search

weitingchen83 / JSTASR-DesnowNet-ECCV-2020

This is the project page of our paper which has been published in ECCV 2020.
https://github.com/weitingchen83/JSTASR-DesnowNet-ECCV-2020
MIT License
26 stars 10 forks source link

ValueError: axes don't match array #1

Open zhilaAI opened 4 years ago

zhilaAI commented 4 years ago

File "./predict.py", line 95, in modelRecoverCombine = model.model.build_combine_model() File "/home/zhila/Project/Research_Journal/JSTASR-DesnowNet-ECCV-2020/model/model.py", line 192, in build_combine_model modelSnow.load_weights('./modelParam/snowmodel.h5',by_name=False) File "/home/zhila/.local/lib/python3.6/site-packages/keras/engine/saving.py", line 492, in load_wrapper return load_function(*args, *kwargs) File "/home/zhila/.local/lib/python3.6/site-packages/keras/engine/network.py", line 1230, in load_weights f, self.layers, reshape=reshape) File "/home/zhila/.local/lib/python3.6/site-packages/keras/engine/saving.py", line 1224, in load_weights_from_hdf5_group reshape=reshape) File "/home/zhila/.local/lib/python3.6/site-packages/keras/engine/saving.py", line 861, in preprocess_weights_for_loading weights = convert_nested_model(weights) File "/home/zhila/.local/lib/python3.6/site-packages/keras/engine/saving.py", line 836, in convert_nested_model original_backend=original_backend)) File "/home/zhila/.local/lib/python3.6/site-packages/keras/engine/saving.py", line 980, in preprocess_weights_for_loading weights[0] = np.transpose(weights[0], (3, 2, 0, 1)) File "<__array_function__ internals>", line 6, in transpose File "/home/zhila/.local/lib/python3.6/site-packages/numpy/core/fromnumeric.py", line 653, in transpose return _wrapfunc(a, 'transpose', axes) File "/home/zhila/.local/lib/python3.6/site-packages/numpy/core/fromnumeric.py", line 58, in _wrapfunc return bound(args, **kwds) ValueError: axes don't match array

zhilaAI commented 4 years ago

Can you please help me?

cctakaet commented 4 years ago

Are you running the following command? python ./predict.py -dataroot ./testImg -predictpath ./p -batch_size 3

If so, do you modify predict.py ? I can't reproduce your issue so I would need more detail.

zhilaAI commented 4 years ago

Hi, Yes I used the same command line to run the program.

I did not change anything in predict.py. I do not know why I have this error.

I followed your instructions step by step to run. The only issue here I did not use TensorFlow 1.6, I am using 1.14. But I do not think this is the problem.

On Mon, Sep 28, 2020 at 3:29 AM cctakaet notifications@github.com wrote:

Are you running the following command? python ./predict.py -dataroot ./testImg -predictpath ./p -batch_size 3

If so, do you modify predict.py ? I can't reproduce your issue so I would need more detail.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/weitingchen83/JSTASR-DesnowNet-ECCV-2020/issues/1#issuecomment-699923014, or unsubscribe https://github.com/notifications/unsubscribe-auth/AKCQ2IK3QBJWVFY2W7XLF5TSIBQRPANCNFSM4RP4FUBA .

-- Yours sincerely, Zhila

UBC University, Okanagan Campus Faculty of Applied Science EME4242 - 1137 Alumni Ave Kelowna, BC Canada V1V 1V7 Ph.D. Student Tel. |+15149926299 Alternative Emails: http://goog_284045116zhilaai@mail.ubc.ca, zhila.bahrami@ http://@canscantech.comcanscantech.com

cctakaet commented 4 years ago

It seems to be an issue of keras/tf. https://github.com/keras-team/keras/issues/13279

Could you try TensorFlow 1.12.0 ? We actually use this verion for training and testing. If you want to use docker, we already have docker image on docker hub. You can get it by docker pull cctakaet/jstasr:latest .

zhilaAI commented 4 years ago

I think you are right the problem is with libraries and packages. I tried TensorFlow 1.12 to run the code. I could not install TensorFlow 1.12.0 as my Cuda is not matched with that version. My Cuda is 10.1. I will focus on using the docker container to see if I can run the model successfully or not. I am so grateful for your help.

On Mon, Sep 28, 2020 at 11:19 AM cctakaet notifications@github.com wrote:

It seems to be an issue of keras/tf. keras-team/keras#13279 https://github.com/keras-team/keras/issues/13279

Could you try TensorFlow 1.12.0 ? We actually use this verion for training and testing. If you want to use docker, we already have docker image on docker hub. You can get it by docker pull cctakaet/jstasr:latest .

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/weitingchen83/JSTASR-DesnowNet-ECCV-2020/issues/1#issuecomment-700201629, or unsubscribe https://github.com/notifications/unsubscribe-auth/AKCQ2IJZFOHFCPBSPA2PJCTSIDHU5ANCNFSM4RP4FUBA .

-- Yours sincerely, Zhila

UBC University, Okanagan Campus Faculty of Applied Science EME4242 - 1137 Alumni Ave Kelowna, BC Canada V1V 1V7 Ph.D. Student Tel. |+15149926299 Alternative Emails: http://goog_284045116zhilaai@mail.ubc.ca, zhila.bahrami@ http://@canscantech.comcanscantech.com