NTCAL/norbert2_sentiment_norec_en_gpu_3000_rader_3

norbert2_sentiment_norec_to_gpu_3000_rader_2

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

• Loss: 0.6214
• Compute Metrics: :
• Accuracy: 0.69
• Balanced Accuracy: 0.5
• F1 Score: 0.8166
• Recall: 1.0
• Precision: 0.69

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: 64
• eval_batch_size: 128
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 512
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Compute Metrics	Accuracy	Balanced Accuracy	F1 Score	Recall	Precision
No log	0.85	5	0.6279	:	0.6827	0.4994	0.8102	0.9816	0.6897
0.719	1.85	10	0.6217	:	0.6897	0.4998	0.8163	0.9995	0.6899
0.719	2.85	15	0.6225	:	0.69	0.5	0.8166	1.0	0.69
0.7646	3.85	20	0.6186	:	0.69	0.5	0.8166	1.0	0.69
0.7646	4.85	25	0.6214	:	0.69	0.5	0.8166	1.0	0.69

Framework versions

• Transformers 4.26.0
• Pytorch 1.13.1+cu117
• Datasets 2.9.0
• Tokenizers 0.13.2