Instructions to use zeeshanali01/cryptotunned with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use zeeshanali01/cryptotunned with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="zeeshanali01/cryptotunned")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("zeeshanali01/cryptotunned")
model = AutoModelForCausalLM.from_pretrained("zeeshanali01/cryptotunned", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use zeeshanali01/cryptotunned with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "zeeshanali01/cryptotunned"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "zeeshanali01/cryptotunned",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/zeeshanali01/cryptotunned

SGLang

How to use zeeshanali01/cryptotunned 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 "zeeshanali01/cryptotunned" \
    --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": "zeeshanali01/cryptotunned",
		"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 "zeeshanali01/cryptotunned" \
        --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": "zeeshanali01/cryptotunned",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use zeeshanali01/cryptotunned with Docker Model Runner:
```
docker model run hf.co/zeeshanali01/cryptotunned
```

cryptotunned / README.md

zeeshanali01

Update README.md

455b3a4 verified almost 2 years ago

preview code

Raw

History Blame Contribute Delete

2.72 kB

metadata

license: apache-2.0
tags:
  - trl
  - sft
library_name: transformers

CatMemo: Fine-Tuning Large Language Models for Financial Applications

Model Overview

This model, CatMemo, is fine-tuned using Data Fusion techniques for financial applications. It was developed as part of the FinLLM Challenge Task and focuses on enhancing the performance of large language models in finance-specific tasks such as question answering, document summarization, and sentiment analysis.

Key Features

Fine-tuned on financial datasets using Supervised Fine-Tuning (SFT) techniques.
Optimized for Transfer Reinforcement Learning (TRL) workflows.
Specialized for tasks requiring domain-specific context in financial applications.

Usage

You can use this model with the Hugging Face Transformers library to perform financial text analysis. Below is a quick example:

from transformers import AutoTokenizer, AutoModelForCausalLM

# Load the model and tokenizer
model_name = "zeeshanali01/cryptotunned"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)

# Tokenize input
inputs = tokenizer("What are the key takeaways from the latest earnings report?", return_tensors="pt")

# Generate output
outputs = model.generate(**inputs)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

Training Details

This model was fine-tuned using Data Fusion methods on domain-specific financial datasets. The training pipeline includes:

Preprocessing financial documents and datasets to enhance model understanding.
Applying Supervised Fine-Tuning (SFT) to optimize the model for financial NLP tasks.
Testing and evaluation on FinLLM benchmark tasks.

Citation

If you use this model, please cite our work:

@inproceedings{cao2024catmemo,
  title={CatMemo at the FinLLM Challenge Task: Fine-Tuning Large Language Models using Data Fusion in Financial Applications},
  author={Cao, Yupeng and Yao, Zhiyuan and Chen, Zhi and Deng, Zhiyang},
  booktitle={Joint Workshop of the 8th Financial Technology and Natural Language Processing (FinNLP) and the 1st Agent AI for Scenario Planning (AgentScen) in conjunction with IJCAI 2023},
  pages={174},
  year={2024}
}

License

This model is licensed under the Apache 2.0 License. See the LICENSE file for details.

Acknowledgments

We thank the organizers of the FinLLM Challenge Task for providing the benchmark datasets and tasks used to develop this model.

Model Card Metadata

License: Apache 2.0
Tags: TRL, SFT
Library Used: Transformers