L3Cube-MahaCorpus and MahaBERT: Marathi Monolingual Corpus, Marathi BERT Language Models, and Resources
Paper • 2202.01159 • Published
How to use l3cube-pune/marathi-gpt with Transformers:
# Use a pipeline as a high-level helper
from transformers import pipeline
pipe = pipeline("text-generation", model="l3cube-pune/marathi-gpt") # Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("l3cube-pune/marathi-gpt")
model = AutoModelForCausalLM.from_pretrained("l3cube-pune/marathi-gpt")How to use l3cube-pune/marathi-gpt with vLLM:
# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "l3cube-pune/marathi-gpt"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
-H "Content-Type: application/json" \
--data '{
"model": "l3cube-pune/marathi-gpt",
"prompt": "Once upon a time,",
"max_tokens": 512,
"temperature": 0.5
}'docker model run hf.co/l3cube-pune/marathi-gpt
How to use l3cube-pune/marathi-gpt with SGLang:
# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
--model-path "l3cube-pune/marathi-gpt" \
--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": "l3cube-pune/marathi-gpt",
"prompt": "Once upon a time,",
"max_tokens": 512,
"temperature": 0.5
}'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 "l3cube-pune/marathi-gpt" \
--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": "l3cube-pune/marathi-gpt",
"prompt": "Once upon a time,",
"max_tokens": 512,
"temperature": 0.5
}'How to use l3cube-pune/marathi-gpt with Docker Model Runner:
docker model run hf.co/l3cube-pune/marathi-gpt
MahaGPT is a Marathi GPT2 model. It is a GPT2 model pre-trained on L3Cube-MahaCorpus and other publicly available Marathi monolingual datasets. [dataset link] (https://github.com/l3cube-pune/MarathiNLP)
More details on the dataset, models, and baseline results can be found in our [paper] (https://arxiv.org/abs/2202.01159)
@article{joshi2022l3cube,
title={L3Cube-MahaCorpus and MahaBERT: Marathi Monolingual Corpus, Marathi BERT Language Models, and Resources},
author={Joshi, Raviraj},
journal={arXiv preprint arXiv:2202.01159},
year={2022}
}