|
|
--- |
|
|
license: mit |
|
|
language: |
|
|
- en |
|
|
metrics: |
|
|
- accuracy |
|
|
- precision |
|
|
- recall |
|
|
- f1 |
|
|
- confusion_matrix |
|
|
pipeline_tag: tabular-classification |
|
|
library_name: sklearn |
|
|
tags: |
|
|
- Water Potability |
|
|
- Random Forest |
|
|
- Standard Scaler |
|
|
--- |
|
|
|
|
|
# 💧 HydraSense - Water Potability Classifier Model (v1.0) |
|
|
|
|
|
A lightweight **Random Forest + StandardScaler** based water potability prediction model developed by **DarkNeuronAI**. |
|
|
It classifies water as **Potable (1)** or **Not Potable (0)** based on chemical and physical features — ideal for simple tabular classification tasks. |
|
|
|
|
|
--- |
|
|
|
|
|
## 🚀 Features |
|
|
- Fast and efficient — runs easily on standard laptops |
|
|
- Trained with real-world water quality datasets |
|
|
- Predicts potability from features like **pH, Hardness, Solids, Chloramines, Sulfate, Conductivity, Organic Carbon, Trihalomethanes, Turbidity** |
|
|
- Uses a **pipeline** to automatically scale and preprocess input data |
|
|
- Easy to use and integrate |
|
|
|
|
|
--- |
|
|
|
|
|
## 🚀 Model Overview |
|
|
- **Algorithm:** Random Forest Classifier |
|
|
- **Preprocessing:** StandardScaler (automatic feature scaling) |
|
|
- **Goal:** Predict whether water is safe to drink (Potable) or unsafe (Not Potable) |
|
|
- **Performance:** Accurate classification on real-world datasets |
|
|
|
|
|
--- |
|
|
|
|
|
## 🧩 Files Included |
|
|
- `water_potability_model.pkl` → Trained Random Forest pipeline (scaler + model) |
|
|
- `example_usage.py` → Example code to use the model |
|
|
- `requirements.txt` → Dependencies list |
|
|
|
|
|
--- |
|
|
|
|
|
## 🏷️ Prediction Labels (Binary) |
|
|
- **0:** Not Potable (Unsafe to drink) |
|
|
- **1:** Potable (Safe to drink) |
|
|
|
|
|
--- |
|
|
|
|
|
## 💡 How to Use (Example Code) |
|
|
```python |
|
|
from huggingface_hub import hf_hub_download |
|
|
import joblib |
|
|
import pandas as pd |
|
|
|
|
|
# Download and load the trained pipeline |
|
|
pipeline_path = hf_hub_download("DarkNeuron-AI/darkneuron-hydrasense-v1", "water_potability_model.pkl") |
|
|
model = joblib.load(pipeline_path) |
|
|
|
|
|
# Example water sample |
|
|
sample_data = { |
|
|
'ph': [7.2], |
|
|
'Hardness': [180], |
|
|
'Solids': [15000], |
|
|
'Chloramines': [8.3], |
|
|
'Sulfate': [350], |
|
|
'Conductivity': [450], |
|
|
'Organic_carbon': [10], |
|
|
'Trihalomethanes': [70], |
|
|
'Turbidity': [3] |
|
|
} |
|
|
|
|
|
sample_df = pd.DataFrame(sample_data) |
|
|
|
|
|
# Predict potability |
|
|
prediction = model.predict(sample_df) |
|
|
|
|
|
print("Prediction:", "💧 Potable" if prediction[0] == 1 else "⚠️ Not Potable") |
|
|
``` |
|
|
|
|
|
# Developed With ❤️ By DarkNeuronAI |
