Guest
network classification in TensorFlow\ 8:31 Introducing more classification evaluation methods 14:36 Finding the accuracy of our classification model 18:54 Creating our first confusion matrix (to see where our Mdl is getting confused) 27:22 Making our confusion matrix prettier Multi-class classification part 41:23 1 Putting things together with Getting the data 52:00 2 Becoming one with the data 59:08 3 Building a multi-class classification model 1:14:47 4 Improving performance with normalisation 1:27:30 5 Comparing normalised and non-normalised data 1:31:44 6 Finding the ideal learning rate 1:42:22 7 Evaluating our model 1:55:39 8 Creating a confusion matrix 2:00:05 9 Visualising random Mdl predictions 2:10:48 What patterns is our Mdl learning Vision and Convolutional Neural Networks in TensorFlow\ 2:26:21 Introduction to Computer Vision with TensorFlow 2:35:57 Introduction to Convolutional Neural Networks (CNNs) with TensorFlow 2:43:57 Downloading an image dataset for our first Food Vision model 2:52:24 Becoming One With Data 2:57:30 Becoming One With Data Part 2 3:09:56 Becoming One With Data Part 3 3:14:19 Building an end to end CNN Model 3:32:37 Using a GPU to run our CNN Mdl 5x faster 3:41:54 Trying a non-CNN Mdl on our image data 3:50:45 Improving our non-CNN Mdl by adding more layers Breaking our CNN Mdl down part 4:00:38 1 Becoming one with the data 4:09:41 2 Preparing to load our data 4:21:27 3 Loading our data with ImageDataGenerator 4:31:22 4 Building a baseline CNN model 4:39:25 5 Looking inside a Conv2D layer 4:54:46 6 Compiling and fitting our baseline CNN 5:02:01 7 Evaluating our CNN’s training curves 5:13:46 8 Reducing overfitting with Max Pooling 5:27:27 9 Reducing overfitting with data augmentation 5:34:19 10 Visualizing our augmented data 5:49:23 11 Training a CNN Mdl on augmented data 5:58:12 12 Discovering the power of shuffling data 6:08:14 13 Exploring options to improve our model 6:13:36 Downloading a custom image to make predictions on 6:18:31 Writing a helper function to load and preprocessing custom images 6:28:32 Making a prediction on a custom image with our trained CNN Multi-class CNN’s part 6:38:40 1 Becoming one with the data 6:53:40 2 Preparing our data (turning it into tensors) 7:00:18 3 Building a multi-class CNN model 7:07:43 4 Fitting a multi-class CNN Mdl to the data 7:13:46 5 Evaluating our multi-class CNN model 7:18:37 6 Trying to fix overfitting by removing layers 7:30:57 7 Trying to fix overfitting with data augmentation 7:42:43 8 Things you could do to improve your CNN model 7:47:07 9 Making predictions with our Mdl on custom images 7:56:29 Saving and loading our trained CNN model Learning in TensorFlow Part 1 Feature extraction\ 8:02:51 What is and why use transfer learning 8:13:03 Downloading and preparing data for our first transfer learning model 8:27:43 Introducing Callbacks in TensorFlow and making a callback to track our models 8:37:45 Exploring the TensorFlow Hub website for pretrained models 8:47:36 Building and compiling a TensorFlow Hub feature extraction model 9:01:36 Blowing our previous models out of the water with transfer learning 9:10:50 Plotting the loss curves of our ResNet feature extraction model 9:18:26 Building and training a pre-trained EfficientNet Mdl on our data 9:28:08 Different Types of Transfer Learning 9:39:49 Comparing Our Model’s Results 9:55:06 Exercise Imposter Syndrome Learning in TensorFlow Part 2 Fine tuning\ 9:58:02 Introduction to Transfer Learning in TensorFlow Part 2 Fine-tuning 10:04:18 Importing a script full of helper functions (and saving lots of space) 10:11:53 Downloading and turning our images into a TensorFlow BatchDataset 10:27:32 Discussing the four (actually five) modelling experiments we’re running 10:29:48 Comparing the TensorFlow Keras Sequential API versus the Functional API 10:32:22 Creating our first Mdl with the TensorFlow Keras Functional API 10:44:01 Compiling and fitting our first Functional API model 10:54:55 Getting a feature vector from our trained model 11:08:34 Drilling into the concept of a feature vector (a learned representation) 11:12:17 Downloading and preparing the data for Mdl 1 (1 percent of training data) 11:22:09 Building a data augmentation layer to use inside our model 11:34:16 Visualizing what happens when images pass through our data augmentation layer 11:45:11 Building Mdl 1 (with a data augmentation layer and 1% of training data)
