Instructions to use multimolecule/malinois with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MultiMolecule
How to use multimolecule/malinois with MultiMolecule:
pip install multimolecule
from multimolecule import AutoModel, AutoTokenizer tokenizer = AutoTokenizer.from_pretrained("multimolecule/malinois") model = AutoModel.from_pretrained("multimolecule/malinois") inputs = tokenizer("ACTCCCCTGCCCTCAACAAGATGTTTTGCCAACTGGCCAAGACCTGCCCTGTGCAGCTGTGGGTTGATTCCACACCCCCGCCCGGCACCCGCGTCCGCGCCATGGCCATCTACAAGCAGTCACAGCACATGACGGAGGTTGTGAGGCGCTGCCCCCACCATGAGCGCTGCTCAGATAGCGATGG", return_tensors="pt") outputs = model(**inputs) embeddings = outputs.last_hidden_state - Notebooks
- Google Colab
- Kaggle
Download pytorch_model.bin from multimolecule/malinois: direct link, hf CLI and curl.
- Browser
- Download file 16.5 MB
-
https://huggingface.co/multimolecule/malinois/resolve/main/pytorch_model.bin
- Command line
-
hf download hf://multimolecule/malinois/pytorch_model.bin
-
curl -L -o pytorch_model.bin https://huggingface.co/multimolecule/malinois/resolve/main/pytorch_model.bin
16.5 MB
- Xet hash:
- 87b801c5fba70372ebc540c45a5fb19dd80334d265604f9454e03e35a4783fa0
- Size of remote file:
- 16.5 MB
- SHA256:
- f36e59e863254b3640ababd5e9286efd49112fd10cf0758f08ce6c74632cac6d
·
Xet efficiently stores Large Files inside Git, intelligently splitting files into unique chunks and accelerating uploads and downloads. More info.