N.B.: Please don't use the assignment and quiz solution at first time, only use when you get stuck really bad situation. Try to solve the problem by yourself.
Discover the tools software developers use to build scalable AI-powered algorithms in TensorFlow, a popular open-source machine learning framework.
In this four-course Specialization, you’ll explore exciting opportunities for AI applications. Begin by developing an understanding of how to build and train neural networks. Improve a network’s performance using convolutions as you train it to identify real-world images. You’ll teach machines to understand, analyze, and respond to human speech with natural language processing systems. Learn to process text, represent sentences as vectors, and input data to a neural network. You’ll even train an AI to create original poetry! - Source
- Huber Loss: https://en.wikipedia.org/wiki/Huber_loss
Introduction to TensorFlow for Artificial Intelligence, Machine Learning, and Deep Learning - Coursera - GitHub - Certificate
- Week 1
- Lesson Topic: Introduction to ML, Introduction to TensorFlow
- Quiz: Basic ML and TensorFlow
- Optional: Housing prices
- Week 2
- Lesson Topic: Introduction to Computer vision, Data processing, MNIST data, Neural Network, Callback function implementation
- Quiz: Data processing, callbacks
- Optional: Handwriting recognition
- Week 3
- Lesson Topic: Convolutions, Pooling, ConvNet, Filters, Padding
- Quiz: Convolution Neural Network
- Optional: MNIST with convolutions
- Week 4
- Lesson Topic: ImageGenerator, Designing of neural network, ConvNet train, Impact of compressing images
- Quiz: Image Generator
- Optional: Handling complex images
Convolutional Neural Networks in TensorFlow - Coursera - GitHub - Certificate
- Week 1
- Lesson Topic: Cats vs Dogs datasets, Visualizing the effect of the convolutions, accuracy and loss
- Quiz: Convolutions
- Optional: Cats Vs Dogs
- Week 2
- Lesson Topic: Image Augmemtation, Overfitting
- Quiz: Image Augmentation
- Optional: Cats vs Dogs using augmemtation
- Week 3
- Lesson Topic: Transfer learning, Dropouts
- Quiz: Transfer learning and Dropouts
- Optional: Horses vs. humans using Transfer Learning
- Week 4
- Lesson Topic: Multiclass Classifications
- Quiz: Multiclass classifications
- Optional: Multi-class classifier
Natural Language Processing in TensorFlow - Coursera - GitHub - Certificate
- Week 1
- Lesson Topic: Tokenizer (tf.keras.preprocessing.text), pad_sequence (tf.keras.preprocessing.sequence)
- Quiz: Tokenizer
- Week 2
- Lesson Topic: IMDB Datasets, Tesorflow Datasets, Subwords Text Encoder
- Quiz: IMDB Datasets
- Optional: BBC news archive
- Week 3
- Lesson Topic: Sequence Modeling, LSTM, Accuracy and Loss, Convolutional Network
- Quiz: LSTM, GRU, Conv1D and NLP
- Optional: Single Layer LSTM, Multi Layer LSTM, 1D Convolutional Layer, Bidirectional LSTM, GRU, Exploring overfitting in NLP
- Week 4
- Lesson Topic: Sequence Models and Literature
- Quiz: Tokenizer, Conv1D, Datasets, Padding, Prediction, LSTM
- Optional: Shakespeare
Sequences, Time Series and Prediction - Coursera - GitHub - Certificate
- Week 1
- Lesson Topic: Introduction of time series, Errors, MSE, RMSE, MAE, MAPE, Forecasting
- Quiz: Different Types of Time Series with Example
- Optional: Create and predict synthetic data
- Week 2
- Lesson Topic: Preparing Features and Labels, Sequence Bias, Feeding Windowed Dataset, Single Layer NN, Time Windows, Prediction, DNN Training-Tuning-Prediction, DNN
- Quiz: Time Windows, Types of Error, Callbacks, Learning Rate
- Optional: Predict with a DNN
- Week 3
- Lesson Topic: RNN, Lambda Layers, Huber Loss, LSTM
- Quiz: RNN, Loss, LSTM
- Optional: Mean Absolute Error
- Week 4
- Lesson Topic: Convolutions, Bi-directional LSTMs, Batch Size, Real Data with Train-Tune-Prediction
- Quiz: Convolution, CSV File, Error, Time Series
- Optional: Sunspots