Qianyu2022 / Siamese_Network_Re

Siamese Networks for One-Shot Learning

Forked from the link https://github.com/kevinzakka/one-shot-siamese
But customized to be executed on current environment and different datasets

Paper Modifications

I've done some slight modifications to the paper to eliminate variables while I debug my code. Specifically, validation and test accuracy currently suck so I'm checking if there's a bug either in the dataset generation or trainer code.

• I'm using Relu -> Maxpool rather than Maxpool - Relu.
• I'm using batch norm between the conv layers.
• I'm using He et. al. initialization.
• I'm using a global learning rate, l2 reg factor, and momentum rather than per-layer parameters.

Omniglot Dataset

Please use the dataset in the Omniglot folder

• Process the data using data_prep_efficient.ipynb

Tips for excuting the codes for training Siamese Network

• All the needed hyperparameters including the dataset folder path are predefined in the config.py as default value
• The training is emulated on GPU, so please make sure the compatible CUDA and cuDNN are installed
• Current configurations are CUDA(11.7), Python(3.9), PyTorch(2.0) in the virtual environment GPU_AmpStruc
• If you want to switch to the tensorflow GPU, please change CUDA version in the CUDA_PATH from 11.7 to 11.2

Checkout Playground.ipynb for a minimal working example.