cleaned up code/print statements/comments

ablation fix

Merge branch 'katherine_branch'

revised readme/setup/added diagnose leakage back

file organization/deleting redundant files

Merge branch 'main' of https://github.com/sashpol111/burnout-tracker

file organization / deletion

rewrite comments in data_loader.py and data_pipeline.py

created data pipeline engineering: 3 sources (apis and synthetic data) merged into 1 data set

fixed error analysis

rewrote / fixed error_analysis. generates png of 4 visualizations

fixed ai coach--multi-step ML pipeline that connects multiple models

implemented regularization techniques: l1/l2 & early stopping

fixed deterministic issue + created preprocessing data pipeline

preprocessing

DistilBERT fine-tuning complete - F1 0.753 AUC 0.925

synthetic data pipeline - 310 entries generated, F1 improves 0.893 to 0.895

add synthetic data generation, error analysis, train with synthetic pipeline

add error analysis with confusion matrix and failure case analysis

ablation study - 5 feature set experiments documented

fix system_prompt bug in get_burnout_advice

add Groq LLM advisor with burnout coaching

add visualizations - feature importance, training curves, model comparison

tune XGBoost with regularization - train 0.971 val 0.954

add data pipeline, baseline, and XGBoost model