This course demystifies convolutional neural network architectures using PyTorch for image classification and object detection.

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

Suppose that we have an image with the dimensions 224x224x3, and we apply the following convolution: `torch.nn.Conv2d(3, 32, 5, stride=1, padding=0)`. What should be the output feature map dimensions?



Suppose that we have an image with the dimensions 224x224x3, and we apply the following convolution: <code>torch.nn.Conv2d(3, 32, 5, stride=1, padding=0)</code>. What should be the output feature map dimensions?

Suppose that we have an image with the dimensions 224x224x3, and we apply the following convolution:
`torch.nn.Conv2d(3, 32, 5, stride=1, padding=“same”)`. What should be the output feature map dimensions?


Suppose that we have an image with the dimensions 224x224x3, and we apply the following convolution:
<code>torch.nn.Conv2d(3, 32, 5, stride=1, padding=“same”)</code>. What should be the output feature map dimensions?

Consider you have 8 output feature maps coming from the last convolutional layer and you have a batch size = 4. Which image's feature map does this code visualize? 

```
output = layer(img)
for feature_map in output:
    layer_viz = output[2, :, :, :]
    layer_viz = layer_viz.data
for i, filter in enumerate(layer_viz):
    plt.subplot(2, 4, i + 1)
    plt.imshow(filter.detach().cpu().numpy(), cmap='gray')
    plt.axis("off")
```

Consider you have 8 output feature maps coming from the last convolutional layer and you have a batch size = 4. Which image’s feature map does this code visualize?
<pre><code>output = layer(img)
for feature_map in output:
 layer_viz = output[2, :, :, :]
 layer_viz = layer_viz.data
for i, filter in enumerate(layer_viz):
 plt.subplot(2, 4, i + 1)
 plt.imshow(filter.detach().cpu().numpy(), cmap='gray')
 plt.axis(&quot;off&quot;)
</code></pre>

What would be the output feature map's dimensions if we apply 2x2 max pooling to an input feature map having the shape 224x224x3?

What would be the output feature map’s dimensions if we apply 2x2 max pooling to an input feature map having the shape 224x224x3?

A filter has kernels inside as much as the input feature map channel size.

Kernels are 2D arrays whereas filters are 3D.

Average pooling is better than maximum pooling for extracting important features.

Kernels in a filter have the same weights and are identical.

Weights are individual for the kernels, whereas there is only 1 bias for 1 layer.

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

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

Test your knowledge of the basics of CNNs.


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