Data matrix for training XMTC machine learning models (TF-IDF) with TNPP lemmatisation (fi)

Data matrix for training XMTC machine learning models (TF-IDF) with TNPP lemmatisation (Turku Neural Parser Pipeline). Based on a Finnish corpus. Textual data follows the Bag-of-Words feature file format of The Extreme Classification Repositoryn (http://manikvarma.org/downloads/XC/XMLRepository.html).

The first line is formatted as:

total_documents number_of_features number_of_labels

All other lines represent one document per line:

label1,label2,...,labelk ft1:ft1_val ft2:ft2_val ft3:ft3_val .. ftd:ftd_val

i.e, comma-separated list of labels followed by all non-zero components of the TF-IDF vector given as component_number:value.