Notes for Tensorflow Certificate Exam Preparation

Skills checklist

Source: https://www.tensorflow.org/site-assets/downloads/marketing/cert/TF_Certificate_Candidate_Handbook.pdf

1. Build and Train Neural Network Models Using TensorFlow 2.x

Understanding the foundational principles of machine learning (ML) and deep learning (DL) using TensorFlow 2.x.

• Use TensorFlow 2.x. ✔️
• Build, compile and train machine learning (ML) models using TensorFlow. ✔️
• Preprocess data to get it ready for use in a model. ✔️
• Use models to predict results. ✔️
• Build sequential models with multiple layers. ✔️
• Build and train models for binary classification.
• Build and train models for multi-class categorization. ✔️
• Plot loss and accuracy of a trained model. ✔️
• Identify strategies to prevent overfitting, including augmentation and dropout. ✔️
• Use pretrained models (transfer learning). ✔️
• Extract features from pre-trained models. ✔️
• Ensure that inputs to a model are in the correct shape. ✔️
• Ensure that you can match test data to the input shape of a neural network. ✔️
• Ensure you can match output data of a neural network to specified input shape for test data. ✔️
• Understand batch loading of data. ✔️
• Use callbacks to trigger the end of training cycles. ✔️
• Use datasets from different sources. ✔️
• Use datasets in different formats, including json(✔️) and csv (✔️).
• Use datasets from tf.data.datasets. ✔️

2. Image Classification

Building image recognition and object detection models with deep neural networks and convolutional neural networks using TensorFlow 2.x.

• Define Convolutional neural networks with Conv2D and pooling layers. ✔️
• Build and train models to process real-world image datasets. ✔️
• Understand how to use convolutions to improve your neural network. ✔️
• Use real-world images in different shapes and sizes. ✔️
• Use image augmentation to prevent overfitting. ✔️
• Use ImageDataGenerator. ✔️
• Understand how ImageDataGenerator labels images based on the directory structure. ✔️

3. Natural Language Processing (NLP)

Using neural networks to solve natural language processing problems using TensorFlow.

• Build natural language processing systems using TensorFlow. ✔️
• Prepare text to use in TensorFlow models. ✔️
• Build models that identify the category of a piece of text using binary categorization ✔️
• Build models that identify the category of a piece of text using multi-class categorization ✔️
• Use word embeddings in your TensorFlow model. ✔️
• Use LSTMs in your model to classify text for either binary or multi-class categorization. ✔️
• Add RNN and GRU layers to your model. ✔️
• Use RNNS, LSTMs, GRUs and CNNs in models that work with text. ✔️
• Train LSTMs on existing text to generate text (such as songs and poetry)

4. Time Series, Sequences and Predictions

Understanding how to solve time series and forecasting problems in TensorFlow.

• Train, tune and use time series, sequence and prediction models. ✔️
• Prepare data for time series learning. ✔️
• Understand Mean Average Error (MAE) and how it can be used to evaluate accuracy of sequence models. ✔️
• Use RNNs and CNNs for time series, sequence and forecasting models. ✔️
• Identify when to use trailing versus centred windows. ✔️
• Use TensorFlow for forecasting. ✔️
• Prepare features and labels. ✔️
• Identify and compensate for sequence bias. ✔️
• Adjust the learning rate dynamically in time series, sequence and prediction models. ✔️