All of my deep learning and machine learning projects developed using python.
Project 1: Implementation of Multi-layer Perceptron for Image Classification on CIFAR-10 Dataset
Project descriptionImplemented the multi-layer perceptron (MLP) algorithm that trains using Backpropagation on various different feed forward fully connected neural networks model to classify the 10 classes in the CIFAR-10 dataset.
- Different Classifiers such as Perceptron model, MLP classifier, DNN classifier were used on the dataset to achieve the highest accuracy.
- Machine Learning Libraries used were tensorflow, keras, pandas, numpy, matplotlib.
- The highest accuracy achieved was 65% using MLP.
- Tool used: Jupyter notebook, Language: Python.
Project 2: PUBG finish placement Prediction using Neural network and Deep learning Nov 2018 – Dec 2018
Project description- Used various data cleaning and data visualization techniques such as heat maps to train the model with best features.
- Used different regression models such as linear regression, MLP regression and deep learning models.
- Worked on different machine learning libraries such as tensorflow, scikit learn, numpy, pandas, matplot , seaborn etc.
- The final Mean absolute error turned out be 0.0258. The rank secured in the kaggle competition was 462 out of 1500 people.
- Tool used: Jupyter notebook, Language: Python.