Text Classification
Transformers
Safetensors
xlm-roberta
reranker
cross-encoder
bengali
banglish
maternal-health
text-embeddings-inference
Instructions to use asim-alam/maasathi-bge-reranker-v2-m3 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use asim-alam/maasathi-bge-reranker-v2-m3 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="asim-alam/maasathi-bge-reranker-v2-m3")# Load model directly from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_pretrained("asim-alam/maasathi-bge-reranker-v2-m3") model = AutoModelForSequenceClassification.from_pretrained("asim-alam/maasathi-bge-reranker-v2-m3", device_map="auto") - Notebooks
- Google Colab
- Kaggle
MaaSathi BGE Reranker v2 M3
Fine-tuned from BAAI/bge-reranker-v2-m3 for MaaSathi maternal-health retrieval reranking.
This is a full fine-tuned reranker model, not a LoRA adapter.
Before Finetuning
{
"model": "BAAI/bge-reranker-v2-m3",
"rows": 227,
"top1_accuracy": 0.8502202643171806,
"mrr": 0.9103157121879589,
"emergency_total": 79,
"emergency_top1_recall": 0.8354430379746836,
"curated_total": 1,
"curated_top1_recall": 1.0
}
After Finetuning
{
"model": "maasathi-bge-reranker-v2-m3"
"rows": 227,
"top1_accuracy": 0.9691629955947136,
"mrr": 0.9845814977973568,
"emergency_total": 79,
"emergency_top1_recall": 0.9873417721518988,
"curated_total": 1,
"curated_top1_recall": 1.0
}
- Downloads last month
- 4
Model tree for asim-alam/maasathi-bge-reranker-v2-m3
Base model
BAAI/bge-reranker-v2-m3