selims

This model is a fine-tuned version of nlptown/bert-base-multilingual-uncased-sentiment on the tweet_eval dataset.

Model description

This is a multilingual model for sentiment analysis that provides outputs ranging from 1 to 5, following the same logic as the 1 to 5-star reviews.

Intended uses & limitations

This sentiment model can be applied to datasets in the following languages: English, Dutch, German, French, Spanish, and Italian.

Training and evaluation data

For fine-tunning of this model, the Tweet_eval dataset was used.

Training procedure

Please refer to the information below:

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 3.0

Framework versions

• Transformers 4.15.0
• Pytorch 1.10.1+cpu
• Datasets 2.0.0
• Tokenizers 0.10.3