StudEmo NLPerspectives - Baseline Experiments

[artikandri]

Continuation of https://github.com/CLARIN-PL/personalized-nlp/pull/6

What's new: experiments/

• personalized_nlp/experiments/goemo_avg_ann_notune.py : The experiment on GoEmo dataset with baseline AVG-ANN method without fine-tuning
• personalized_nlp/experiments/goemo_avg_ann_tune.py : The experiment on GoEmo dataset with baseline AVG-ANN method with fine-tuning
• personalized_nlp/experiments/goemo_single_ann_notune.py : The experiment on GoEmo dataset with baseline SINGLE-ANN method without fine-tuning
• personalized_nlp/experiments/goemo_single_ann_tune.py : The experiment on GoEmo dataset with baseline SINGLE-ANN method with fine-tuning
• personalized_nlp/experiments/studemo_avg_ann_notune.py : The experiment on StudEmo dataset with baseline AVG-ANN method without fine-tuning
• personalized_nlp/experiments/studemo_avg_ann_tune.py : The experiment on StudEmo dataset with baseline AVG-ANN method with fine-tuning
• personalized_nlp/experiments/studemo_single_ann_notune.py : The experiment on StudEmo dataset with baseline SINGLE-ANN method without fine-tuning
• personalized_nlp/experiments/studemo_single_ann_tune.py : The experiment on StudEmo dataset with baseline SINGLE-ANN method with fine-tuning

models/

• personalized_nlp/models/baseline_tune.py : The model for finetuning baseline method

What's changed:

• personalized_nlp/datasets/datamodule_base.py : added a parameter (is_averaged) to average annotations and modified compute_major_votes function in relation to the newly added parameter
• personalized_nlp/models/init.py : added the baseline_tune model
• personalized_nlp/models/baseline.py : added a parameter to set max_seq_len