Skip to content

Deep Learning notes and practical implementation with Tensorflow and keras. Text Analytics and practical application implementation with NLTK, Spacy and Gensim.

Notifications You must be signed in to change notification settings

nursnaaz/Deeplearning-and-NLP

Repository files navigation

Deeplearning

I will update this repository to learn Deeplearning with Tensorflow and Keras

Day - 1: 25-8-2019 We Learnt about

  1. Basic building blocks of Neural Network
  2. Perceptron
  3. Neurons
  4. Hidden Layers
  5. Linear regression with Neural Networks
  6. Logistic regression with Neural Networks
  7. No Linear Activation Function
  8. tanh, step, logit, relu, elu
  9. Back propagation
  10. Vanishing and Exploding gradient descent
  11. Ways to avoid Vanishing and Exploding gradient descent
  12. How to mitigate over fitting ?
  13. Tensorflow - Keras practical

Day - 2: 31-8-2019

  1. Parameter explotion in image recognition
  2. Convolution layer - kernel , filter, Stride, Padding, feature map
  3. Pooling Layer - max, min, average
  4. CNN architecture
  5. Keras implementation
  6. Image recognition in comparison with Basis NN and CNN
  7. Advanced Deep CNN
  8. Pre Trained Models
  9. Transfer Learning - Resnet50
  10. Image Agumentation
  11. Tensor board
  12. Opencv, Yolo3
  13. Sample Hackathon

Day - 3: 01-9-2019

  1. Neural Network so far can only know what was passed in current time
  2. What if we want to remember last output to predict the future if it is a sequence data
  3. Neuron with memory
  4. RNN architecture
  5. Back Propagation Through Time (BPTT)
  6. Problem with BPTT
  7. Vanishing and Exploding gradient descent
  8. Truncated BPTT
  9. LSTM
  10. LSTM Architecture
  11. Keras LSTM implementation

References: https://github.com/omerbsezer/LSTM_RNN_Tutorials_with_Demo#SampleStock https://github.com/fchollet/deep-learning-with-python-notebooks/blob/master/8.1-text-generation-with-lstm.ipynb https://github.com/dipanjanS/nlp_workshop_odsc19 https://github.com/buomsoo-kim/Easy-deep-learning-with-Keras