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 function will export a PyTorch model to an ONNX model?

Which function will export a PyTorch model to an ONNX model?

Which package provides support for ONNX inference and training?

Which package provides support for ONNX inference and training?

**(Select all that apply.)** Which functions are useful for exporting an ONNX model to a TensorFlow model?

(Select all that apply.) Which functions are useful for exporting an ONNX model to a TensorFlow model?

```
import numpy

def softmax(vec):
    exponential = 2 * np.pi
    probabilities = exponential * np.sum(exponential)
    return probabilities
```

```
import numpy

def softmax(vec):
    exponential = np.exp(vec)
    probabilities = exponential * np.sum(exponential)
    return probabilities
```

```
import numpy

def softmax(vec):
    exponential = 2 * np.pi
    probabilities = exponential / np.sum(exponential)
    return probabilities
```

```
import numpy

def softmax(vec):
    exponential = np.exp(vec)
    probabilities = exponential / np.sum(exponential)
    return probabilities
```

Which is the correct implementation for a `softmax` utility function?

Which is the correct implementation for a <code>softmax</code> utility function?

Which functions permute the dimensions of a tensor?

Which functions permute the dimensions of a tensor?

```
import numpy as np

def to_numpy(img):
    np_array = np.asarray(img, dtype=np.float32)
    np_array = np.transpose(np_array, (1, 0, 2))
    np_array = np.expand_dims(np_array, axis=0)
    np_array /= 255.0

    return np_array
```

```
import numpy as np

def to_numpy(img):
    np_array = np.asarray(img, dtype=np.float32)
    np_array = np.transpose(np_array, (2, 0, 1))
    np_array = np.expand_dims(np_array, axis=0)

    return np_array
```

```
import numpy as np

def to_numpy(img):
    np_array = np.asarray(img, dtype=np.float32)
    np_array = np.transpose(np_array, (2, 0, 1))
    np_array = np.expand_dims(np_array, axis=0)
    np_array /= 255.0

    return np_array
```

```
import numpy as np

def to_numpy(img):
    np_array = np.asarray(img, dtype=np.float32)
    np_array = np.transpose(np_array, (2, 0, 1))
    np_array /= 255.0

    return np_array
```

Which is the correct implementation to convert a `PIL.Image` into a `numpy` array?

Which is the correct implementation to convert a <code>PIL.Image</code> into a <code>numpy</code> array?

```
import numpy as np

k = 4
top_ids = np.argsort(probabilities)[k:]
```

```
import numpy as np

k = 4
top_ids = np.argsort(probabilities)[-k:]
```

```
import numpy as np

k = 4
top_ids = np.argsort(probabilities)[:k]
```

```
import numpy as np

k = 4
top_ids = np.argsort(probabilities)[:-k]
```

Which is the correct implementation to get `topk` predictions using `np.argsort`?


Which is the correct implementation to get <code>topk</code> predictions using <code>np.argsort</code>?

```
import tensorflow as tf

tensor = tf.io.decode_image('test.jpg', channels=3, dtype=tf.dtypes.float32)
```

```
import tensorflow as tf

img = tf.io.read_file('test.jpg')
tensor = tf.io.decode_image(img, channels=3, dtype=tf.dtypes.float32)
```

```
import tensorflow as tf

img = tf.io.read_file('test.jpg', mode='RGB')
```

```
import tensorflow as tf

img = tf.io.read_file('test.jpg')
tensor = tf.io.decode_image(img, mode='RGB', dtype=tf.dtypes.float32)
```

Which is the correct implementation to load an image and convert it to RGB format using `tf.io`?

Which is the correct implementation to load an image and convert it to RGB format using <code>tf.io</code>?

Model Conversion

Model Conversion

Test your knowledge of the model conversion processes in the PyTorch Image Model.

Introduction

Basic Concepts

Augmentation

Loss

Training

Model Conversion

Deployment

Appendix

Conclusion

Quiz (Model Conversion)