pytorch-example
Example of setup of a pytorch project. This was part of a lecture for MIT's Adv Nat Language Processing course, 6.864.
Project Structure
- All neural net code is setup as module for convienece
- data: defines dataset object and its utils
- models: define models and their utils
- train: defines code for training network and logging perfromance
- Run all scripts from a script directory
- main.py: entry point to run model
- requirements.txt
- requirements to run this code on a linux machine with CUDA 8. For alternate OS's, please check pytorch documentation.