Closed joehoover closed 1 year ago
Thanks for your notice. We have updated now. By the way, we would clean up the code in the next version which no longer rely on apex dependency. Stay Tuned.
Wow, thanks for the fast response!
I might just wait for the next version. Do you have an ETA for that?
Unfortunately, I ran into another issue with dependency conflicts.
torchvision==0.13.1
is not compatible with torch==1.13.1
.
What version of torchvision
would you recommend?
You might use the following script here to install the torch and torchvision for your reference.
conda install pytorch==1.13.1 torchvision==0.14.1 torchaudio==0.13.1 pytorch-cuda=11.7 -c pytorch -c nvidia
For the next version, we aim to natively support the huggingface library! Besides, we are working on multi-lingual support of that. We would release it as soon as the training is done.
If you have any further question or need help, we can set up a slack / discord channel to help you. 😊
Wow, thanks for the fast response!
I might just wait for the next version. Do you have an ETA for that?
Unfortunately, I ran into another issue with dependency conflicts.
torchvision==0.13.1
is not compatible withtorch==1.13.1
.What version of
torchvision
would you recommend?
Hi, just quick notice for the next version have been released with much easier usage!
There are conflicts between the dependencies specified in the README and
env.yaml
.In the README,
torchvision
is not specified andtorch
is pinned at1.13.1
. Further, inenv.yaml
peft
is not listed as a dependency.However, when I tried to run the demo like:
I got an import error for
torchvision
.I then referred to
env.yaml
for dependencies, but there torch is pinned at1.12.1
This conflicts with the version specified in the README.