search

junyanz / interactive-deep-colorization

Deep learning software for colorizing black and white images with a few clicks.
https://richzhang.github.io/ideepcolor/
MIT License
2.69k stars 447 forks source link

some wrong when run to globalmodel #43

Closed cccusername closed 6 years ago

cccusername commented 6 years ago

I run DemoGlobalHistogramTransfer.ipynb by jupyter.My jupyter keral dead, so i copy all of them to a PY file and error appear as fellow:

[libprotobuf ERROR google/protobuf/text_format.cc:245] Error parsing text-format caffe.NetParameter: 5:15: Message type "caffe.LayerParameter" has no field named "input_param".

i think my caffe version and the .prototxt of the demo do not match.would you tell me the version of your caffe and prototxt?thank you!

cccusername commented 6 years ago

I rebuild the caffe，but new error appear as fellow:

**_ColorizeImageCaffe instantiated
gpu_id = 0, net_path = ./models/global_model/deploy_nodist.prototxt, model_path = ./models/global_model/global_model.caffemodel

---------------------------------------------------------------------------
ImportError                               Traceback (most recent call last)
ImportError: No module named caffe_traininglayers

---------------------------------------------------------------------------
SystemError                               Traceback (most recent call last)
<ipython-input-29-29cd1cf45678> in <module>()
      4 # Colorization network
      5 cid = CI.ColorizeImageCaffeGlobDist(Xd)
----> 6 cid.prep_net(gpu_id,prototxt_path='./models/global_model/deploy_nodist.prototxt',            caffemodel_path='./models/global_model/global_model.caffemodel')
      7 
      8 # Global distribution network - extracts global color statistics from an image

/home/choi/HandsomeCHAN/interactive-deep-colorization-master/data/colorize_image.py in prep_net(self, gpu_id, prototxt_path, caffemodel_path)
    233         if gpu_id == -1:
    234             caffe.set_mode_cpu()
--> 235         else:
    236             caffe.set_device(gpu_id)
    237             caffe.set_mode_gpu()

SystemError: NULL result without error in PyObject_Call_**

I will very appreciate that if you can tell me how to fix this error，thank you！

junyanz commented 6 years ago

Are you able to run standard caffe test (make runtest)? @richzhang

richzhang commented 6 years ago

Did you add the ./caffe_files to your PYTHONPATH?

cccusername commented 6 years ago

I did, but it doesn`t work. Now i have solve my problem by install the lastest caffe. Thank you!

kaustubholpadkar commented 5 years ago

I rebuild the caffe，but new error appear as fellow:

**_ColorizeImageCaffe instantiated
gpu_id = 0, net_path = ./models/global_model/deploy_nodist.prototxt, model_path = ./models/global_model/global_model.caffemodel

---------------------------------------------------------------------------
ImportError                               Traceback (most recent call last)
ImportError: No module named caffe_traininglayers

---------------------------------------------------------------------------
SystemError                               Traceback (most recent call last)
<ipython-input-29-29cd1cf45678> in <module>()
      4 # Colorization network
      5 cid = CI.ColorizeImageCaffeGlobDist(Xd)
----> 6 cid.prep_net(gpu_id,prototxt_path='./models/global_model/deploy_nodist.prototxt',            caffemodel_path='./models/global_model/global_model.caffemodel')
      7 
      8 # Global distribution network - extracts global color statistics from an image

/home/choi/HandsomeCHAN/interactive-deep-colorization-master/data/colorize_image.py in prep_net(self, gpu_id, prototxt_path, caffemodel_path)
    233         if gpu_id == -1:
    234             caffe.set_mode_cpu()
--> 235         else:
    236             caffe.set_device(gpu_id)
    237             caffe.set_mode_gpu()

SystemError: NULL result without error in PyObject_Call_**

I will very appreciate that if you can tell me how to fix this error，thank you！

I am facing the same issue even with latest caffe version. Can you suggest any solution?

cccusername commented 5 years ago

@kaustubholpadkar My caffe version is 1.0.0, maybe you could try it.If it still does`t work, you could try this work of the pytorch version. https://github.com/richzhang/colorization-pytorch

kaustubholpadkar commented 5 years ago

@cccusername I am using the same caffe version. I want to do Global Histogram Style Transfer and I could not find PyTorch repo has that support right now. I made my code work by commenting out lines 407, 408, 409 from data/colorize_image.py

Benitoite commented 5 years ago

407, 408, 409 from data/colorize_image.py

@kaustubholpadkar were you referring to a different commit than the latest master?

kaustubholpadkar commented 5 years ago

407, 408, 409 from data/colorize_image.py

@kaustubholpadkar were you referring to a different commit than the latest master?

Nope. I am using the latest master only.