Please enable JavaScript.
Coggle requires JavaScript to display documents.
PyTorch Paper Replicating - Coggle Diagram
- PyTorch Paper Replicating
- Make predictions on a custom image
- Replicating the ViT paper: an overview
-
-
-
- Equation 2: Multi-Head Attention (MSA)
-
-
-
- Putting it all together to create ViT
-
- Using a pretrained ViT from torchvision.models on the same dataset
-
-
-
-
-
-
-
- Create the Transformer Encoder
-
-
-
-
- Equation 1: Split data into patches and creating the class, position and patch embedding
-
-
-
-
-
-
-
-
- Equation 3: Multilayer Perceptron(MLP)
-
-
- Setting up training code for ViT model
-
-
-
-
-