Update lextreme.py

added config name

fixed typo

remove non-critical classes from datasets predicting the degree of criticality

Update lextreme.py

Add criticality prediction configs (#8)

Added Judgment Prediction Dataset for Turkish Court of Cassation (#5)

added lextreme citation

Fixed Bug for multi_eurlex

Fixed Bug for multi_eurlex

Fixed Bug in loading script

Fixed Bug for multi_eurlex

adapted data loader

made fine grained labels upper case as well because of problems with seqeval otherwise

fixed comma problem

updated NER labelsets

added language

fixed label classes of greek_legal_code_subject

added first version of lextreme benchmark