# Focal Loss Optimization

Multilabel sentiment classification model fine-tuned from BERT-base-uncased, supporting 28 emotion recognitions

A BERT-based Chinese sentiment analysis model for determining the emotional tendency of text (positive, negative, or neutral).

RetinaNet is an accurate and fast single-stage object detection model that uses a feature pyramid network and focal loss function to address class imbalance issues.

A sentiment classification model based on the BERT pre-trained model, specifically designed to identify emotional categories in text, supporting multiple sentiment label classifications.