Prediction on the famous MNIST dataset.The MNIST database of handwritten digits, It is a subset of a larger set available from NIST. The digits have been size-normalized and centered in a fixed-size image. Predicts the number from the image of numbers provided.
- Image Data preprocessing.
- Data Visualization and Exploratory Data Analysis.
- Image Data Normalization and Scaling.
- Training using Sequential Model with Conv2D,MaxPool2D,Flatten and Dense Layers , with output layer having Softmax activation.
- Compiled using AdamOptmizer as optmizer and categorical_crossentropy as the loss function as its multi class classification problem.
- Solving Overfitting issues using EarlyStoppings Callbacks and Dropout Layers in Network.
- Hyperparameter Tuning the Algorithms yielding best results.
- Testing the model on custom reallife Image data.
- Buitin Dataset from Keras
- TensorFlow 2
- Keras
- Scikit-Learn
- Seaborn
- Matplotlib
- Pandas
- Numpy