zero-to-mastery-tensorflow-developer-certificate-in-2022-updated-5-2022-2

Learning in TensorFlow Part 2 Fine tuning\ 1:07 Building Mdl 2 (with a data augmentation layer and 10% of training data) 17:44 Creating a ModelCheckpoint to save our model’s weights during training 25:09 Fitting and evaluating Mdl 2 (and saving its weights using ModelCheckpoint) 32:23 Loading and comparing saved weights to our existing trained Mdl 2 39:41 Preparing Mdl 3 (our first fine-tuned model) 1:00:08 Fitting and evaluating Mdl 3 (our first fine-tuned model) 1:07:54 Comparing our model’s results before and after fine-tuning 1:18:21 Downloading and preparing data for our biggest experiment yet (Mdl 4) 1:24:46 Preparing our final modelling experiment (Mdl 4) 1:36:47 Fine-tuning Mdl 4 on 100% of the training data and evaluating its results 1:47:06 Comparing our modelling experiment results in TensorBoard 1:57:52 How to view and delete previous TensorBoard experiments Learning with TensorFlow Part 3 Scaling Up\ 1:59:57 Introduction to Transfer Learning Part 3 Scaling Up 2:06:17 Getting helper functions ready and downloading data to model 2:19:51 Outlining the Mdl we’re going to build and building a ModelCheckpoint callback 2:25:30 Creating a data augmentation layer to use with our model 2:30:09 Creating a headless EfficientNetB0 Mdl with data augmentation built in 2:39:08 Fitting and evaluating our biggest transfer learning Mdl yet 2:47:05 Unfreezing some layers in our base Mdl to prepare for fine-tuning 2:58:34 Fine-tuning our feature extraction Mdl and evaluating its performance 3:06:58 Saving and loading our trained model 3:13:24 Downloading a pretrained Mdl to make and evaluate predictions with 3:19:58 Making predictions with our trained Mdl on 25,250 test samples 3:32:45 Unravelling our test dataset for comparing ground truth labels to predictions 3:38:50 Confirming our model’s predictions are in the same order as the test labels 3:44:07 Creating a confusion matrix for our model’s 101 different classes 3:56:15 Evaluating every individual class in our dataset 4:10:32 Plotting our model’s F1-scores for each separate class 4:18:08 Creating a function to load and prepare images for making predictions 4:30:17 Making predictions on our test images and evaluating them 4:46:23 Discussing the benefits of finding your model’s most wrong predictions 4:52:33 Writing code to uncover our model’s most wrong predictions 5:03:49 Plotting and visualising the samples our Mdl got most wrong 5:14:26 Making predictions on and plotting our own custom images Project 1 Food Vision Big™\ 5:24:16 Making sure we have access to the right GPU for mixed precision training 5:34:33 Getting helper functions ready 5:37:40 Introduction to TensorFlow Datasets (TFDS) 5:49:43 Exploring and becoming one with the data (Food101 from TensorFlow Datasets) 6:05:39 Creating a preprocessing function to prepare our data for modelling 6:21:29 Batching and preparing our datasets (to make them run fast) 6:35:17 Exploring what happens when we batch and prefetch our data 6:42:06 Creating modelling callbacks for our feature extraction model 6:49:21 Turning on mixed precision training with TensorFlow 6:59:26 Creating a feature extraction Mdl capable of using mixed precision training 7:12:09 Checking to see if our Mdl is using mixed precision training layer by layer 7:20:05 Introducing your Milestone Project 1 challenge build a Mdl to beat DeepFood Fundamentals in TensorFlow\ 7:27:53 Introduction to Natural Language Processing (NLP) and Sequence Problems 7:40:45 Example NLP inputs and outputs 7:48:08 The typical architecture of a Recurrent Neural Network (RNN) 7:57:11 Preparing a notebook for our first NLP with TensorFlow project 8:06:04 Becoming one with the data and visualising a text dataset 8:22:45 Splitting data into training and validation sets 8:29:12 Converting text data to numbers using tokenisation and embeddings (overview) 8:38:35 Setting up a TensorFlow TextVectorization layer to convert text to numbers 8:55:45 Mapping the TextVectorization layer to text data and turning it into numbers 9:06:48 Creating an Embedding layer to turn tokenised text into embedding vectors 9:19:16 Discussing the various modelling experiments we’re going to run 9:28:13 Mdl 0 Building a baseline Mdl to try and improve upon 9:37:39 Creating a function to track and evaluate our model’s results 9:49:53 Mdl 1 Building, fitting and evaluating our first deep Mdl on text data 10:10:45 Visualising our model’s learned word embeddings with TensorFlow’s projector tool 10:31:29 High-level overview of Recurrent Neural Networks (RNNs) where to learn more 10:41:03 Mdl 2 Building, fitting and evaluating our first TensorFlow RNN Mdl (LSTM),a GRU-cell powered RNN,a bidirectional RNN model 11:35:51 Discussing the intuition behind Conv1D neural networks for text and sequences 11:55:23 Mdl 5 Building, fitting and evaluating a 1D CNN for text