Ivsucram
/
ATL_CIKM-2019
mirror of https://github.com/Ivsucram/ATL_Python.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
ATL_CIKM-2019/README.md

4.7 KiB
Raw Blame History

Reference

Paper

ATL: Autonomous Knowledge Transfer from Many Streaming Processes

ArXiv

ResearchGate

ACM Digital Library

Bibtex

@inproceedings{10.1145/3357384.3357948,
author = {Pratama, Mahardhika and de Carvalho, Marcus and Xie, Renchunzi and Lughofer, Edwin and Lu, Jie},
title = {ATL: Autonomous Knowledge Transfer from Many Streaming Processes},
year = {2019},
isbn = {9781450369763},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
url = {https://doi.org/10.1145/3357384.3357948},
doi = {10.1145/3357384.3357948},
booktitle = {Proceedings of the 28th ACM International Conference on Information and Knowledge Management},
pages = {269–278},
numpages = {10},
keywords = {concept drif, transfer learning, deep learning, multistream learning},
location = {Beijing, China},
series = {CIKM ’19}
}

Notes

If you want to see the original code used for this paper, access ATL_Matlab

ATL_Python is a reconstruction of ATL_Matlab made by the same author, but using Python 3.6 and PyTorch (with autograd enabled and GPU support). The code is still not one-to-one and some differences in results can be found (specially on the data split methods in DataManipulator, however the network structure is correct and can be used by whoever is interested on this work in order to understand the structure or to build comparative results with your own research work.

Having said that, expect ATL_Python to be updated in the following weeks, including functions refactoring and functions documentation.

ATL_Python

ATL: Autonomous Knowledge Transfer From Many Streaming Processes ACM CIKM 2019

  1. Clone ATL_Python git to your computer, or just download the files.

  2. Install anaconda or miniconda.

  3. Open Anaconda prompt and travel until ATL folder.

  4. Run the following command conda env create -f environment.yml. This will create an environment called atl with every python packaged/library needed to run ATL.

  5. Enable ATL environment by running the command activate atl or conda activate atl.

  6. Provide a dataset by replacing the file data.csv The current data.csv holds SEA dataset. data.csv must be prepared as following:

- Each row presents a new data sample
- Each column presents a data feature
- The last column presents the label for that sample. Don't use one-hot encoding. Use a format from 1 onwards
  1. Run python ATL.py

ATL will automatically normalize your data and split your data into 2 streams (Source and Target data streams) with a bias between them, as described in the paper.

ATL statues are printed at the end of every minibatch, where you will be able to follow useful information as:

- Training time (maximum, mean, minimum, current and accumulated)
- Testing time (maximum, mean, minimum, current and accumulated)
- Classification Rate for the Source (maximum, mean, minimum and current)
- Classification Rate for the Target (maximum, mean, minimum and current)
- Classification Loss for the Source (maximum, mean, minimum and current)
- Classification Loss for the Target (maximum, mean, minimum and current)
- Reconstruction Loss for the Target (maximum, mean, minimum and current)
- Kullback-Leibler Loss (maximum, mean, minimum and current)
- Number of nodes (maximum, mean, minimum and current)
- And a quick review of ATL structure (both discriminative and generative phases), where you can see how many automatically generated nodes were created.

At the end of the process, ATL will plot 6 graphs:

- The processing time per mini-batch and the total processing time as well, both for training and testing
- The evolution of nodes over time
- The target and source classification rate evolution, as well as the final mean accuracy of the network

Thank you.

Download all datasets used on the paper

As some datasets are too big, we can't upload them to GitHub. GitHub has a size limit of 35MB per file. Because of that, you can find all the datasets in a csv format on the anonymous link below. To test it, copy the desired dataset to the same folder as ATL and rename it to data.csv.