Using ResNet18 To Classify Dogs

Image classified using a ResNet18 architecture. 
Image Source: Stanford-Dogs Dataset

What is a ResNet?

A ResNet, short form of “Residual Network,” created to solve the vanishing gradient (modifying the parameters does not result in an increase in performance) and exploding gradient (modifying the parameters results in large updates to the parameters and fails in the learning process) problems. This allows us to create larger, and more deep networks.


They solved this problem by using a skip connection, where they skip some layers to find as potentially more layers are unable to learn the identity well, and allows it to try and find the difference between the output and the input. Then, you add the transformed input (output of the network) to the initial input to pass back into the network again.

Implementation

*This code is a modification from the PyTorch website (https://pytorch.org/hub/pytorch_vision_resnet). You can get my modified code on GitHub (https://github.com/smrtdylan/Dog-Identification-Network). The Dataset is from Stanford-Dogs, and downloaded from Ajinkya Kolhe in Hugging Face.


1. First, let’s load up the model and set the model to evaluation mode (used to tell the model we’re testing it).



2. Next, we’re going to load in a version of the dataset from Huggin Face. We will be using the train split since we can actually see whether or not the model was actually correct. We will also shuffle the images in the dataset so that we will always get a random image.


3. Next, we’re going to download the labels for the ResNet. This does hold labels that are not in the Stanford-Dogs Dataset.


Now comes the fun part of pre-processing and passing the image into the network, getting the top 5 probabilities, and checking whether or not the model was correct. 


 4. Create the network and preprocess the image.

5. Now, we’re going to pass the images (in the same function) into the model and generate our probabilities. 


6. Now, create a new function with two parameters: the probabilities from the network and the true value/answer to the image. We are also going to print out the top 5 probabilities of the network.

7. Now, we’re going to create the ability (in the same function) to check whether or not the top answer by the model was correct.


8. Now, let’s create a loop to test on some images (10 in our case)! 


Here’s the final result, 



Works Referenced/Extra Resources

These were the works I referenced while researching for this article. I also included some other resources that I thought would be beneficial for everybody. 


“Exploding Gradient Problem” by DeepAI.

“Vanishing and Exploding Gradients Problems in Deep Learning” by GeeksForGeeks. 

“ResNet” by PyTorch Team.

"What are Skip Connections in Deep Learning?" by Sivaram (Analytics Vidhya).




Comments

Post a Comment

Popular posts from this blog

What is a Multimodal LLM?

Image
Introduction Multimodal is the ability for a model to process several different types of inputs. For example, ChatGPT can support text input as well as image input as one combined input. You could also have a model take in .pdf, .md, and input text as inputs (yes, these count as different modalities).  This allows for users to give represent different ideas in its rawest form possible, such as an image, instead of describing it. If you try to describe it, you could hit hurdles in trying to properly explain it and making sure it interpreted the description correctly.  How do Multimodal LLMs work?  Typically, these modalities are "fused" together. Here's an architecture diagram,  Image from NVIDIA In this case, we have an encoder for a specific modality (e.g. video encoder, image encoder, audio encoder). This encodes them into embeddings the LLM can understand and use. After encoding each specific modality, we then fuse them together. Then, the output is projected and ...
Read more

Top 3 Breakthroughs in Computer Vision in 2024

Image
Introduction Over the past few year, AI has become popular; however, no year matches the pace of the past few years. 2024 was an extremely crucial year for Computer Vision, with new models and breakthroughs occurring rapidly. Today, I'll be talking about the 3 biggest advancements in the Computer Vision field during 2024.  3. Vision Language Modeling Simply put, Vision Language Models, or VLMs for short, are Large Language Models (LLMs) that take an image and/or text as input. Think of ChatGPT, where you import an image and then type out a question that you have over the image. It will be able to understand both and output an answer. These are also what you would call a "multimodal model," which is a model that can understand multiple different types of sources/inputs.  Although Vision Language Models have existed for a while, they have recently become better and more accurate. However, this doesn't mean it's all sunshine an rainbows: there are still several barri...
Read more

A Mathematical Explanation of Gradient Descent

Image
Example of a close-to-optimal Gradient descent for a quadratic function. Credit: Dylan Introduction Although I wrote an article recently explaining Gradient Descent, I feel like a mathematical explanation of how it works would be beneficial not only to understand how it works, but also how you derive the functions.  Explanation First, we need a way to calculate how far off the model is. For our example, we will use the L2 Loss Function,  Let N represent the total amount of data points we have, i the iteration we are on, yᵢ represent the expected output, and ŷ represent the output predicted by the model.  Let's look more into ŷ. Since ŷ is the predicted value, this is the function that needs to have their parameters changed. Our goal is to change these parameters enough to minimize the loss of ŷ. For our example, let's set ŷ to be a linear line.  The values in "a" and "b" are our parameters. These will be the values we will have to change within f(x) to chang...
Read more