hidden_representation_BERT.ipynb - In this notebook, I have implemented a probing classifier for POS tagging NLP task by measuring how many parameters from the input representations are necessary to learn the task to certain degrees by dropping out some systematically. Keywords - PyTorch, Fine tuning BERT, Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA), POS Tagging
sentiment_analysis_multiclass_BERT.ipynb - In this notebook, I have investiagted the performance of a pre-trained BERT model against baseline models, including Logistic Regression and Support Vector Machines. This comprehensive analysis involved performing sentiment analysis on the Rotten Tomato movie review dataset, providing valuable insights into the efficiency and superiority of the advanced BERT model in capturing nuanced sentiment patterns. Keywords - Pre Trained BERT Model, Downstream Classification Task, PyTorch, Pre-processing Data, Linear Regression, Support Vector Machine Classification, GPU Acceleration