Instructions to use nferruz/ProtGPT2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use nferruz/ProtGPT2 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="nferruz/ProtGPT2")# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("nferruz/ProtGPT2") model = AutoModelForCausalLM.from_pretrained("nferruz/ProtGPT2") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use nferruz/ProtGPT2 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "nferruz/ProtGPT2" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "nferruz/ProtGPT2", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/nferruz/ProtGPT2
- SGLang
How to use nferruz/ProtGPT2 with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "nferruz/ProtGPT2" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "nferruz/ProtGPT2", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "nferruz/ProtGPT2" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "nferruz/ProtGPT2", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use nferruz/ProtGPT2 with Docker Model Runner:
docker model run hf.co/nferruz/ProtGPT2
Generate mutation by define specific stations
Hi,
First of all - thank you very much for sharing ProtGPT2.
I read the article as part of my thesis, and started playing with the model.
My question:
Is it possible to define which station/s in natural sequence I want ProtGPT2 will change? if yes - how?
For example:
I want to change only the 3th station in sequence, so that:
Natural seq: "DQSV..."
Possible generated mutant: "DQGV..."
Best Regards,
Aviv.
Hi Aviv,
Sorry I had missed your message. I’m afraid it is not possible to change directly some residues. What you could do though, is to input the first residues, and see what tokens the model considers most likely afterwards. This won’t directly mutate your sequence but it would tell you what residue is most likely after a set of residues. I haven’t tried this myself but it should be possible. Let me know how it goes!
Noelia
Hi Noelia,
I want to generate several sequences with "Cys" as the first residue.I wonder if it is possible to finetune the model with sequences like that. Or there are any other ways?
Thanks!
Gandi.
Hi Gandi11,
yes it would be possible. Find a dataset of sequences that start with 'C' and fine-tune the model. It should then most likely generate sequences that follow that pattern.
Best wishes