Assessing patterns of agreement across monolingual and codeswitched data.

Model: https://github.com/google-research/bert/blob/master/multilingual.md

use python3 pls pip3 install {pytorch_pretrained_bert, seaborn, pandas}

from mbert-aa-cs: python3 generate_data.py > data/prep_anim.txt

Data will now be in mbert-aa-cs/data/prep_anim.txt

from mbert-aa-cs: python3 eval_bert.py --data data/prep_anim.txt > results/enfrru_prep_anim_bert.txt

Results will now be in mbert-aa-cs/results

R code is also located here: https://colab.research.google.com/drive/1t2p3bejrjSH45UPFDMUrXSL4W5MBSRCF?usp=sharing play around with analyze_results.py my sincerest apologies for the messiness