Binary Classification as a Phase Separation Process (data repository)

<strong>For version 0.0.2 (from 2021) see below:</strong> This is a data repository for the paper "Binary classification as a phase separation process", by Rafael Monteiro. Website with description of this project: https://rafael-a-monteiro-math.github.io/Binary_classification_phase_separation/index.html Github: https://github.com/rafael-a-monteiro-math/Binary_classification_phase_separation This is a second version, which I wrote using tensorflow. It is much smaller (5 Gb when decompressed), a remarkable improvement when compared to the more than 100 Gb of the previous version). The new files are PSBC_BCs.tar.gz PSBC_classifier_PCA.tar.gz PSBC_dataset.tar.gz PSBC_libs_grids_statistics.tar.gz PSBC_notebooks.tar.gz Their content is explained in the file README_v2.pdf <strong>UPDATE: a Google Colab folder is also available. You can also find all the data and libraries there, unpacked.</strong> For usage, see the Git-hub. <strong>NOTE)</strong> I will keep the content for the previous version available in my Github as well. It is still a "nice exercise" to do all that is done in this new version in numpy, as done there. <strong><em>(Or, I should say, they should be studied as a cautionary tale of what to avoid.)</em></strong> <strong>For version 0.0.1 (from 2020) see below:</strong> This is a data repository for the paper "Binary classification as a phase separation process", by Rafael Monteiro. Website with description of this project: https://rafael-a-monteiro-math.github.io/Binary_classification_phase_separation/index.html Github: https://github.com/rafael-a-monteiro-math/Binary_classification_phase_separation Therein you will find Examples 1D toy model examples Computational statistics Several trained PSBC on MNIST dataset, with different parameter configurations Extra simulations, investigating normalization properties, low dimensional models that fail due to "too much" model compression, and comparison among ANNs, KNNs, and the PSBC in 1D If you want to know how to read the data how to access computational statistics, raw data, and examples how to use the data stored in this data repository see the guide README.pdf on GitHub page at Binary_Classification_Phase_Separation, where a script that downloads (and organizes) all this data is also available ("download_PSBC.sh). I did not include a copy of the train-test set (0-1dubset of the MNIST database) in every folder with simulations. But you can find a copy of the normalized dataset in the tar ball "PSBC_Examples.tar.gz" as data_test_normalized_MNIST.csv and data_train_normalized_MNIST.csv.**版本0.0.2（2021年发布）详见下文：** 本数据集仓库对应拉斐尔·蒙特罗（Rafael Monteiro）发表的论文《二元分类作为相分离过程（Binary classification as a phase separation process）》。本项目的说明网站：https://rafael-a-monteiro-math.github.io/Binary_classification_phase_separation/index.html；GitHub仓库地址：https://github.com/rafael-a-monteiro-math/Binary_classification_phase_separation。 本版本为第二版，基于TensorFlow开发，体积大幅缩减，解压后仅占5GB，相较前一版本超过100GB的体量实现了显著优化。 新增文件包括：PSBC_BCs.tar.gz、PSBC_classifier_PCA.tar.gz、PSBC_dataset.tar.gz、PSBC_libs_grids_statistics.tar.gz以及PSBC_notebooks.tar.gz。各文件的内容说明可查阅README_v2.pdf文档。 **更新：** 现已同步提供Google Colab文件夹，其中包含所有已解压的数据集与依赖库。 使用方法请查阅对应GitHub仓库。 **注意：** 我仍会在GitHub仓库中保留前一版本的全部内容。若使用NumPy复现本新版本的全部功能，仍是一项颇具价值的练习。 **（或者说，这也可作为一则规避不良实践的警示案例进行学习。）** **版本0.0.1（2020年发布）详见下文：** 本数据集仓库对应拉斐尔·蒙特罗（Rafael Monteiro）发表的论文《二元分类作为相分离过程（Binary classification as a phase separation process）》。本项目的说明网站与GitHub仓库地址同上。 其中包含以下内容：一维玩具模型示例、计算统计结果、基于MNIST数据集（MNIST dataset）训练得到的多款不同参数配置的相分离二元分类器（Phase Separation Binary Classifier, PSBC）模型、针对归一化特性的额外仿真实验、因「过度模型压缩」导致失效的低维模型，以及人工神经网络（Artificial Neural Networks, ANNs）、K近邻（K-Nearest Neighbors, KNNs）与一维PSBC模型的性能对比。 若您希望了解如何读取数据集、获取计算统计结果与原始数据，或是如何使用本仓库存储的数据，请查阅GitHub仓库Binary_Classification_Phase_Separation中的README.pdf指南文档，其中同时提供了用于下载并整理全部数据集的脚本`download_PSBC.sh`。 本仓库的每个仿真文件夹中均未附带训练集与测试集（MNIST数据库的0-1子集）的副本，但您可在压缩包PSBC_Examples.tar.gz中找到归一化后的数据集副本，文件名为data_test_normalized_MNIST.csv与data_train_normalized_MNIST.csv。