This repository is created for general understanding coding and its related projects on machine learning and deep learning topics. These projects involve working on machine learning and deep learning concepts using Keras, PyTorch, TensorFlow libraries using Python Code. This repository is updated on regular basis dependent upon project studies. Some of the projects are undertaken from Deep Learning with Python by Jason Brownlee, from Coursera guided projects, and Kaggle.com.
-
The dataset used for this section is on Pima Indians Diabetes
-
Evaluating the performance of deep learning models
-
Using Keras models integrated with scikit-learn for general machine learning and deep learning concepts
-
Tuning deep learning algorithms, optimizers and parameters using GridSearchCV
Data Exploration, Data Mining Projects:
- Exploratory Data Analysis [Project Folder], [Project Description],[Notebook],[Dataset]
- Exploratory Data Analysis Using Seaborn - Breast Cancer Diagnosis [Project Folder], [Project Description],[Notebook],[Dataset], [Notes]
Machine Learning Projects:
- Statistical Data Visualization with Seaborn - Breast Cancer Diagnosis [Project Folder], [Project Description],[Notebook],[Dataset]
- Credit Card Fraud Detection [Project Folder], [Project Description],[Notebook]
- Mercedes-Benz Greener Manufacturing [Project Folder], [Project Description],[Notebook],[Training Dataset],[Testing Dataset]
Deep Learning Projects
- Basic Image Classification with TensorFlow [Project Folder], [Project Description],[Notebook]
- Image Classification with CNNs using Keras [Project Folder], [Project Description],[Notebook 1],[Notebook 2]
- Facial Expression Recognition with Keras [Project Folder], [Project Description],[Notebook 1],[Notebook 2]
- Image Noise Reduction with Auto-encoders using TensorFlow [Project Folder], [Project Description],[Notebook]
- Dimensionality Reduction using an Autoencoder in Python [Project Folder], [Project Description],[Notebook]
- Anomaly Detection in Time Series Data with Keras [Project Folder], [Project Description], [Notebook V1], [Notebook V2]
- V1 - Underfit Model - Dropout Regularization of 0.2 and Threshold of 0.65
- V2 - Fit Model - Dropout Regularization of 0.02 and Threshold of 1
- Generate Synthetic Images with DCGANs in Keras [Project Folder], [Project Description], [Notebook V1], [Notebook V2], [Images V1], [Images V2], [plot_utils.py]
- Text Classification Using Word2Vec and LSTM on Keras [Project Folder], [Project Description],[Notebook], [Dataset]