README.md

metadata

tags:
  - generated_from_trainer
metrics:
  - accuracy
  - f1
  - precision
  - recall
base_model: DeepPavlov/distilrubert-tiny-cased-conversational-v1
model-index:
  - name: distilrubert-tiny-cased-conversational-v1_finetuned_empathy_classifier
    results: []

distilrubert-tiny-cased-conversational-v1_finetuned_empathy_classifier

This model is a fine-tuned version of DeepPavlov/distilrubert-tiny-cased-conversational-v1 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.6624
• Accuracy: 0.6780
• F1: 0.6878
• Precision: 0.7175
• Recall: 0.6780

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: 0.0001
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=0.0001
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
1.09	1.0	9	1.0661	0.4407	0.4464	0.6498	0.4407
1.0292	2.0	18	0.9658	0.5678	0.5223	0.5179	0.5678
0.942	3.0	27	0.8659	0.5932	0.5807	0.5723	0.5932
0.8614	4.0	36	0.7864	0.6186	0.5924	0.5879	0.6186
0.8002	5.0	45	0.7766	0.6017	0.5946	0.6086	0.6017
0.7633	6.0	54	0.7545	0.6186	0.6022	0.6151	0.6186
0.7249	7.0	63	0.7649	0.6356	0.6381	0.6921	0.6356
0.6687	8.0	72	0.7115	0.6695	0.6741	0.7154	0.6695
0.6426	9.0	81	0.6554	0.6864	0.6761	0.6807	0.6864
0.6144	10.0	90	0.6649	0.6864	0.6909	0.7172	0.6864
0.6252	11.0	99	0.8685	0.6186	0.6118	0.6880	0.6186
0.5988	12.0	108	0.6306	0.6949	0.7015	0.7107	0.6949
0.56	13.0	117	0.6919	0.6610	0.6662	0.7061	0.6610
0.5468	14.0	126	0.6563	0.6949	0.6980	0.7188	0.6949
0.5658	15.0	135	0.6351	0.6949	0.7048	0.7280	0.6949
0.5262	16.0	144	0.6902	0.6780	0.6821	0.7173	0.6780
0.4777	17.0	153	0.6237	0.6949	0.6981	0.7056	0.6949
0.4771	18.0	162	0.6688	0.6780	0.6799	0.7035	0.6780
0.4737	19.0	171	0.6482	0.6864	0.6957	0.7219	0.6864
0.5033	20.0	180	0.6624	0.6780	0.6878	0.7175	0.6780

Framework versions

• Transformers 4.21.2
• Pytorch 1.12.1+cu113
• Datasets 2.4.0
• Tokenizers 0.12.1