Update README.md

README.md

@@ -19,23 +19,28 @@ tags:
 # Depth Estimation Using ResNet50 and XGBoost
 
 ## Overview
-This project demonstrates a depth estimation model that predicts the average depth of images using features extracted from a pre-trained ResNet50 model and an XGBoost regressor. The model was trained using the **NYUv2 dataset** hosted on Hugging Face ([0jl/NYUv2](https://huggingface.co/datasets/0jl/NYUv2)). The trained model is saved as `model.pkl` using Python's `pickle` library for easy deployment and reuse.
+This project demonstrates a depth estimation XgBoost Regressor model that predicts the average depth of images provided using features extracted from a pre-trained ResNet50 model.The model was trained upon the **NYUv2 dataset** ([0jl/NYUv2](https://huggingface.co/datasets/0jl/NYUv2)). The trained model is saved as using Python's `pickle` library for easy deployment and reuse.
+
+## License
+This project is licensed under the Apache License 2.0.
+
+## Author
+**Vishal Adithya.A**
+
 ### Loading the Model
 The model is saved as `model.pkl` using `pickle`. You can load and use it as follows:
 
 ```python
-import pickle
-
-# Load the trained model
 with open("model.pkl", "rb") as f:
     model = pickle.load(f)
 
-# Example usage
-features = extract_features("path/to/image.jpg")  # Use the same feature extraction pipeline
+features = extract_features("path/to/image.jpg") 
 predicted_depth = model.predict([features])
-print("Predicted Depth:", predicted_depth[0])
+print(predicted_depth[0])
 ```
-## Features
+extract_features() is a predefined function in the original code which uses ResNet50 to extract features out of the image
+
+## Key Features
 - **Model Architecture**:
   - Feature extraction: ResNet50 (pre-trained on ImageNet, with the top layers removed and global average pooling).
   - Regression: XGBoost, optimized for structured data prediction.
@@ -47,7 +52,7 @@ print("Predicted Depth:", predicted_depth[0])
 - Format: The dataset includes RGB images and corresponding depth maps.
 - Preprocessing:
   - Images were resized to 224x224 pixels to match the input requirements of ResNet50.
-  - Depth maps were converted into single average depth values for each image by taking the mean of the depth map.
+  - Depth maps were converted into single average depth values.
 
 ## Model Training
 1. **Feature Extraction**:
@@ -58,12 +63,12 @@ print("Predicted Depth:", predicted_depth[0])
    - Hyperparameters were tuned using cross-validation techniques for optimal performance.
 
 ## Results
-- **R² Score**: 0.841.
+- **R² Score**: 0.841
 - Performance is reasonable for a first few implementation and can be further improved with additional tuning or by improving feature extraction methods.
 
 ## How to Use
 ### Requirements
-1. Python 3.8+
+1. Python 3.10+
 2. Required libraries:
    - `numpy`
    - `pickle`
@@ -72,7 +77,6 @@ print("Predicted Depth:", predicted_depth[0])
    - `tensorflow`
    - 'scikit-learn'
 
-
 Install the dependencies using pip:
 ```bash
 pip install numpy tensorflow xgboost pickle-mixin datasets scikit-learn
@@ -80,7 +84,7 @@ pip install numpy tensorflow xgboost pickle-mixin datasets scikit-learn
 
 ### Training Pipeline
 If you want to retrain the model, follow these steps:
-NOTE: This pipeline has just the base fundamental code more additional parameter tunings and preprocessing steps were being conducted during the training of the original model 
+ 
 1. Download the **NYUv2 dataset** from Hugging Face:
    ```python
    from datasets import load_dataset
@@ -106,12 +110,7 @@ NOTE: This pipeline has just the base fundamental code more additional parameter
    with open("model.pkl", "wb") as f:
        pickle.dump(regressor, f)
    ```
-
-## License
-This project is licensed under the Apache License 2.0.
-
-## Author
-**Vishal Adithya.A**
+NOTE: This pipeline has just the base fundamental code more additional parameter tunings and preprocessing steps were being conducted during the training of the original model
 
 ## Acknowledgments
 - Hugging Face for hosting the NYUv2 dataset.