Getting Started with Image Classification with PyTorch

Description

In this course, you will learn :

• A basic overview of the PyTorch Image Model
• The ability to fine-tune custom image classification models
• A working knowledge of deploying models as REST API
• A familiarity with converting PyTorch models into ONNX format

Syllabus :

1. Introduction

• Overview of Image Classification
• Image Classification Techniques
• Image Classification Metrics
• PyTorch Image Model Framework

2. Basic Concepts

• Create Model
• List and Search Supported Models
• Dataset Format
• Load and Preprocess Images
• Prediction Using Pre-trained Resnet50
• Prediction Using Pre-trained EfficientNet

3. Augmentation

• Training a Neural Network with Augmentation
• Mixup and CutmixRandAugment
• Random Erase
• AutoAugment
• Random Resized Crop and Interpolation

4. Loss

• Loss Functions
• Asymmetric Loss
• Jensen-Shannon Divergence and Cross-Entropy Loss

5. Training

• Training Workflow
• ResNet50
• EfficientNet
• Fine-tuning Custom Model
• Training with an EMA (Exponential Moving Average)

6. Model Conversion

• Serving a Pytorch Model
• Convert PyTorch Model into an ONNX Model
• Serving an ONNX Model
• Convert ONNX into a Tensorflow Model
• Serving a TensorFlow Model
• Convert a TensorFlow Model into a TFLite Model

7. Deployment

• Overview of FastAPI
• HTTP Methods
• REST API Integration

Show more