samet-akcay / ganomaly

GANomaly: Semi-Supervised Anomaly Detection via Adversarial Training

MIT License

853 stars 213 forks source link

Cannot install dependencies: mkl-fft, mkl-random, mkl-service #58

Open heytan opened 4 years ago

heytan commented 4 years ago

I cannot install dependencies: mkl-fft==1.0.12, mkl-random==1.0.2, and mkl-service==2.0.2 using ''pip install --user --requirement requirements.txt'' and the errors are: ERROR: Could not find a version that satisfies the requirement mkl-fft==1.0.12 (from versions: none) ERROR: No matching distribution found for mkl-fft==1.0.12 i don't know how to do, can anyone help? thank u.

Constant-TL commented 4 years ago

I have also met this problem, and I solved it by usingconda install -c intel mkl_fft=1.0.12. You can install the other two packages by the similar commands @heytan

Minqi824 commented 4 years ago

I have also met this problem, and I solved it by usingconda install -c intel mkl_fft=1.0.12. You can install the other two packages by the similar commands @heytan

This actually works! Thanks a lot!

Minqi824 commented 4 years ago

I have also met this problem, and I solved it by usingconda install -c intel mkl_fft=1.0.12. You can install the other two packages by the similar commands @heytan

Well, when I download the mkl_fft package, it seems too slow and I change the mirror, but it still doesn't work. Do you meet this problem?

Minqi824 commented 4 years ago

I cannot install dependencies: mkl-fft==1.0.12, mkl-random==1.0.2, and mkl-service==2.0.2 using ''pip install --user --requirement requirements.txt'' and the errors are: ERROR: Could not find a version that satisfies the requirement mkl-fft==1.0.12 (from versions: none) ERROR: No matching distribution found for mkl-fft==1.0.12 i don't know how to do, can anyone help? thank u.

another notification: I use conda install to get all the above packages in pycharm. However, some packages (e.g.,mkl-service==2.3.0) seems too new to satisfy this requirement. Maybe modify the version code in the requirement file can fix this problem.

huangpan2507 commented 4 years ago

I have also met this problem, and I solved it by usingconda install -c intel mkl_fft=1.0.12. You can install the other two packages by the similar commands @heytan

Well, when I download the mkl_fft package, it seems too slow and I change the mirror, but it still doesn't work. Do you meet this problem?

conda install -c conda-forge mkl_fft.
Remind! It woks well! Download those tools very quickly.