yang-song/score_sde_pytorch

Unable to recreate working python environment to run this codebase using requirements.txt

rutujagurav opened this issue · 5 comments

rutujagurav commented

Hello Authors,

For some reason I am unable to create a working python environment using the provided requirements.txt. Could you please test the environment requirements you provide?

For me the error me the main issue seems to be with a dependency of tensorflow_gan==2.0.0 called tensorflow_probability which defaults to 0.19.0 and seems to be incompatible with tensorflow==2.4.0. But even after downgrading tensorflow_probability, I don't get a stable environment that's able to run main.py.

Any insights would be appreciated.

zq-OwO-qz commented

Hello Authors,

For some reason I am unable to create a working python environment using the provided requirements.txt. Could you please test the environment requirements you provide?

For me the error me the main issue seems to be with a dependency of tensorflow_gan==2.0.0 called tensorflow_probability which defaults to 0.19.0 and seems to be incompatible with tensorflow==2.4.0. But even after downgrading tensorflow_probability, I don't get a stable environment that's able to run main.py.

Any insights would be appreciated.

Hello rutujagurav, I meet the same issue as yours. I can't built the environment for the ipynb file and tried for some days. Could you tell me how to do if you resolve it? Thank you so much.

mengchuangji commented

tensorflow 2.4.0 requires numpy~=1.19.2, but you have numpy 1.24.4 which is incompatible.
Successfully installed jax-0.2.8 jaxlib-0.4.13 numpy-1.24.4

mengchuangji commented

tensorflow 2.4.0 requires numpy~=1.19.2, but you have numpy 1.24.4 which is incompatible. Successfully installed jax-0.2.8 jaxlib-0.4.13 numpy-1.24.4

The conflict is caused by:
tensorflow-datasets 3.1.0 depends on numpy
tensorflow 2.4.0 depends on numpy~=1.19.2
tensorboard 2.4.0 depends on numpy>=1.12.0
torchvision 0.13.1 depends on numpy
tensorflow-probability 0.12.2 depends on numpy>=1.13.3
jax 0.2.8 depends on numpy>=1.12
jaxlib 0.4.13 depends on numpy>=1.21
tensorflow-datasets 3.1.0 depends on numpy
tensorflow 2.4.0 depends on numpy~=1.19.2
tensorboard 2.4.0 depends on numpy>=1.12.0
torchvision 0.13.1 depends on numpy
tensorflow-probability 0.12.2 depends on numpy>=1.13.3
jax 0.2.8 depends on numpy>=1.12
jaxlib 0.4.12 depends on numpy>=1.21
tensorflow-datasets 3.1.0 depends on numpy
tensorflow 2.4.0 depends on numpy~=1.19.2
tensorboard 2.4.0 depends on numpy>=1.12.0
torchvision 0.13.1 depends on numpy
tensorflow-probability 0.12.2 depends on numpy>=1.13.3
jax 0.2.8 depends on numpy>=1.12
jaxlib 0.4.11 depends on numpy>=1.21
tensorflow-datasets 3.1.0 depends on numpy
tensorflow 2.4.0 depends on numpy~=1.19.2
tensorboard 2.4.0 depends on numpy>=1.12.0
torchvision 0.13.1 depends on numpy
tensorflow-probability 0.12.2 depends on numpy>=1.13.3
jax 0.2.8 depends on numpy>=1.12
jaxlib 0.4.10 depends on numpy>=1.21
tensorflow-datasets 3.1.0 depends on numpy
tensorflow 2.4.0 depends on numpy~=1.19.2
tensorboard 2.4.0 depends on numpy>=1.12.0
torchvision 0.13.1 depends on numpy
tensorflow-probability 0.12.2 depends on numpy>=1.13.3
jax 0.2.8 depends on numpy>=1.12
jaxlib 0.4.9 depends on numpy>=1.21
tensorflow-datasets 3.1.0 depends on numpy
tensorflow 2.4.0 depends on numpy~=1.19.2
tensorboard 2.4.0 depends on numpy>=1.12.0
torchvision 0.13.1 depends on numpy
tensorflow-probability 0.12.2 depends on numpy>=1.13.3
jax 0.2.8 depends on numpy>=1.12
jaxlib 0.4.7 depends on numpy>=1.21
tensorflow-datasets 3.1.0 depends on numpy
tensorflow 2.4.0 depends on numpy~=1.19.2
tensorboard 2.4.0 depends on numpy>=1.12.0
torchvision 0.13.1 depends on numpy
tensorflow-probability 0.12.2 depends on numpy>=1.13.3
jax 0.2.8 depends on numpy>=1.12
jaxlib 0.4.6 depends on numpy>=1.20
tensorflow-datasets 3.1.0 depends on numpy
tensorflow 2.4.0 depends on numpy~=1.19.2
tensorboard 2.4.0 depends on numpy>=1.12.0
torchvision 0.13.1 depends on numpy
tensorflow-probability 0.12.2 depends on numpy>=1.13.3
jax 0.2.8 depends on numpy>=1.12
jaxlib 0.4.4 depends on numpy>=1.20
tensorflow-datasets 3.1.0 depends on numpy
tensorflow 2.4.0 depends on numpy~=1.19.2
tensorboard 2.4.0 depends on numpy>=1.12.0
torchvision 0.13.1 depends on numpy
tensorflow-probability 0.12.2 depends on numpy>=1.13.3
jax 0.2.8 depends on numpy>=1.12
jaxlib 0.4.3 depends on numpy>=1.20
tensorflow-datasets 3.1.0 depends on numpy
tensorflow 2.4.0 depends on numpy~=1.19.2
tensorboard 2.4.0 depends on numpy>=1.12.0
torchvision 0.13.1 depends on numpy
tensorflow-probability 0.12.2 depends on numpy>=1.13.3
jax 0.2.8 depends on numpy>=1.12
jaxlib 0.4.2 depends on numpy>=1.20
tensorflow-datasets 3.1.0 depends on numpy
tensorflow 2.4.0 depends on numpy~=1.19.2
tensorboard 2.4.0 depends on numpy>=1.12.0
torchvision 0.13.1 depends on numpy
tensorflow-probability 0.12.2 depends on numpy>=1.13.3
jax 0.2.8 depends on numpy>=1.12
jaxlib 0.4.1 depends on numpy>=1.20

dinuowei commented

i use this requierments
ml-collections==0.1.0
tensorflow-gan==2.0.0
tensorflow_io
tensorflow_datasets==3.1.0
tensorflow==2.4.0
tensorflow-addons==0.12.0
tensorboard==2.4.0
tensorflow-estimator==2.4.0
tensorflow-probability==0.12.2
absl-py==0.10.0
torch==1.8.0
torchvision
ninja
numpy==1.19.5
and cuda 11.1 but still show only support vs17-19 , for me my vs is 2022, so it is hard to test. if you can pls test it

dinuowei commented

tensorflow 2.4.0 requires numpy~=1.19.2, but you have numpy 1.24.4 which is incompatible. Successfully installed jax-0.2.8 jaxlib-0.4.13 numpy-1.24.4

The conflict is caused by: tensorflow-datasets 3.1.0 depends on numpy tensorflow 2.4.0 depends on numpy~=1.19.2 tensorboard 2.4.0 depends on numpy>=1.12.0 torchvision 0.13.1 depends on numpy tensorflow-probability 0.12.2 depends on numpy>=1.13.3 jax 0.2.8 depends on numpy>=1.12 jaxlib 0.4.13 depends on numpy>=1.21 tensorflow-datasets 3.1.0 depends on numpy tensorflow 2.4.0 depends on numpy~=1.19.2 tensorboard 2.4.0 depends on numpy>=1.12.0 torchvision 0.13.1 depends on numpy tensorflow-probability 0.12.2 depends on numpy>=1.13.3 jax 0.2.8 depends on numpy>=1.12 jaxlib 0.4.12 depends on numpy>=1.21 tensorflow-datasets 3.1.0 depends on numpy tensorflow 2.4.0 depends on numpy~=1.19.2 tensorboard 2.4.0 depends on numpy>=1.12.0 torchvision 0.13.1 depends on numpy tensorflow-probability 0.12.2 depends on numpy>=1.13.3 jax 0.2.8 depends on numpy>=1.12 jaxlib 0.4.11 depends on numpy>=1.21 tensorflow-datasets 3.1.0 depends on numpy tensorflow 2.4.0 depends on numpy~=1.19.2 tensorboard 2.4.0 depends on numpy>=1.12.0 torchvision 0.13.1 depends on numpy tensorflow-probability 0.12.2 depends on numpy>=1.13.3 jax 0.2.8 depends on numpy>=1.12 jaxlib 0.4.10 depends on numpy>=1.21 tensorflow-datasets 3.1.0 depends on numpy tensorflow 2.4.0 depends on numpy~=1.19.2 tensorboard 2.4.0 depends on numpy>=1.12.0 torchvision 0.13.1 depends on numpy tensorflow-probability 0.12.2 depends on numpy>=1.13.3 jax 0.2.8 depends on numpy>=1.12 jaxlib 0.4.9 depends on numpy>=1.21 tensorflow-datasets 3.1.0 depends on numpy tensorflow 2.4.0 depends on numpy~=1.19.2 tensorboard 2.4.0 depends on numpy>=1.12.0 torchvision 0.13.1 depends on numpy tensorflow-probability 0.12.2 depends on numpy>=1.13.3 jax 0.2.8 depends on numpy>=1.12 jaxlib 0.4.7 depends on numpy>=1.21 tensorflow-datasets 3.1.0 depends on numpy tensorflow 2.4.0 depends on numpy~=1.19.2 tensorboard 2.4.0 depends on numpy>=1.12.0 torchvision 0.13.1 depends on numpy tensorflow-probability 0.12.2 depends on numpy>=1.13.3 jax 0.2.8 depends on numpy>=1.12 jaxlib 0.4.6 depends on numpy>=1.20 tensorflow-datasets 3.1.0 depends on numpy tensorflow 2.4.0 depends on numpy~=1.19.2 tensorboard 2.4.0 depends on numpy>=1.12.0 torchvision 0.13.1 depends on numpy tensorflow-probability 0.12.2 depends on numpy>=1.13.3 jax 0.2.8 depends on numpy>=1.12 jaxlib 0.4.4 depends on numpy>=1.20 tensorflow-datasets 3.1.0 depends on numpy tensorflow 2.4.0 depends on numpy~=1.19.2 tensorboard 2.4.0 depends on numpy>=1.12.0 torchvision 0.13.1 depends on numpy tensorflow-probability 0.12.2 depends on numpy>=1.13.3 jax 0.2.8 depends on numpy>=1.12 jaxlib 0.4.3 depends on numpy>=1.20 tensorflow-datasets 3.1.0 depends on numpy tensorflow 2.4.0 depends on numpy~=1.19.2 tensorboard 2.4.0 depends on numpy>=1.12.0 torchvision 0.13.1 depends on numpy tensorflow-probability 0.12.2 depends on numpy>=1.13.3 jax 0.2.8 depends on numpy>=1.12 jaxlib 0.4.2 depends on numpy>=1.20 tensorflow-datasets 3.1.0 depends on numpy tensorflow 2.4.0 depends on numpy~=1.19.2 tensorboard 2.4.0 depends on numpy>=1.12.0 torchvision 0.13.1 depends on numpy tensorflow-probability 0.12.2 depends on numpy>=1.13.3 jax 0.2.8 depends on numpy>=1.12 jaxlib 0.4.1 depends on numpy>=1.20

可以看一下我刚发的