nojedag/xlm-roberta-finetuned-financial-news-sentiment-analysis-european

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

xlm-roberta-finetuned-financial-news-sentiment-analysis-european

This model is a fine-tuned version of FacebookAI/xlm-roberta-base on an unknown dataset. It achieves the following results on the evaluation set:

• —eval_loss: 0.4898
• —evalmodelpreparation_time: 0.0035
• —eval_accuracy: 0.8563
• —evalmacroprecision: 0.8561
• —evalmacrorecall: 0.8700
• —evalmacrof1: 0.8586
• —evalneutralprecision: 0.9346
• —evalneutralrecall: 0.7894
• —evalneutralf1: 0.8559
• —evalpositiveprecision: 0.8988
• —evalpositiverecall: 0.9350
• —evalpositivef1: 0.9165
• —evalnegativeprecision: 0.7350
• —evalnegativerecall: 0.8858
• —evalnegativef1: 0.8034
• —eval_runtime: 28.86
• —evalsamplesper_second: 287.942
• —evalstepsper_second: 18.018
• —step: 0

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• —learning_rate: 2e-05
• —trainbatchsize: 16
• —evalbatchsize: 16
• —seed: 42
• —gradientaccumulationsteps: 2
• —totaltrainbatch_size: 32
• —optimizer: Use OptimizerNames.ADAMWTORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizerargs=No additional optimizer arguments
• —lrschedulertype: linear
• —lrschedulerwarmup_steps: 846
• —num_epochs: 7
• —mixedprecisiontraining: Native AMP

Framework versions

• —Transformers 4.51.3
• —Pytorch 2.7.0+cu128
• —Datasets 3.6.0
• —Tokenizers 0.21.1