Instructions to use AI4PD/ZymCTRL with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use AI4PD/ZymCTRL with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="AI4PD/ZymCTRL")# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("AI4PD/ZymCTRL") model = AutoModelForCausalLM.from_pretrained("AI4PD/ZymCTRL") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use AI4PD/ZymCTRL with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "AI4PD/ZymCTRL" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "AI4PD/ZymCTRL", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/AI4PD/ZymCTRL
- SGLang
How to use AI4PD/ZymCTRL 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 "AI4PD/ZymCTRL" \ --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": "AI4PD/ZymCTRL", "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 "AI4PD/ZymCTRL" \ --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": "AI4PD/ZymCTRL", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use AI4PD/ZymCTRL with Docker Model Runner:
docker model run hf.co/AI4PD/ZymCTRL
Hit handling for bulk generation
#9
by fmoorhof - opened
Hey there,
thanks for the very easy to use tool. To make the handling of output files as straight forward I wanted to share a quick way of handling the many output files: The script is extracting the lowest perplexity hits for each EC number (in the folder_path) and writes them as lowest_perplexity_{ec_number_batch}.fasta:
import os
from collections import defaultdict
from Bio import SeqIO
# Folder containing .fasta files
folder_path = "Your_path_goes_here"
# Dictionary to store records grouped by EC number batch
batch_records = defaultdict(list)
# Iterate through all .fasta files in the folder
for filename in os.listdir(folder_path):
if filename.endswith("_0.fasta"):
file_path = os.path.join(folder_path, filename)
with open(file_path, "r") as fasta_file:
for record in SeqIO.parse(fasta_file, "fasta"):
# Extract perplexity value from the record description
description_parts = record.description.split()
perplexity = float(description_parts[-1])
# Extract EC number batch from the filename
ec_number_batch = filename.split("_")[0]
# Store the record and perplexity value as a tuple in the dictionary
batch_records[ec_number_batch].append((record, perplexity))
# Iterate through the batches and select the record with the lowest perplexity
for ec_number_batch, records in batch_records.items():
records.sort(key=lambda x: x[1]) # Sort records by perplexity
lowest_perplexity_record = records[0][0] # Get the record with the lowest perplexity
# Write the lowest perpl-exity record to a file for each batch
output_file_path = f"{folder_path}/lowest_perplexity_{ec_number_batch}.fasta"
with open(output_file_path, "w") as output_file:
SeqIO.write(lowest_perplexity_record, output_file, "fasta")