UrFavB0i's picture
Update README.md
6b779f0 verified
metadata
language:
  - en
license: apache-2.0
library_name: transformers
tags:
  - text-generation-inference
datasets:
  - UrFavB0i/skincare-ecommerce-FAQ
pipeline_tag: text-generation
base_model: tiiuae/falcon-7b

Model Card for Model ID

Falcon-7B Fine-Tuned Chatbot Model This repository contains the fine-tuned Falcon-7B model for a chatbot application. The model has been fine-tuned using the PEFT method to provide robust responses for e-commerce customer support. It guides buyers in product selection, recommends sizes, checks product stock, suggests similar products, and presents reviews and social media video links.

Model Details

  • Base Model: Falcon 7B (tiiuae/falcon-7b)
  • Fine-Tuning Method: Parameter-Efficient Fine-Tuning (PEFT)
  • Training Data : Custom dataset including skincare e-commerce related dialogues. (UrFavB0i/skincare-ecommerce-FAQ)

Features

  • 24/7 customer support
  • Product selection guidance
  • Size recommendations
  • Product stock checks
  • Similar product suggestions
  • Reviews and social media video link presentation

Usage

Installation

To use the model, you need to install the necessary dependencies. Make sure you have Python 3.7+ and pip installed.

pip install transformers
pip install peft

Loading the Model

You can load the fine-tuned model using the transformers library:

from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "your-huggingface-username/falcon-7b-chatbot"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)

Example usage

inputs = tokenizer("Hello, how can I assist you today?", return_tensors="pt")
outputs = model.generate(**inputs)
print(tokenizer.decode(outputs[0]))

Training Details

The model was fine-tuned using the PEFT method on a dataset specifically curated for e-commerce scenarios. The training process involved:

  • Data Preparation: Gathering and preprocessing e-commerce-related dialogues.
  • Fine-Tuning: Training the base model using PEFT to adapt it to the specific needs of the e-commerce domain.

Evaluation

The fine-tuned model was evaluated based on its ability to handle various e-commerce related queries, providing accurate and contextually appropriate responses.

Limitations

While the model performs well in many scenarios, it might not handle extremely rare or out-of-domain queries perfectly. Continuous training and updating with more data can help improve its performance further.

Contributing

We welcome contributions to improve this model. If you have any suggestions or find any issues, please create an issue or a pull request.

License

This project is licensed under the Apache 2.0 License. See the [LICENSE] file for more details.

Acknowledgements

Special thanks to the Falcon team and the creators of the tiiuae/falcon-7b model for providing the base model and the tools necessary for fine-tuning.