JHurricane96/emo_lex

Creating emotion lexicons for 250+ low-resource languages # emo_lex Experiment runner for paper Cross-Lingual Emotion Lexicon Induction using Representation Alignment in Low-Resource Settings. # Usage ### Setup 1. Clone this repo 2. Clone fastText, multilingual-nlm and vecmap. 3. Install python packages `numpy`, `cupy`, `torch` and pot ## Embedding Alignment Some conventions: - Languages are referred to by their 3 letter ISO code. - Each bible file should be named ` .txt`, e.g. Spanish would be `spa.txt`. - Each line in a Bible file should contain 1 sentence/verse. The text should be pre-processed to be lowercased and contain space-separated words (no punctuation unless it's in the middle of a word, e.g. hypenation). The 3 algorithms used for embedding alignment: ### Wasserstein-Procrustes ``` python align.py \ --langs \ --bible_dir \ --align_dir \ --emb_dir \ --num_gpus \ --algorithm fb \ --fasttext_dir ``` ### Neural Language Model ``` python align.py \ --langs \ --bible_dir \ --align_dir \ --num_gpus \ --algorithm nlm \ --nlm_dir \ --nlm_preproc_dir \ --nlm_preprocess \ --nlm_modified ``` ### Orthogonal Refinement ``` python align.py \ --langs \ --bible_dir \ --sid_bible_dir \ --align_dir \ --emb_dir \ --num_gpus \ --algorithm vecmap \ --vecmap_dir \ --fasttext_dir ``` The sentence ID Bibles are like the normal Bibles, except each line is prefixed with a sentence ID followed by a tab. Sentence ID for translations of the same sentence across different language should be the same. Omit the `--sid_bible_dir` argument to run the original vecmap algorithm. ## Emotion Lexicon Induction and Evaluation ``` python eval.py \ --langs \ --align_dir \ --exp_id \ --trans_dir \ --emo_lex_dir \ --nns_dir \ --reports_dir ``` - The path provided to `nns_dir` and `reports_dir` is suffixed by `exp_id` so that multiple runs can share the same paths and just have different experiment IDs. - The ground-truth word translation files should be named ` _ .txt`, e.g. `spa_eng.t …