finetuned_iebert-base_on_shEMO_transcripts

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

• Loss: 1.3279
• Accuracy: 0.5367

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: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.4217	1.0	75	1.2872	0.4933
1.2468	2.0	150	1.1732	0.5433
1.202	3.0	225	1.0937	0.5833
1.1722	4.0	300	1.1343	0.5767
1.1248	5.0	375	1.0780	0.58
1.0578	6.0	450	1.0778	0.59
0.9632	7.0	525	1.0967	0.59
0.8325	8.0	600	1.1147	0.5933
0.8339	9.0	675	1.1271	0.61
0.7494	10.0	750	1.1886	0.59
0.6416	11.0	825	1.2204	0.5833
0.5611	12.0	900	1.3593	0.5567
0.5421	13.0	975	1.3113	0.56
0.53	14.0	1050	1.3207	0.5733
0.4981	15.0	1125	1.3327	0.5567

Framework versions

• Transformers 4.34.1
• Pytorch 2.1.0+cu118
• Datasets 2.14.6
• Tokenizers 0.14.1