- Always look at the open issues
- Make sure to add comments before pushing
- Use variable name that make sense
- Open Pull Requests everytime you push
-
ResNet and FAISS : convert the image dataset, into vectors and build the FAISS Index over that
- (Jiwon & Rosie)
- Dataset (CIFAR-10) : https://www.cs.toronto.edu/~kriz/cifar.html
- Download the dataset to /localdisk3/ on node 2x16a
- Go through ResNet Code : https://github.com/KaimingHe/deep-residual-networks
- Download the code and convert the CIFAR-10 images to vector embeddings
- Load the vector embeddings into a FAISS Index build on the GPU
- FAISS Index code can be found here : https://github.com/DataIntelligenceCrew/koios-semantic-search
-
Algorithm and Complexity Analysis : pseudocode for the algorithm and complexity analysis
- (Pranay)
- Write out the algorithm