Upload folder using huggingface_hub

README.md

@@ -5,7 +5,6 @@ tags:
 - food
 - reranking
 - sequence-classification
-- text-classification
 library_name: transformers
 pipeline_tag: text-classification
 license: mit
@@ -13,25 +12,41 @@ license: mit
 
 # Food Re-ranker Model
 
-This is a fine-tuned DistilBERT model trained for re-ranking food descriptions based on semantic similarity. The model is optimized for matching food item descriptions in a nutrition and diet tracking context.
+This is a fine-tuned DistilBERT model trained for binary classification of food description pairs. The model determines whether two food descriptions refer to the same item, enabling accurate re-ranking of search results.
 
-## Model Description
+## Model Details
 
-- **Base Model**: DistilBertForSequenceClassification
-- **Training Data**: Food description pairs with binary labels for matching/non-matching
-- **Task**: Binary classification to determine if two food descriptions refer to the same item
+### Model Description
+- **Base Model**: distilbert-base-uncased
+- **Model Type**: distilbert
+- **Task**: Binary classification (food description matching)
 - **Output**: Binary classification scores (0 = different foods, 1 = same food)
 
-## Use Case
+### Architecture
+- **Hidden Dimension**: 768
+- **Number of Layers**: 6
+- **Number of Attention Heads**: 12
+- **Intermediate Size**: 3072
+- **Maximum Position Embeddings**: 512
+- **Vocabulary Size**: 30522
 
-This model is designed to improve food search accuracy by re-ranking initial search results. It takes pairs of food descriptions and determines their semantic similarity, helping to surface the most relevant matches.
+## Use Case
 
-## Training
+Designed for improving food search accuracy by re-ranking initial search results, this model:
+- Takes pairs of food descriptions as input
+- Determines if they refer to the same food item
+- Enables more accurate matching of food descriptions
+- Helps surface the most relevant matches in search results
 
-The model was trained on carefully curated food description pairs with binary classification:
-- Label 1: Pairs of descriptions that refer to the same food item
-- Label 0: Pairs of descriptions that refer to different food items
-- Uses cross-entropy loss for binary classification
+## Training Configuration
+- **Batch Size**: 32
+- **Learning Rate**: 5e-05
+- **Number of Epochs**: 3
+- **Warmup Steps**: 0
+- **Weight Decay**: 0.0
+- **Dropout**: 0.1
+- **Attention Dropout**: 0.1
+- **Classification Dropout**: 0.2
 
 ## Example Usage
 
@@ -50,3 +65,4 @@ inputs = tokenizer(query, candidate, padding=True, truncation=True, return_tenso
 # Get prediction
 outputs = model(**inputs)
 score = outputs.logits.softmax(dim=1)[0][1].item()  # Score for positive class
+```