Gain insights into using PyTorch for image classification and detection, delve into model implementation, and explore deployment on edge devices with ONNX and OpenVINO.

Image Classification and Object Detection using CNNs-01.png

pytorch.tar.gz

Basics_Of_CNN-ex1

Basics_Of_CNN-ex2

Basics_Of_CNN-vgg16

Basics_Of_CNN-inceptionv1

Basics_Of_CNN-residuals

Basics_Of_CNN-depthwise

Train-data

Train-training

Train-trainingFT

Train-transfer

deploy-onnx

deploy-openvino

deploy-comparison

prepare_dataset_openimages

train_yolov7

inference_yolov7

onnx_yolov7

Image classification and object detection have gained widespread use in recent years. Content categorization and monitoring, disease diagnosis from medical images, identifying terrain in satellite images, and detecting road elements for self-driving cars are classification problems at their core. PyTorch is a popular framework for these tasks—offering a useful mix of user-friendliness, deep learning functionalities, customization, and optimization.

In this course, you will cover the fundamentals of classification and object detection models and apply them to actual datasets using PyTorch. You’ll learn popular architectures and how to implement and fine-tune them for better results. Finally, you’ll learn to convert models to ONNX and OpenVINO to deploy in edge devices.

By the end of this course, you will have acquired the necessary skills to be able to use PyTorch for image identification and object detection in real-world applications

Using PyTorch for Image Classification and Object Detection

If we apply dropout with a 0.3 rate to a layer with 10 neurons, only 3 of them will be selected randomly to carry the signal and will be updated with their weights in each iteration.

Residual blocks, first discovered in ResNet, are especially good for vanishing gradient problems.

Weights are individual for the kernels, whereas only one bias exists for one filter.

While an exploding gradient is about having too small weights, the vanishing gradient is about having too large weights, and in both cases, we have an underfitting problem.

**(Select all that apply.)** Which statement is _true_?

(Select all that apply.) Which statement is true?

Suppose we have a 24x24x16$24 feature map, apply a 5x5x32 convolution on it, and obtain a 20x20x32 output feature map. What is the FLOP of this convolution?

Suppose we have a 24x24x16$24 feature map, apply a 5x5x32 convolution on it, and obtain a 20x20x32 output feature map. What is the FLOP of this convolution?

The width of a network means the input image width.

The depth of a network means the number of layers in the neural network.

While MobileNet focuses on compressing models with individual width and resolution multipliers, EfficientNet aims to compress the whole model with a common factor.

The width multiplier is used to scale the input image's size.

We have a 7x7x3 input image and send it to a 3x3x128 depth-wise convolutional layer. With stride = 1 and padding = 1, what is the output feature map that we obtain?

We have a 7x7x3 input image and send it to a 3x3x128 depth-wise convolutional layer. With stride = 1 and padding = 1, what is the output feature map that we obtain?

Test your knowledge about image classification basics and its popular architectures.



Before We Start

Basics of Convolutional Neural Networks

Popular Neural Network Architectures for Image Classification

Using PyTorch for Image Classification

Model Deployment

Basics of Object Detection

Two-Stage Object Detection Architectures

One-Stage Object Detection Architectures

YOLOv7 Model Train and Inference on Edge

Conclusion

Appendix

Quiz on the Basics of CNNs and their Architectures