search

NVIDIA / FastPhotoStyle

Style transfer, deep learning, feature transform
Other
11.11k stars 1.2k forks source link

Docker build frozen #75

Open AlexanderKozhevin opened 5 years ago

AlexanderKozhevin commented 5 years ago

I'm trying to build docker image. But everytime it's stocked on step 12. 

Step 11/16 : RUN conda install -y pytorch=0.4.1 torchvision cuda91 -c pytorch
 ---> Running in 79a08f6afcdc
Fetching package metadata .............
Solving package specifications: .

Package plan for installation in environment /opt/anaconda3:

The following NEW packages will be INSTALLED:

    blas:        1.0-mkl                                      
    cuda91:      1.0-h4c16780_0                        pytorch
    ninja:       1.8.2-py36h6bb024c_1                         
    pytorch:     0.4.1-py36_py35_py27__9.0.176_7.1.2_2 pytorch
    torchvision: 0.2.1-py_2                            pytorch

The following packages will be UPDATED:

    anaconda:    5.0.1-py36hd30a520_1                          --> custom-py36hbbc8b67_0
    conda:       4.3.30-py36h5d9f9f4_0                         --> 4.6.8-py36_0         
    conda-env:   2.6.0-h36134e3_1                              --> 2.6.0-1              
    pycosat:     0.6.2-py36h1a0ea17_1                          --> 0.6.3-py36h14c3975_0 

blas-1.0-mkl.t 100% |###############################| Time: 0:00:00   4.96 MB/s
conda-env-2.6. 100% |###############################| Time: 0:00:00   3.94 MB/s
cuda91-1.0-h4c 100% |###############################| Time: 0:00:00   3.49 MB/s
anaconda-custo 100% |###############################| Time: 0:00:00   5.84 MB/s
ninja-1.8.2-py 100% |###############################| Time: 0:00:00  37.70 MB/s
pycosat-0.6.3- 100% |###############################| Time: 0:00:00  40.20 MB/s
pytorch-0.4.1- 100% |###############################| Time: 0:00:22  21.77 MB/s
torchvision-0. 100% |###############################| Time: 0:00:00  17.99 MB/s
conda-4.6.8-py 100% |###############################| Time: 0:00:00  38.25 MB/s
Removing intermediate container 79a08f6afcdc
 ---> f73722974e7f
Step 12/16 : RUN conda install -y -c anaconda pip
 ---> Running in 2a75a902441d
Collecting package metadata: ...working... done
Solving environment: ...working...
TheIllusion commented 5 years ago

I am also facing with this issue. Does anyone know how to fix this?

mchalecki commented 5 years ago

There are some problems with conda. For this problem you need to change channel. Also the next step of installing opencv is also not working. You need to install it following way

RUN conda install -y  --override-channels -c main pip
RUN pip install opencv-python