Gain insights into image classification with PyTorch. Learn about data preprocessing, model training, fine-tuning, and deploying models using ONNX for real-world applications.

Introduction to PyTorch image classification-01.png

wfngenv.tar.gz

wfngenv

terminalenv

PyTorch is a machine learning framework used in a wide array of popular applications, including Tesla’s Autopilot and Pyro, Uber’s probabilistic modeling engine.

This course is an introduction to image classification using PyTorch’s computer vision models for training and tuning your own model. You’ll start with the fundamental concepts of applying machine learning and its applications to image classification before exploring the process of training your AI model. You’ll prepare data for intake by the computer vision model with image pre-processing, set up pipelines for training your model, and fine-tune the variables to improve predictive performance. You’ll finish by deploying the image classification model by converting to ONNX format and serving it via REST API.

By the end of this course, you’ll be able to build and deploy your own image classification models from scratch.

Getting Started with Image Classification with PyTorch

Which statement about the PyTorch Image Model is correct?

Which statement about the PyTorch Image Model is correct?

```
model = timm.new_model('resnet50', pretrained=True)
```

```
model = timm.create_model('resnet50')
```

```
model = timm.instantiate('resnet50', pretrained=True)
```

```
model = timm.create_model('resnet50', pretrained=True)
```

Which statement will instantiate a new pre-trained ResNet50 model?

Which statement will instantiate a new pre-trained ResNet50 model?

**(Select all that apply.)** Which keyword arguments are available for the create_model function?

(Select all that apply.) Which keyword arguments are available for the create_model function?

Which function should we use to search the list of available models in the Pytorch Image Model framework?

Which function should we use to search the list of available models in the Pytorch Image Model framework?

```
densenet_models = timm.list_models('?densenet?')
```

```
densenet_models = timm.list_models('*densenet*')
```

```
densenet_models = timm.list_models('!densenet!')
```

```
densenet_models = timm.list_models('<densenet>')
```

Which is the correct code snippet to search for available `densenet` models via wildcard?

Which is the correct code snippet to search for available <code>densenet</code> models via wildcard?

**(Select all that apply.)** Which `Dataset` classes are supported by the PyTorch Image Model?

(Select all that apply.) Which <code>Dataset</code> classes are supported by the PyTorch Image Model?

```
fruit
├── training
│   ├── banana
│   ├── apple
│   ├── orange
└── testing
    ├── banana
    ├── apple
    └── orange
```

```
fruit
├── train
│   ├── banana
│   ├── apple
│   ├── orange
└── val
    ├── grape
    ├── pear
    └── strawberry
```

```
fruit
├── train
│   ├── banana
│   ├── apple
│   ├── orange
└── val
    ├── banana
    ├── apple
    └── orange
```

```
fruit
└── test
    ├── banana
    ├── apple
    ├── orange
    ├── grape
    ├── pear
    └── strawberry
```

Which folder structure can be loaded by the `ImageDataset` class?

Which folder structure can be loaded by the <code>ImageDataset</code> class?

```
from PIL import open_image

img = open_image("test.png")
```

```
from PIL import Image

img = Image.load("test.png")
```

```
from PIL import Image

img = Image.open("test.png")
```

```
from PIL import Image

img = open("test.png")
```

Which is the correct command to load an image with PIL?

Which is the correct command to load an image with PIL?

Which attribute of `PIL.Image` can be used to obtain the width and height of an image?

Which attribute of <code>PIL.Image</code> can be used to obtain the width and height of an image?

Which `mode` should we use to load 1-bit pixels (black and white) using `PIL`?

Which <code>mode</code> should we use to load 1-bit pixels (black and white) using <code>PIL</code>?

Given the following code snippet:

```
from PIL import Image

img = Image.open("test.png")
print(img.format)
```

Which is the expected output for the script?

Given the following code snippet:
<pre><code>from PIL import Image

img = Image.open(&quot;test.png&quot;)
print(img.format)
</code></pre>
Which is the expected output for the script?

```
resize = transforms.Resize(224, 224)
```

```
resize = transforms.Resize(width=224, height=224)
```

```
resize = transforms.Resize({'width': 224, 'height': 224})
```

```
resize = transforms.Resize([224, 224])
```

Which is the correct code snippet to resize an image with `torchvision`?

Which is the correct code snippet to resize an image with <code>torchvision</code>?

```
tensor_func = transforms.ToTensorArray()
```

```
tensor_func = transforms.Tensor()
```

```
tensor_func = transforms.ConvertTensor()
```

```
tensor_func = transforms.ToTensor()
```

Which is the correct function to convert a PIL image, or `numpy.ndarray`, to tensor while using `torchvision`?

Which is the correct function to convert a PIL image, or <code>numpy.ndarray</code>, to tensor while using <code>torchvision</code>?

Which is the correct shape for a tensor image in PyTorch?

Which is the correct shape for a tensor image in PyTorch?

Test your knowledge of the Pytorch Image Model.

Introduction

Basic Concepts

Augmentation

Loss

Training

Model Conversion

Deployment

Appendix

Conclusion

Quiz (Basic)