|
|
--- |
|
|
base_model: meta-llama/Llama-3.1-8B-Instruct |
|
|
library_name: peft |
|
|
license: mit |
|
|
datasets: |
|
|
- imnim/multiclass-email-classification |
|
|
language: |
|
|
- en |
|
|
tags: |
|
|
- Email-classifier |
|
|
- Email-labelling |
|
|
- Fine-tuning |
|
|
- peft |
|
|
- lora |
|
|
--- |
|
|
|
|
|
# Model Card for Model ID |
|
|
|
|
|
<!-- Provide a quick summary of what the model is/does. --> |
|
|
Model is finetuned for the task of email labelling. It labels the given email into one or more than one categories based on email subject and email body. |
|
|
|
|
|
|
|
|
|
|
|
## Model Details |
|
|
|
|
|
### Model Description |
|
|
|
|
|
<!-- Provide a longer summary of what this model is. --> |
|
|
The model classifies emails into the following 10 categories: "Business", "Personal", "Promotions", "Customer Support", "Job Application", |
|
|
"Finance & Bills", "Events & Invitations", "Travel & Bookings", "Reminders", "Newsletters" |
|
|
|
|
|
I have prepared a synthetic but realistic dataset of 2,105 labeled emails. Each email includes a subject, body, and one or more categories. |
|
|
|
|
|
|
|
|
|
|
|
- **Developed by:** imnim |
|
|
- **Model type:** text-to-text |
|
|
- **Language(s) (NLP):** English |
|
|
- **Finetuned from model:** Llama-3.1-8B-Instruct |
|
|
|
|
|
|
|
|
### Model Sources |
|
|
|
|
|
<!-- Provide the basic links for the model. --> |
|
|
|
|
|
- **Repository:** https://github.com/contributerMe/multi-label-email-classifier |
|
|
- **Demo:** https://huggingface.co/spaces/imnim/Multi-labelEmailClassifier |
|
|
|
|
|
|
|
|
## Technical Specifications |
|
|
|
|
|
### Model Architecture and Objective |
|
|
Auto-regressive language model that uses an optimized transformer architecture. |
|
|
|
|
|
|
|
|
### Compute Infrastructure |
|
|
|
|
|
Kaggle Notebook |
|
|
|
|
|
#### Hardware |
|
|
|
|
|
Trained on Kaggle's P100 GPU |
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- PEFT 0.15.2 |
