---
license: mit
datasets:
  - custom
language:
  - en
metrics:
  - accuracy
  - f1
pipeline_tag: text-classification
library_name: sklearn
tags:
  - agriculture
  - logistic-regression
  - tfidf
  - binary-classification
  - crop-health
model_name: Agriculture Text Classifier
model_creator: PopeJohn
model_type: sklearn
model_description: A logistic regression classifier trained on agricultural text using TF-IDF features.
---
🌱 Agriculture Text Classifier

**Model owner:** [PopeJohn](https://huggingface.co/PopeJohn)  
**Repository:** [PopeJohn/agriculture-model](https://huggingface.co/PopeJohn/agriculture-model)

---

## 📝 Overview
This model is a **Logistic Regression** classifier trained on agricultural text data, using **TF–IDF vectorization** for feature extraction.  
It predicts predefined agriculture-related categories from short text inputs, making it useful for tasks like farmer query routing, agronomic content tagging, and agricultural market analysis.

---

## 📂 Files in this repository
- `agriculture_model.pkl` — Trained Logistic Regression model
- `vectorizer.pkl` — Fitted TF–IDF vectorizer for text preprocessing

---

## 🔍 Intended Use
This model is designed for:
- Classifying farmer questions into crop/disease categories
- Indexing or tagging agricultural content
- Supporting NLP pipelines in agriculture-focused applications

Not intended for:
- Real-time critical decision-making without human verification
- Non-agriculture domains without fine-tuning

---

## ⚙️ How to Use

```python
from huggingface_hub import hf_hub_download
import joblib

# Download files from Hugging Face Hub
model_path = hf_hub_download("PopeJohn/agriculture-model", "agriculture_model.pkl")
vectorizer_path = hf_hub_download("PopeJohn/agriculture-model", "vectorizer.pkl")

# Load
model = joblib.load(model_path)
vectorizer = joblib.load(vectorizer_path)

# Predict
sample_text = ["Healthy maize crop after seasonal rains"]
prediction = model.predict(vectorizer.transform(sample_text))
print(prediction[0])