Closed JinwnK closed 1 month ago
Hi, JinwnK
Did you create an enviroment and install? It seems like that you have install torch==2.1.2 (requires python3.9), which has conflict with our default python version (python3.8).
Hi, JinwnK
Did you create an enviroment and install? It seems like that you have install torch==2.1.2 (requires python3.9), which has conflict with our default python version (python3.8).
Yes, I did following the guide:
git clone https://github.com/CatIIIIIIII/RNAErnie_baselines.git
cd ./RNAErnie_baselines
conda env create -f environment.yml ( The file name is actually environment.yaml )
I checked the file "environment.yaml" and found "dependencies:
Oh, you are installing RNAErnie_baselines repo. I'll check it later and give you a response ASAP.
But I believe that the version of torch is not that necessary, you could install any versions that have no conflict with other packages and have a try.
Oh, you are installing RNAErnie_baselines repo. I'll check it later and give you a response ASAP.
But I believe that the version of torch is not that necessary, you could install any versions that have no conflict with other packages and have a try.
I'll try it. Thanks!
Hi, JinwnK
After check, torch 2.0.1 should work. You could install it by conda:
conda install pytorch==2.0.1 torchvision==0.15.2 torchaudio==2.0.2 pytorch-cuda=11.8 -c pytorch -c nvidia
or pip:
pip install torch==2.0.1 torchvision==0.15.2 torchaudio==2.0.2 --index-url https://download.pytorch.org/whl/cu118
Or you could check out my updated environment.yaml
Hi, JinwnK
After check, torch 2.0.1 should work. You could install it by conda:
conda install pytorch==2.0.1 torchvision==0.15.2 torchaudio==2.0.2 pytorch-cuda=11.8 -c pytorch -c nvidia
or pip:pip install torch==2.0.1 torchvision==0.15.2 torchaudio==2.0.2 --index-url https://download.pytorch.org/whl/cu118
Or you could check out my updated environment.yaml
I tried to install with the updated environment.yaml but still get the same error :
Pip subprocess error:
ERROR: Ignored the following versions that require a different python version: 0.42.0 Requires-Python >=3.9; 0.42.0rc1 Requires-Python >=3.9; 0.43.0rc1 Requires-Python >=3.9; 0.59.0 Requires-Python >=3.9; 0.59.0rc1 Requires-Python >=3.9; 0.59.1 Requires-Python >=3.9; 0.60.0rc1 Requires-Python >=3.9; 1.11.0 Requires-Python <3.13,>=3.9; 1.11.0rc1 Requires-Python <3.13,>=3.9; 1.11.0rc2 Requires-Python <3.13,>=3.9; 1.11.1 Requires-Python <3.13,>=3.9; 1.11.2 Requires-Python <3.13,>=3.9; 1.11.3 Requires-Python <3.13,>=3.9; 1.11.4 Requires-Python >=3.9; 1.12.0 Requires-Python >=3.9; 1.12.0rc1 Requires-Python >=3.9; 1.12.0rc2 Requires-Python >=3.9; 1.13.0 Requires-Python >=3.9; 1.13.0rc1 Requires-Python >=3.9; 1.4.0 Requires-Python >=3.9; 1.4.0rc1 Requires-Python >=3.9; 1.4.1.post1 Requires-Python >=3.9; 1.4.2 Requires-Python >=3.9; 1.5.0rc1 Requires-Python >=3.9; 2.1.0 Requires-Python >=3.9; 2.1.0rc0 Requires-Python >=3.9; 2.1.1 Requires-Python >=3.9; 2.1.2 Requires-Python >=3.9; 2.1.3 Requires-Python >=3.9; 2.1.4 Requires-Python >=3.9; 2.2.0 Requires-Python >=3.9; 2.2.0rc0 Requires-Python >=3.9; 2.2.1 Requires-Python >=3.9; 2.2.2 Requires-Python >=3.9; 3.2 Requires-Python >=3.9; 3.2.1 Requires-Python >=3.9; 3.2rc0 Requires-Python >=3.9; 3.3 Requires-Python >=3.10; 3.3rc0 Requires-Python >=3.10
ERROR: Could not find a version that satisfies the requirement torch==2.0.1+cu118 (from versions: 1.4.0, 1.5.0, 1.5.1, 1.6.0, 1.7.0, 1.7.1, 1.8.0, 1.8.1, 1.9.0, 1.9.1, 1.10.0, 1.10.1, 1.10.2, 1.11.0, 1.12.0, 1.12.1, 1.13.0, 1.13.1, 2.0.0, 2.0.1, 2.1.0, 2.1.1, 2.1.2, 2.2.0, 2.2.1, 2.2.2, 2.3.0)
ERROR: No matching distribution found for torch==2.0.1+cu118
failed
CondaEnvException: Pip failed
Instead, I installed it successfully with:
pip install torch==2.0.1 torchvision==0.15.2 torchaudio==2.0.2 --index-url https://download.pytorch.org/whl/cu118
Considering most of the researchers will prefer to use transformers and pytorch as backend. So, I transfer my work to transformers and train a pytorch model from scratch. The new model is trained with more powerful settings: The max model length is up to 2048 now and the pretraining dataset is the newest version of rnacentral, which contains about 31 million RNA sequences after length filtering (<2048). This pytorch version model has been uploaded to huggingface at https://huggingface.co/WANGNingroci/RNAErnie and the training framework/tokenization is located at https://github.com/CatIIIIIIII/RNAErnie2. (NOTE: the tokenization is a little different from the original paddle implementation). Moreover, Multimolecule are implementing current most powerful RNA language model with transformers and pytorch. Our model also could be accessed at https://huggingface.co/multimolecule/rnaernie.
I tried to install it but encountered the following error: