distilbert-finetuned-financial-news-sentiment-analysis-european

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

eval_loss: 0.7528
eval_model_preparation_time: 0.002
eval_accuracy: 0.7628
eval_macro_precision: 0.7622
eval_macro_recall: 0.7619
eval_macro_f1: 0.7611
eval_neutral_precision: 0.7921
eval_neutral_recall: 0.7675
eval_neutral_f1: 0.7796
eval_positive_precision: 0.8106
eval_positive_recall: 0.7607
eval_positive_f1: 0.7849
eval_negative_precision: 0.6838
eval_negative_recall: 0.7575
eval_negative_f1: 0.7188
eval_runtime: 17.582
eval_samples_per_second: 472.643
eval_steps_per_second: 29.576
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
train_batch_size: 16
eval_batch_size: 16
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 32
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 846
num_epochs: 7
mixed_precision_training: Native AMP