Overview

This course aims to teach the basics of computer vision with deep learning using TensorFlow. By the end of the course, learners will be able to build neural networks, convolutional neural networks, and advanced models for tasks like car price prediction, malaria diagnosis, and human emotions detection. The course covers topics such as tensors, variables, building models, evaluating classification models, improving model performance, data augmentation, advanced TensorFlow topics, Tensorboard integration, MLOps with Weights and Biases, modern convolutional neural networks, transfer learning, understanding the blackbox, and transformers in vision. The intended audience for this course is individuals interested in computer vision, deep learning, and TensorFlow, with a focus on practical implementation and hands-on experience.

Syllabus

⌨️ Welcome
⌨️ Prerequisite
⌨️ What we shall Learn
⌨️ Basics
⌨️ Initialization and Casting
⌨️ Indexing
⌨️ Maths Operations
⌨️ Linear Algebra Operations
⌨️ Common TensorFlow Functions
⌨️ Ragged Tensors
⌨️ Sparse Tensors
⌨️ String Tensors
⌨️ Variables
⌨️ Task Understanding
⌨️ Data Preparation
⌨️ Linear Regression Model
⌨️ Error Sanctioning
⌨️ Training and Optimization
⌨️ Performance Measurement
⌨️ Validation and Testing
⌨️ Corrective Measures
⌨️ Task Understanding
⌨️ Data Preparation
⌨️ Data Visualization
⌨️ Data Processing
⌨️ How and Why ConvNets Work
⌨️ Building Convnets with TensorFlow
⌨️ Binary Crossentropy Loss
⌨️ Training Convnets
⌨️ Model Evaluation and Testing
⌨️ Loading and Saving Models to Google Drive
⌨️ Functional API
⌨️ Model Subclassing
⌨️ Custom Layers
⌨️ Precision, Recall and Accuracy
⌨️ Confusion Matrix
⌨️ ROC Plots
⌨️ TensorFlow Callbacks
⌨️ Learning Rate Scheduling
⌨️ Model Checkpointing
⌨️ Mitigating Overfitting and Underfitting
⌨️ Augmentation with tf.image and Keras Layers
⌨️ Mixup Augmentation
⌨️ Cutmix Augmentation
⌨️ Data Augmentation with Albumentations
⌨️ Custom Loss and Metrics
⌨️ Eager and Graph Modes
⌨️ Custom Training Loops
⌨️ Data Logging
⌨️ View Model Graphs
⌨️ Hyperparameter Tuning
⌨️ Profiling and Visualizations
⌨️ Experiment Tracking
⌨️ Hyperparameter Tuning
⌨️ Dataset Versioning
⌨️ Model Versioning
⌨️ Data Preparation
⌨️ Modeling and Training
⌨️ Data Augmentation
⌨️ TensorFlow Records
⌨️ AlexNet
⌨️ VGGNet
⌨️ ResNet
⌨️ Coding ResNet from Scratch
⌨️ MobileNet
⌨️ EfficientNet
⌨️ Feature Extraction
⌨️ Finetuning
⌨️ Visualizing Intermediate Layers
⌨️ Gradcam method
⌨️ Understanding ViTs
⌨️ Building ViTs from Scratch
⌨️ FineTuning Huggingface ViT
⌨️ Model Evaluation with Wandb
⌨️ Converting TensorFlow Model to Onnx format
⌨️ Understanding Quantization
⌨️ Practical Quantization of Onnx Model
⌨️ Quantization Aware Training
⌨️ Conversion to TensorFlow Lite
⌨️ How APIs work
⌨️ Building an API with FastAPI
⌨️ Deploying API to the Cloud
⌨️ Load Testing with Locust
⌨️ Introduction to Object Detection
⌨️ Understanding YOLO Algorithm
⌨️ Dataset Preparation
⌨️ YOLO Loss
⌨️ Data Augmentation
⌨️ Testing
⌨️ Introduction to Image Generation
⌨️ Understanding Variational Autoencoders
⌨️ VAE Training and Digit Generation
⌨️ Latent Space Visualization
⌨️ How GANs work
⌨️ The GAN Loss
⌨️ Improving GAN Training
⌨️ Face Generation with GANs
⌨️ What's Next