updated the dependency list

It looks like this now: ``` BUILD OPTIONS: SSL library: GnuTLS PKCS#11 support: GnuTLS DTLS support: yes ESP support: yes libproxy support: yes RSA SecurID support: no PSKC OATH file...

Updated the dependency list

That is right. As another note, I installed `libgnutls28-dev` as you've mentioned in the comment there.

Error D8016 '/RTC1' and '/O2' command-line options are incompatible

Right click on the project in the Solution Explorer and click on **Properties** and go to: **Configuration Properties** => **C/C++** => **Code Generation** Set the "Basic Runtime Checks" to "Default"....