Instructions to use ScriptEdgeAI/MarathiSentiment-Bloom-560m with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use ScriptEdgeAI/MarathiSentiment-Bloom-560m with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="ScriptEdgeAI/MarathiSentiment-Bloom-560m")# Load model directly from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_pretrained("ScriptEdgeAI/MarathiSentiment-Bloom-560m") model = AutoModelForSequenceClassification.from_pretrained("ScriptEdgeAI/MarathiSentiment-Bloom-560m") - Notebooks
- Google Colab
- Kaggle
metadata
language:
- mr
tags:
- mr
- Sentiment-Analysis
license: cc-by-nc-4.0
widget:
- text: मला तुम्ही आवडता. मी तुझ्यावर प्रेम करतो.
Marathi-Bloom-560m is a Bloom fine-tuned model trained by ScriptEdge on MahaNLP tweets dataset from L3Cube-MahaNLP.
Worked on by:
Trained by:
- Venkatesh Soni.
Assistance:
- Rayansh Srivastava.
Supervision:
- Akshay Ugale, Madhukar Alhat.
Usage -
- It is intended for non-commercial usages.
Model best metrics
| Model | Data | Accuracy |
|---|---|---|
| bigscience/bloom-560m | Validation | 34.7 |
| bigscience/bloom-560m | Test | 34.8 |
| ScriptEdgeAI/MarathiSentiment-Bloom-560m | Validation | 76.0 |
| ScriptEdgeAI/MarathiSentiment-Bloom-560m | Test | 77.0 |
Citation to L3CubePune by the dataset usage.
@article {joshi2022l3cube,
title= {L3Cube-MahaNLP: Marathi Natural Language Processing Datasets, Models, and Library},
author= {Joshi, Raviraj},
journal= {arXiv preprint arXiv:2205.14728},
year= {2022}
}