Notebook for learning deep learning
- Learn concepts involved in Deep Learning
- Try multiple classification and recognition problems
- Use tf, keras
- Learn Models, Layers, Convolution, Normalization, Pooling, etc.
- Learn mathematical concepts used in these layers.
- Tried image classification example
- Saw video on Types of layers of keras. Convolution, Pooling, Normalization and
- Tried Text classification.
- Reading about relu, CNN and other concepts
- Understanding keras.layers.Dense: CNN layer op = activation(in, kernel) + bias
- Units: size of output
- Activation: function to apply linear or non-linear
- Initializer: kernel and bias initial values.
- All keras layers: https://keras.io/layers/core/ I will focus on only some of them: Dense, Flatten, Dropouts
- Learnt concept of epoch and its effect and training.
- Tried fuel efficiency regression example.
- Take other example
- Create script to try combinaations of units, epoch and activation functions. tabular result of each ones accuracy.
- Understand What happens when we apply keras layer.
- https://www.tensorflow.org/tutorials/keras/classification
- https://medium.com/tensorflow/standardizing-on-keras-guidance-on-high-level-apis-in-tensorflow-2-0-bad2b04c819a
- https://www.tensorflow.org/guide/keras#build_a_simple_model
- https://keras.io/layers/core/
- https://machinelearningmastery.com/better-deep-learning/
- https://machinelearningmastery.com/rectified-linear-activation-function-for-deep-learning-neural-networks/
- https://github.com/Kulbear/deep-learning-nano-foundation/wiki/ReLU-and-Softmax-Activation-Functions
- https://deepai.org/machine-learning-glossary-and-terms/epoch