srujanponnur / erroneous_data_detection

Erroneous_data_detection

Python3 - Version 3.7

Libraries Used for the project are:

1. pandas
2. tqdm
3. numpy
4. datetime
5. sklearn
6. imblearn
7. scipy
8. tensorflow
9. os

Project Flow:

1. All the preprocessing steps like one hot encoding, Normalization and feature engineering are performed in "PreProcessing.ipynb". This will generate two csv files, Train_updated.csv and Test_updated.csv
2. Now all the model (Decision Tree, Random Forest, Logistic Regression) training and hyperparameter tuning is performed in the file, "ModelRandomSearch.ipynb" file, this also generates a prediction.csv file which was submitted as part of the competition.
3. A ANN Model is trained and later tuned in the file, "NeuralNetwork.py".

Commands to Run:

1. After Installing all the libraries mentioned in the above list.
2. Run the Jupiter Notebooks mentioned above in the respective order.
3. Also, Run the python file NeuralNetwork.py using python3 NeuralNetwork.py