The code of this repository is referenced to Deep-Tutorial-for-PyTorch
These tutorials is the implementation of some typical papers. Below is the code directories and their corresponding papers.
| Tutorial | Paper |
|---|---|
| Image Captioning | Show, Attend, and Tell |
| Sequence Labeling | Empower Sequence Labeling with Task-Aware Neural Language Model |
| Object Detection | SSD: Single Shot MultiBox Detector |
| Text Classification | Hierarchical Attention Networks for Document Classification |
| Super-Resolution | Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network |
| Machine Translation | Attention Is All You Need |
Take ImageCaptioning as an example to introduce the file dictionary structure, the others are similar.
.
|--ImageCaptioning
| |--create_input_files.py // Process source data files
| |--utils.py // Utility module
| |--datasets.py // Create data source for GeneratorDataset
| |--models.py // Model file
| |--train.py // Train the model
| |--eval.py // Evaluate the model
| |--caption.py // Caption the input image