Upload content-moderation model for NEED AI

README.md

@@ -1,55 +1,68 @@
 ---
+language: en
 license: mit
 tags:
 - text-generation
-- transformers
+- question-answering
+- sentence-similarity
+- text-classification
 - need-ai
-- service-marketplace
+base_model: unitary/toxic-bert
 pipeline_tag: text-classification
-widget:
-- text: "This is a great service!"
 ---
 
 # NEED AI - Content Moderation
 
-This model is part of the NEED AI system for service marketplace intelligence.
+BERT model for detecting toxic and inappropriate content in user messages and reviews.
 
 ## Model Details
-- **Task**: text-classification
-- **Base Model**: Toxic-BERT
-- **Fine-tuned for**: NEED Service App
+
+- **Base Model**: unitary/toxic-bert
+- **Task**: Text Classification
+- **Fine-tuned for**: NEED Service Marketplace Platform
+- **Language**: English
+- **License**: MIT
 
 ## Usage
 
 ```python
-from transformers import AutoModelForSeq2SeqLM, AutoTokenizer
+from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
+# or AutoModelForSequenceClassification for moderation
+# or SentenceTransformer for semantic-search
 
-model = AutoModelForSeq2SeqLM.from_pretrained("yogami9/need-content-moderation")
-tokenizer = AutoTokenizer.from_pretrained("yogami9/need-content-moderation")
+model_name = "yogami9/need-content-moderation"
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
 
-input_text = "This is a great service!"
+# Example usage
+input_text = "Your input here"
 inputs = tokenizer(input_text, return_tensors="pt")
 outputs = model.generate(**inputs)
 result = tokenizer.decode(outputs[0], skip_special_tokens=True)
 print(result)
 ```
 
-## API Usage
+## Training Data
 
-```python
-import requests
+Trained on curated datasets specific to the NEED platform's service categories and user interactions.
 
-API_URL = "https://api-inference.huggingface.co/models/yogami9/need-content-moderation"
-headers = {"Authorization": "Bearer YOUR_HF_TOKEN"}
+## Limitations
 
-def query(payload):
-    response = requests.post(API_URL, headers=headers, json=payload)
-    return response.json()
+- Optimized for English language
+- Best performance on NEED platform-specific queries
+- May require fine-tuning for other domains
 
-result = query({"inputs": "This is a great service!"})
-print(result)
-```
+## Contact
 
-## Links
-- **GitHub**: https://github.com/Need-Service-App/need-ai-model
+- **Organization**: NEED Service App
 - **Email**: needserviceapp@gmail.com
+- **GitHub**: https://github.com/Need-Service-App
+
+## Related Models
+
+All NEED AI models:
+- [Category Recommendation](https://huggingface.co/yogami9/need-category-recommendation)
+- [Chat Support](https://huggingface.co/yogami9/need-chat-support)
+- [Service Description](https://huggingface.co/yogami9/need-service-description)
+- [Semantic Search](https://huggingface.co/yogami9/need-semantic-search)
+- [Content Moderation](https://huggingface.co/yogami9/need-content-moderation)