It gives a good knowledge of Python libraries required for Artificial Intelligence such as -
- Numpy
- Pandas
- MatplotLib and Seaborn
- Pytorch
Along with this, it taught the basic maths behind all this,
- Linear Algebra
- Calculus
Problem 1: The first project is to make a dog classifier in a pre trained classifier.
Problem 2: The second project is to make a image classifier for species of flowers.
The project indentifies the breed of a dog with 84% accuracy and if the image is dog or not with more than 87% accuracy.
The project indentifies the species of a flower with 85% accuracy by a deep learning pre trained model vgg19. To use the project on any new image of a flower, clone the repo to your local machine and run the script "train.py" with the appropriate parameters (will be update in readme soon) under the path "AIwithPython\6. Image Classifier\2. Developing Wrapper for NN\train.py" .
After the model completes training, run "predict.py" to test the model on a brand new image.