hbertv2-emotion-logit_KD_new

This model is a fine-tuned version of gokuls/bert_12_layer_model_v2_complete_training_new_48 on the emotion dataset. It achieves the following results on the evaluation set:

• Loss: 0.6082
• Accuracy: 0.8855

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: 5e-05
• train_batch_size: 64
• eval_batch_size: 64
• seed: 33
• distributed_type: multi-GPU
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.1324	1.0	250	1.0855	0.801
0.8121	2.0	500	0.7251	0.872
0.5982	3.0	750	0.6929	0.8695
0.4694	4.0	1000	0.6529	0.8775
0.3873	5.0	1250	0.7370	0.873
0.3477	6.0	1500	0.6082	0.8855
0.3169	7.0	1750	0.6202	0.885
0.2855	8.0	2000	0.5843	0.88
0.2669	9.0	2250	0.6290	0.8825
0.2493	10.0	2500	0.7612	0.8785
0.2326	11.0	2750	0.6896	0.883

Framework versions

• Transformers 4.35.2
• Pytorch 1.14.0a0+410ce96
• Datasets 2.15.0
• Tokenizers 0.15.0