Pytorch Generalization

Pytorch implementations of deep learning libraries.

Table of Contents

1. About
2. Model Converting Instructions
3. References
4. Related Libraries

About

I was really inspired by the work of Yiding Jiang and his group in the paper "Predicting the generalization gap in deep networks with margin distributions", and sought to replicate the work in pytorch. Along the way I noticed that many papers that use similiar methods to predict generalization lacked full pytorch implimentations, and sought to convert them.

Model Converting Instructions

1. Clone this repo
2. Download Starting kit and Public Data from Codalab
3. Add path to "ingestion_program" , in line 7 of 'convert_models.py'
4. Add path to "input_data", in line 19 of 'convert_models.py'

License

Distributed under the MIT License. See LICENSE for more information.

References

• [1]Y. Jiang, D. Krishnan, H. Mobahi, and S. Bengio, “Predicting the Generalization Gap in Deep Networks with Margin Distributions,” arXiv:1810.00113 [cs, stat], Jun. 2019, Accessed: Jul. 27, 2020. [Online]. Available: http://arxiv.org/abs/1810.00113.

Related Libraries

generalizatio gap features tensorflow