遇见数据集

Towards generalizable deep ptychography neural networks - Data and Model artifacts

收藏
Zenodo2025-09-26 更新2026-05-26 收录
官方服务:

资源简介:

This data repository contains the training datasets for all ptychography experiments used in the manuscript. To ensure full replicability, it is recommended each of these files be moved into the root directory of the accompanying repository located here: [https://github.com/AdvancedPhotonSource/PtychoPINN-torch-pub]. They must be unzipped and placed in the root directory to be read/accessible. All files are named the same as their corresponding references in the manuscript, and further details on the experiments themselves are in the manuscript. The model artifacts needed for full replicability are also included, which were created using mlflow as experiment tracking software. The accompanying notebook in the code repository, "recreate_results.ipynb", contains dictionaries that reference specific file uris that can be used to re-train or perform inference using the same model/data setup as was reported.

提供机构:
Zenodo
创建时间:
2025-09-26
二维码
社区交流群
二维码
科研交流群
商业服务