# Multi-metric optimization

This model is a fine-tuned version based on microsoft/deberta-v3-base, achieving an outstanding accuracy of 99.87% on the evaluation set.

A sentiment analysis model fine-tuned from vinai/phobert-base-v2, achieving 93.6% accuracy on the evaluation set

A weather classification model fine-tuned based on Google's ViT model, excelling in image classification tasks with an accuracy of 96.79%.

This model is an image caption generation model based on the GIT architecture, fine-tuned on the HL Narratives dataset, specifically designed for generating high-level narrative image descriptions.

English pronunciation assessment model based on wav2vec2-large-xlsr-53 for evaluating English pronunciation quality

This model is a fine-tuned version based on ibert-roberta-base, specifically designed for detecting abusive or threatening speech.

A vision model fine-tuned from google/vit-base-patch16-224-in21k for facial expression recognition tasks

A Korean sentiment analysis model fine-tuned based on KcELECTRA-base-v2022, which achieved good results on the evaluation set.

A visual classification model fine-tuned on an unknown dataset based on Google's ViT model, excelling in image classification tasks

An image classification model fine-tuned based on microsoft/resnet-18, performing excellently on image folder datasets.

A BERT-based fine-tuned named entity recognition model achieving an F1 score of 95.91% on the evaluation set

A Slovak POS tagging model fine-tuned on Universal Dependencies sk_snk dataset based on distil-slovakbert

This is a FinBERT model fine-tuned for Finnish sentiment analysis tasks, trained and evaluated using the FinnSentiment dataset.

A Dutch malicious comment detection model based on the RobBERT model, used to identify toxic or potentially harmful comment content

A text classification model fine-tuned on the Amazon product review sentiment analysis dataset based on the BERT base model, achieving 94.65% accuracy

Named Entity Recognition model fine-tuned based on RoBERTa-base architecture, specifically optimized for Twitter text