Deep-Learning-With-TensorFlow

TensorFlow Developer Certificate by ZeroToMastery

Lesson 0: Deep Learning Fundamentals

1. What is Deep Learning?
2. Why use Machine Leaning (Deep Learning)?
3. When to used Deep Learning or Not
4. Tensor and TensorFlow
5. Course Contents:
   • TensorFlow fundamentals
   • Preprocessing data (turning into tensors)
   • Building and using pretrained deep learning models
   • Fitting a model to the data (learning pattern)
   • Making prediction with a model(using pattern)
   • Evaluating model predictions, Saving and loading models
   • Using trained model to make prediciton on custom data.

Lesson 1: Tensor Fundamentals

Click here to read the documentation.

Topics covered are: Creating tensor, Random and Shuffling tensor, Tensor from numpy, Getting tensor attributes, Indexing and Expanding tensor, Tensor operation and manipulation, Casting tensor, Aggregating tensor, squeezing tensor, hot encoding, Tensor Vs Numpy, Accessing GPU runtime.

Lesson 2: Neural Network Regression with TensorFlow

Click here to read the documentation.

Topics covered are: Building hands-on Neural Network Regression model; Neural Network architectures, steps in modelling with tensorflow, Improving a model - by increasing epochs, changing optimizer, changing learning rate, adding hidden layers; Evaluating a model - visualizing model, visualizing model's prediction, Evaluating metrics, traking experiment models, save and load model.

Lesson 3: Neural Network Classification with TensorFlow.

Click here to read the documentation.

Topics covered are: Building hands-on Neural Network Classification model; Neural Network architectures, Steps in modelling with tensorflow, Improving the model, Non linear activations (Sigmoid, ReLU), Replicating non-linear activation function from scratch, Evaluating on testing data, Visualizing history - loss and accuracy curve, Find the best learning rate, Evaluation methods - accuracy, precision, recall, F1-score, confusion matrix; Anatomy of confusion matrix, Multiclass Neural Network Classification.

Lesson 4: Convolutional Neural Network with TensorFlow

Click here to read the documentation.

Topics covered are: Computer vision applications, Building hands-on CNN mode, Convolutional Neural Network architectures; Conv2D and MaxPool2D explanation, Food 101 datasets, Walking through directory of data using os module, Visualizing random food images programatically, Normalize the data (divide by 255.), Building CNN model, Non-CNN model for image data, Breaking down CNN model, Batches creation, Inducing overfitting, Data augmentation,