geek-ai/irgan

Issue in QA task, baseline.py

Closed this issue · 3 comments

weicatme commented

Hi,
After I and my superviser checked cafefully with your baseline.py, There are lots of issues happening. The description are the following:

  1. in line 184, actually you did not provide the model mentioned.
  2. in line 186, you did not claim Discriminator.Discriminator_pair_wise
  3. so if I change the Discriminator.Discriminator_pair_wise into Discriminator only, error will appear in line 197 stating that "pair" is not recognized.

Could you help solve the above problems or provide a precise description?

I am looking forward to your response and thank you for your time,
Ryan

wabyking commented

I may upload the wrong version of baseline.py

  1. in line 184, the pre-trained model is in the folder named "model", you should rename the DIS_MODEL_FILE as "model/pre-trained.model".
  2. in line 186, "Discriminator.Discriminator_pair_wise" should be "Discriminator.Discriminator "
  3. Sorry, I cannot repeat your error, would you please provide more detail?

I have fixed these problem in this repository, I am sorry for my error. If there are any other problems, please do not hesitate to contact me.
Thanks for your issue.

weicatme commented

Hello wabyking,
Thank you for your quick response.
After I replaced your updated baseline.py, the execution moved forward a little bit, and I got the following error showing in the picture attached. As a novice of Tensorflow, I am not sure where this type of error comes from. One guess is that the python version does matter in your case because I was using the 2.7 version.
Best,
Ryan

error1

I have fixed some of the bugs. There are the updates:

  1. in line 101, you called the concat functions in a wrong way, which should be tf.concat(output,3)
  2. Due to recent updated tensorflow 1.0 version, all the tf.mul should be tf.multiply and all the tf.sub should be tf.subtract.

Best,
Ryan

wabyking commented

Thank Ryan,
I test the code in TensorFlow 0.12, it works. Sorry for that it does not support the newest version.
It may cost a little time to transfer it to the newest Tensorflow. I may update it if not busy. But I cannot confirm the time schedule.
Benyou