distilrubert-tiny-cased-conversational-v1_empathy_preprocessed_punct_lowercasing

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: 0.6036
• Accuracy: 0.7458
• F1: 0.7409
• Precision: 0.7420
• Recall: 0.7458

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
1.0953	1.0	9	1.0692	0.4661	0.3740	0.3740	0.4661
1.066	2.0	18	1.0242	0.5593	0.5491	0.5446	0.5593
1.0119	3.0	27	0.9259	0.6102	0.6106	0.6147	0.6102
0.9118	4.0	36	0.8659	0.5847	0.5349	0.5835	0.5847
0.8921	5.0	45	0.7925	0.6356	0.6133	0.6275	0.6356
0.83	6.0	54	0.7776	0.6271	0.6087	0.6199	0.6271
0.8015	7.0	63	0.7675	0.6695	0.6601	0.6871	0.6695
0.7334	8.0	72	0.7133	0.6780	0.6659	0.6748	0.6780
0.696	9.0	81	0.6939	0.6864	0.6758	0.6833	0.6864
0.6349	10.0	90	0.6555	0.7119	0.7057	0.7085	0.7119
0.6482	11.0	99	0.6585	0.7288	0.7202	0.7339	0.7288
0.5924	12.0	108	0.6223	0.7373	0.7332	0.7343	0.7373
0.5437	13.0	117	0.6364	0.7288	0.7231	0.7296	0.7288
0.5653	14.0	126	0.6158	0.7373	0.7266	0.7342	0.7373
0.5314	15.0	135	0.6104	0.7458	0.7439	0.7435	0.7458
0.4912	16.0	144	0.6119	0.7458	0.7433	0.7442	0.7458
0.4819	17.0	153	0.6040	0.7458	0.7452	0.7448	0.7458
0.4873	18.0	162	0.6113	0.7288	0.7248	0.7275	0.7288
0.4729	19.0	171	0.6035	0.7373	0.7292	0.7341	0.7373
0.4654	20.0	180	0.6036	0.7458	0.7409	0.7420	0.7458

Framework versions

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