Vrizzo/agri_slm
Vrizzo/agri_slm is a 1.4 billion parameter language model, fine-tuned from microsoft/phi-1_5, specifically designed for crop recommendation. It processes soil nutrient (N-P-K), pH, temperature, humidity, and rainfall data to suggest one of 22 specific crops. This model excels at mapping environmental conditions to agricultural recommendations, providing a focused solution for crop selection.
Loading preview...
Agri-SLM India: Crop Recommendation Model
Vrizzo/agri_slm is a specialized 1.4 billion parameter language model, fine-tuned using a LoRA adapter on the microsoft/phi-1_5 architecture. Its primary function is to provide crop recommendations based on specific environmental and soil conditions.
Key Capabilities
- Crop Recommendation: Given detailed soil nutrient (N-P-K), pH, temperature, humidity, and rainfall data, the model recommends one of 22 specific crops.
- Data-Driven: Trained on a dataset of 2,200 real records from the Kaggle Crop Recommendation Dataset by Atharva Ingle.
Limitations and Specificity
This model is highly specialized and does not provide general farming advice, pest/disease guidance, or yield predictions. It is crucial to understand that the model will generate plausible-sounding but fabricated information for queries outside its trained scope (e.g., specific numeric yield figures). Users should only rely on its crop recommendation output.
Usage
To interact with the model, use the prompt format: Question: <soil/climate description>\nAnswer: