OCCODE files

These are the files, results, model and source code for the OCCODE project. They include: The results from running the occupational column from the full county 1950 Norwegian census through the OCCODE pipeline. An overview of the coding system used by Statistics Norway to encode the occupations of the 1950 Norwegian full count census.This overview was created by Statistics Norway, and the original can be found here: https://www.ssb.no/a/folketellinger/ The model trained on the Random sample dataset. Previous results from running the occupational column from the full county 1950 Norwegian census through the OCCODE pipeline. Note that these results are worse than the current ones, found in RandomSample_total_confidence_scores.csv The source code for the OCCODE pipeline as of 2021.05.11本数据集包含OCCODE项目的各类文件、实验结果、训练模型及源代码，具体如下：针对挪威1950年全郡人口普查数据中的职业字段运行OCCODE流水线（OCCODE pipeline）所得的实验结果；挪威统计局（Statistics Norway）编制的挪威1950年全量人口普查职业编码体系概览，该文档的原始版本可通过以下链接获取：https://www.ssb.no/a/folketellinger/；基于随机样本数据集训练得到的模型；此前针对挪威1950年全郡人口普查职业字段运行OCCODE流水线所得的实验结果，需注意该批次结果劣于当前版本结果，当前结果存储于RandomSample_total_confidence_scores.csv文件中；截至2021年5月11日的OCCODE流水线源代码。