Humans are visual learners. The human brain is not well-equipped to process a large corpus of data but is rather good at identifying changes and patterns visually. If you’re given a large paragraph of text describing a scenario and a picture of that scenario for the same amount of time, then you would retain more information from the visual object i.e. the picture. We believe the learning system in engineering should also be fundamentally structured around visualizing the problem domain first, and then, if necessary, be supplemented by texts.
Emphasizing the need for visualization, we envision developing software that would visualize learning algorithms specified by the users and serve as a learning aid to anyone interested in Machine Learning. The algorithms that are visualized:
Linear Regression
Logistic Regression
Neural Network
Linear Support Vector Machine
Non-Linear Support Vector Machine
K-means
Naive Bayes
Decision Tree
Principal Component Analysis
Using Anaconda
conda create -n ENV_NAME python=3.7
Where ENV_NAME is the name of the environment. After creating the environment, activate it using:
conda activate ENV_NAME
Using venv
python3 -m venv ENV_NAME
Where ENV_NAME is the name of the environment. After creating the environment, activate it using:
source ENV_NAME/bin/activate
Install all the requirements from requirements.txt using:
pip install -r requirements.txt
Open Visualizer.ipynb and run all cells.