Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+webapp/static/source.gif filter=lfs diff=lfs merge=lfs -text

README.md

@@ -1,3 +1,84 @@
----
-license: mit
----
+# Pazham     🎯
+A machine learning model that predicts multiple features of a banana based on its physical characteristics:
+1. Number of seeds
+2. Curvature (in degrees)
+
+
+## Basic Details
+### Team Name: (AB)²
+
+
+### Team Members
+- Team Lead: Atul Biju - Adi Shankara Institute of Engineering and Technology
+- Member 2: Amal Babu - Adi Shankara Institute of Engineering and Technology
+
+
+## Overview
+
+This project uses a Random Forest Regressor to predict multiple banana characteristics based on various physical features. The model achieves good accuracy (R² scores > 0.80) on synthetic data and can be retrained with real-world data.
+
+
+## Features
+
+### Input Features
+The model takes the following measurements as input:
+- Length (centimeters)
+- Width (centimeters)
+- Weight (grams)
+- Ripeness level (scale 1-5)
+- Color (1=green, 2=yellow, 3=brown)
+
+### Predictions
+The model predicts:
+1. Number of seeds
+2. Curvature (degrees)
+
+## Requirements
+
+- Python 3.x
+- Required packages:
+  - numpy
+  - pandas
+  - scikit-learn
+
+## Usage
+
+The model is implemented in a Jupyter notebook (`model.ipynb`). To use it:
+
+1. Open `model.ipynb` in Jupyter or VS Code
+2. Run all cells to train the model
+3. Use the `predict_seeds()` function with your banana measurements
+
+Example usage:
+```python
+predictions = predict_banana_features(
+    length=16,    # cm
+    width=3.2,    # cm
+    weight=130,   # g
+    ripeness=4,   # scale 1-5
+    color=2       # yellow
+)
+
+print(f"Predicted seeds: {predictions['seeds']}")
+print(f"Predicted curvature: {predictions['curvature']}°")
+```
+
+## Model Performance
+
+Current model metrics on synthetic data:
+- Mean Squared Error: 0.20
+- R² Score: 0.80
+
+Note: These metrics are based on synthetic training data. Performance may vary with real-world data.
+
+## Future Improvements
+
+- Replace synthetic data with real banana measurements
+- Add image processing to automatically extract features
+- Implement cross-validation
+- Add visualization of feature importance
+- Create a simple web interface for predictions
+
+## License
+
+[MIT License](LICENSE)

best_model.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a8ae069d5c491106c116b515a527e3918d3932e336a846b99cc5c78f86966472
+size 46365583

requirements.txt

@@ -0,0 +1,10 @@
+torch>=2.7.1
+torchvision>=0.22.1
+Flask>=3.1.1
+Pillow>=11.3.0
+numpy>=2.2.6
+pandas>=2.3.1
+scikit-learn>=1.7.1
+werkzeug>=3.1.3
+opencv-python>=4.12.0
+matplotlib>=3.10.3

training_data/clean_dataset.csv

@@ -0,0 +1,32 @@
+image_filename,length_cm,width_cm,weight_g,ripeness,color_code,seed_count,curvature_degrees
+image/p4.jpeg,19.5,4.5,170.0,4,2,380,285.0
+image/p5.jpeg,22.0,4.2,170.0,4,2,427,190.0
+image/p6.jpeg,20.5,4.31,152.8,5,2,103,250.5
+image/p7.jpeg,19.86,4.68,157.2,4,1,392,271.6
+image/p8.jpeg,20.65,4.4,143.4,4,2,273,202.4
+image/p9.jpeg,21.52,4.24,142.1,5,2,274,188.6
+image/p11.jpeg,19.77,4.22,180.3,4,1,179,265.2
+image/p14.jpeg,21.58,4.43,158.9,4,1,102,205.3
+image/p15.jpeg,20.77,4.11,175.1,5,1,221,172.5
+image/p16.jpeg,19.53,4.2,165.4,5,2,267,282.3
+image/p17.jpeg,20.54,4.43,150.3,1,2,209,211.7
+image/p18.jpeg,19.54,4.51,165.4,1,1,247,227.3
+image/p19.jpeg,19.53,4.43,183.1,2,1,296,268.9
+image/p20.jpeg,20.24,4.38,159.5,1,1,373,247.8
+image/p21.jpeg,18.09,4.35,183.5,3,1,252,255.6
+image/p24.jpeg,18.28,4.18,120.7,2,1,233,187.2
+image/p25.jpeg,19.44,4.29,172.3,5,1,172,230.4
+image/p26.jpeg,18.99,4.33,161.3,4,3,304,250.5
+image/p27.jpeg,20.31,4.56,155.5,4,1,343,189.7
+image/p28.jpeg,19.09,4.45,161.4,4,1,329,182.7
+image/p29.jpeg,18.59,4.14,130.2,5,2,374,287.7
+image/p30.jpeg,21.47,4.45,156.7,2,1,421,227.2
+image/p31.jpeg,19.77,4.34,165.4,2,3,147,193.9
+image/p34.jpeg,20.07,4.3,182.2,2,3,247,289.5
+image/p35.jpeg,18.58,4.49,152.2,2,1,289,285.5
+image/p36.jpeg,19.46,4.55,147.9,2,1,193,232.6
+image/p37.jpeg,20.11,4.54,152.5,1,1,146,274.7
+image/p38.jpeg,18.85,4.27,173.7,1,1,250,184.5
+image/p39.jpeg,20.38,4.35,164.9,3,1,173,196.3
+image/p40.jpeg,19.4,4.45,152.1,2,1,236,251.0
+image/p41.jpeg,19.71,4.55,167.7,3,1,375,270.1

training_data/dataset.csv

@@ -0,0 +1,33 @@
+image_filename,length_cm,width_cm,weight_g,ripeness,color_code,seed_count,curvature_degrees
+image/p4.jpeg,19.5,4.5,170,4,2,380,285
+image/p5.jpeg,22,4.2,170,4,2,427,190
+image/p6.jpeg,20.5,4.31,152.8,5,2,103,250.5
+image/p7.jpeg,19.86,4.68,157.2,4,1,392,271.6
+image/p8.jpeg,20.65,4.4,143.4,4,2,273,202.4
+image/p9.jpeg,21.52,4.24,142.1,5,2,274,188.6
+image/p0.jpeg,19.77,4.52,172.2,5,1,175,252.4
+image/p11.jpeg,19.77,4.22,180.3,4,1,179,265.2
+image/p14.jpeg,21.58,4.43,158.9,4,1,102,205.3
+image/p15.jpeg,20.77,4.11,175.1,5,1,221,172.5
+image/p16.jpeg,19.53,4.2,165.4,5,2,267,282.3
+image/p17.jpeg,20.54,4.43,150.3,1,2,209,211.7
+image/p18.jpeg,19.54,4.51,165.4,1,1,247,227.3
+image/p19.jpeg,19.53,4.43,183.1,2,1,296,268.9
+image/p20.jpeg,20.24,4.38,159.5,1,1,373,247.8
+image/p21.jpeg,18.09,4.35,183.5,3,1,252,255.6
+image/p24.jpeg,18.28,4.18,120.7,2,1,233,187.2
+image/p25.jpeg,19.44,4.29,172.3,5,1,172,230.4
+image/p26.jpeg,18.99,4.33,161.3,4,3,304,250.5
+image/p27.jpeg,20.31,4.56,155.5,4,1,343,189.7
+image/p28.jpeg,19.09,4.45,161.4,4,1,329,182.7
+image/p29.jpeg,18.59,4.14,130.2,5,2,374,287.7
+image/p30.jpeg,21.47,4.45,156.7,2,1,421,227.2
+image/p31.jpeg,19.77,4.34,165.4,2,3,147,193.9
+image/p34.jpeg,20.07,4.3,182.2,2,3,247,289.5
+image/p35.jpeg,18.58,4.49,152.2,2,1,289,285.5
+image/p36.jpeg,19.46,4.55,147.9,2,1,193,232.6
+image/p37.jpeg,20.11,4.54,152.5,1,1,146,274.7
+image/p38.jpeg,18.85,4.27,173.7,1,1,250,184.5
+image/p39.jpeg,20.38,4.35,164.9,3,1,173,196.3
+image/p40.jpeg,19.4,4.45,152.1,2,1,236,251
+image/p41.jpeg,19.71,4.55,167.7,3,1,375,270.1