distilbert-base-uncased_fold_1_binary_v1

This model is a fine-tuned version of distilbert-base-uncased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.7296
• F1: 0.8038

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss	F1
No log	1.0	288	0.4152	0.7903
0.3956	2.0	576	0.4037	0.8083
0.3956	3.0	864	0.5601	0.7996
0.181	4.0	1152	0.8571	0.8023
0.181	5.0	1440	0.9704	0.7822
0.0935	6.0	1728	0.9509	0.8074
0.0418	7.0	2016	1.1813	0.7736
0.0418	8.0	2304	1.2619	0.7859
0.0134	9.0	2592	1.4275	0.7863
0.0134	10.0	2880	1.4035	0.8019
0.0127	11.0	3168	1.4903	0.7897
0.0127	12.0	3456	1.5853	0.7919
0.0061	13.0	3744	1.6628	0.7957
0.0058	14.0	4032	1.5736	0.8060
0.0058	15.0	4320	1.6226	0.7929
0.0065	16.0	4608	1.6395	0.8010
0.0065	17.0	4896	1.6556	0.7993
0.002	18.0	5184	1.7075	0.8030
0.002	19.0	5472	1.6925	0.7964
0.0058	20.0	5760	1.6511	0.8030
0.0013	21.0	6048	1.6135	0.8037
0.0013	22.0	6336	1.6739	0.8028
0.0001	23.0	6624	1.7014	0.8109
0.0001	24.0	6912	1.7015	0.8045
0.002	25.0	7200	1.7296	0.8038

Framework versions

• Transformers 4.21.0
• Pytorch 1.12.0+cu113
• Datasets 2.4.0
• Tokenizers 0.12.1