Skip to content

Commit

Permalink
Create README.md
Browse files Browse the repository at this point in the history
  • Loading branch information
adi271001 authored Jul 21, 2024
1 parent 9c25cb9 commit 2cc8bdb
Showing 1 changed file with 212 additions and 0 deletions.
212 changes: 212 additions & 0 deletions E-commerce Shipping Data Analysis/Model/README.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,212 @@
### Models README with Conclusion
# Models Implemented

In this project, the following machine learning models were implemented and evaluated on the preprocessed datasets:

## Models List
1. Random Forest
2. Logistic Regression
3. Gradient Boosting
4. AdaBoost
5. CatBoost
6. LightGBM
7. XGBoost
8. Extra Trees
9. K-Nearest Neighbors
10. Decision Tree

## Performance of the Models based on Accuracy Scores

### Non-PCA and Outliers Deleted Data
- **Random Forest**
- Train Accuracy: `0.7036`
- Test Accuracy: `0.6506`

- **Logistic Regression**
- Train Accuracy: `0.5954`
- Test Accuracy: `0.6121`

- **Gradient Boosting**
- Train Accuracy: `0.6958`
- Test Accuracy: `0.6533`

- **AdaBoost**
- Train Accuracy: `0.6555`
- Test Accuracy: `0.6414`

- **CatBoost**
- Train Accuracy: `0.8320`
- Test Accuracy: `0.6356`

- **LightGBM**
- Train Accuracy: `0.8444`
- Test Accuracy: `0.6353`

- **XGBoost**
- Train Accuracy: `0.9137`
- Test Accuracy: `0.6146`

- **Extra Trees**
- Train Accuracy: `1.0000`
- Test Accuracy: `0.6097`

- **K-Nearest Neighbors**
- Train Accuracy: `0.7518`
- Test Accuracy: `0.6146`

- **Decision Tree**
- Train Accuracy: `1.0000`
- Test Accuracy: `0.6015`

### Non-PCA and Winsorized Data
- **Random Forest**
- Train Accuracy: `0.7181`
- Test Accuracy: `0.6942`

- **Logistic Regression**
- Train Accuracy: `0.6336`
- Test Accuracy: `0.6386`

- **Gradient Boosting**
- Train Accuracy: `0.7210`
- Test Accuracy: `0.6904`

- **AdaBoost**
- Train Accuracy: `0.6882`
- Test Accuracy: `0.6846`

- **CatBoost**
- Train Accuracy: `0.8590`
- Test Accuracy: `0.6636`

- **LightGBM**
- Train Accuracy: `0.8685`
- Test Accuracy: `0.6656`

- **XGBoost**
- Train Accuracy: `0.9343`
- Test Accuracy: `0.6678`

- **Extra Trees**
- Train Accuracy: `1.0000`
- Test Accuracy: `0.6394`

- **K-Nearest Neighbors**
- Train Accuracy: `0.7780`
- Test Accuracy: `0.6532`

- **Decision Tree**
- Train Accuracy: `1.0000`
- Test Accuracy: `0.6380`

### PCA and Outliers Deleted Data
- **Random Forest**
- Train Accuracy: `0.7075`
- Test Accuracy: `0.6219`

- **Logistic Regression**
- Train Accuracy: `0.6039`
- Test Accuracy: `0.6082`

- **Gradient Boosting**
- Train Accuracy: `0.6958`
- Test Accuracy: `0.6332`

- **AdaBoost**
- Train Accuracy: `0.6489`
- Test Accuracy: `0.6158`

- **CatBoost**
- Train Accuracy: `0.8112`
- Test Accuracy: `0.6262`

- **LightGBM**
- Train Accuracy: `0.8199`
- Test Accuracy: `0.6219`

- **XGBoost**
- Train Accuracy: `0.8997`
- Test Accuracy: `0.6112`

- **Extra Trees**
- Train Accuracy: `1.0000`
- Test Accuracy: `0.6096`

- **K-Nearest Neighbors**
- Train Accuracy: `0.7460`
- Test Accuracy: `0.6104`

- **Decision Tree**
- Train Accuracy: `1.0000`
- Test Accuracy: `0.6018`

### PCA and Winsorized Data
- **Random Forest**
- Train Accuracy: `0.7172`
- Test Accuracy: `0.6752`

- **Logistic Regression**
- Train Accuracy: `0.6039`
- Test Accuracy: `0.6096`

- **Gradient Boosting**
- Train Accuracy: `0.6983`
- Test Accuracy: `0.6806`

- **AdaBoost**
- Train Accuracy: `0.6688`
- Test Accuracy: `0.6702`

- **CatBoost**
- Train Accuracy: `0.8554`
- Test Accuracy: `0.6888`

- **LightGBM**
- Train Accuracy: `0.8706`
- Test Accuracy: `0.6938`

- **XGBoost**
- Train Accuracy: `0.9093`
- Test Accuracy: `0.6612`

- **Extra Trees**
- Train Accuracy: `1.0000`
- Test Accuracy: `0.5996`

- **K-Nearest Neighbors**
- Train Accuracy: `0.7567`
- Test Accuracy: `0.6260`

- **Decision Tree**
- Train Accuracy: `1.0000`
- Test Accuracy: `0.6040`

![result 1](https://github.com/adi271001/ML-Crate/blob/e-commerce-shipping/E-commerce%20Shipping%20Data%20Analysis/Images/__results___34_1.png?raw=true)
![result 1](https://github.com/adi271001/ML-Crate/blob/e-commerce-shipping/E-commerce%20Shipping%20Data%20Analysis/Images/__results___34_11.png?raw=true)
![result 1](https://github.com/adi271001/ML-Crate/blob/e-commerce-shipping/E-commerce%20Shipping%20Data%20Analysis/Images/__results___34_13.png?raw=true)
![result 1](https://github.com/adi271001/ML-Crate/blob/e-commerce-shipping/E-commerce%20Shipping%20Data%20Analysis/Images/__results___34_15.png?raw=true)
![result 1](https://github.com/adi271001/ML-Crate/blob/e-commerce-shipping/E-commerce%20Shipping%20Data%20Analysis/Images/__results___34_17.png?raw=true)
![result 1](https://github.com/adi271001/ML-Crate/blob/e-commerce-shipping/E-commerce%20Shipping%20Data%20Analysis/Images/__results___34_19.png?raw=true)
![result 1](https://github.com/adi271001/ML-Crate/blob/e-commerce-shipping/E-commerce%20Shipping%20Data%20Analysis/Images/__results___34_21.png?raw=true)
![result 1](https://github.com/adi271001/ML-Crate/blob/e-commerce-shipping/E-commerce%20Shipping%20Data%20Analysis/Images/__results___34_23.png?raw=true)
![result 1](https://github.com/adi271001/ML-Crate/blob/e-commerce-shipping/E-commerce%20Shipping%20Data%20Analysis/Images/__results___34_25.png?raw=true)
![result 1](https://github.com/adi271001/ML-Crate/blob/e-commerce-shipping/E-commerce%20Shipping%20Data%20Analysis/Images/__results___34_27.png?raw=true)
![result 1](https://github.com/adi271001/ML-Crate/blob/e-commerce-shipping/E-commerce%20Shipping%20Data%20Analysis/Images/__results___34_29.png?raw=true)
![result 1](https://github.com/adi271001/ML-Crate/blob/e-commerce-shipping/E-commerce%20Shipping%20Data%20Analysis/Images/__results___34_3.png?raw=true)
![result 1](https://github.com/adi271001/ML-Crate/blob/e-commerce-shipping/E-commerce%20Shipping%20Data%20Analysis/Images/__results___34_31.png?raw=true)
![result 1](https://github.com/adi271001/ML-Crate/blob/e-commerce-shipping/E-commerce%20Shipping%20Data%20Analysis/Images/__results___34_33.png?raw=true)
![result 1](https://github.com/adi271001/ML-Crate/blob/e-commerce-shipping/E-commerce%20Shipping%20Data%20Analysis/Images/__results___34_35.png?raw=true)
![result 1](https://github.com/adi271001/ML-Crate/blob/e-commerce-shipping/E-commerce%20Shipping%20Data%20Analysis/Images/__results___34_37.png?raw=true)
![result 1](https://github.com/adi271001/ML-Crate/blob/e-commerce-shipping/E-commerce%20Shipping%20Data%20Analysis/Images/__results___34_39.png?raw=true)
![result 1](https://github.com/adi271001/ML-Crate/blob/e-commerce-shipping/E-commerce%20Shipping%20Data%20Analysis/Images/__results___34_51.png?raw=true)


## Conclusion
The evaluation of various machine learning models on different preprocessed datasets revealed that models like CatBoost and XGBoost consistently achieved higher accuracies across multiple preprocessing scenarios. These models showed robustness in handling different data transformations, with XGBoost showing superior performance overall. On the other hand, models such as Extra Trees and Decision Tree exhibited high training accuracies but struggled with generalization, as reflected in their lower test accuracies. This indicates a tendency towards overfitting. The choice of preprocessing and model selection is crucial for improving predictive performance and reducing overfitting.

## Signature
Aditya D
* Github: [https://www.github.com/adi271001](https://www.github.com/adi271001)
* LinkedIn: [https://www.linkedin.com/in/aditya-d-23453a179/](https://www.linkedin.com/in/aditya-d-23453a179/)
* Topmate: [https://topmate.io/aditya_d/](https://topmate.io/aditya_d/)
* Twitter: [https://x.com/ADITYAD29257528](https://x.com/ADITYAD29257528)

0 comments on commit 2cc8bdb

Please sign in to comment.