roberta-base

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.2676
• Law Precision: 0.8739
• Law Recall: 0.9065
• Law F1: 0.8899
• Law Number: 107
• Violated by Precision: 0.8254
• Violated by Recall: 0.7324
• Violated by F1: 0.7761
• Violated by Number: 71
• Violated on Precision: 0.5077
• Violated on Recall: 0.5156
• Violated on F1: 0.5116
• Violated on Number: 64
• Violation Precision: 0.6460
• Violation Recall: 0.6979
• Violation F1: 0.6710
• Violation Number: 374
• Overall Precision: 0.6890
• Overall Recall: 0.7192
• Overall F1: 0.7037
• Overall Accuracy: 0.9504

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Law Precision	Law Recall	Law F1	Law Number	Violated by Precision	Violated by Recall	Violated by F1	Violated by Number	Violated on Precision	Violated on Recall	Violated on F1	Violated on Number	Violation Precision	Violation Recall	Violation F1	Violation Number	Overall Precision	Overall Recall	Overall F1	Overall Accuracy
No log	1.0	85	0.7040	0.0	0.0	0.0	107	0.0	0.0	0.0	71	0.0	0.0	0.0	64	0.0	0.0	0.0	374	0.0	0.0	0.0	0.7707
No log	2.0	170	0.3668	0.0	0.0	0.0	107	0.0	0.0	0.0	71	0.0	0.0	0.0	64	0.2416	0.2888	0.2631	374	0.2416	0.1753	0.2032	0.8896
No log	3.0	255	0.2618	0.3077	0.1869	0.2326	107	0.0	0.0	0.0	71	0.0	0.0	0.0	64	0.4626	0.5455	0.5006	374	0.4427	0.3636	0.3993	0.9171
No log	4.0	340	0.2232	0.7091	0.7290	0.7189	107	0.5316	0.5915	0.56	71	0.3523	0.4844	0.4079	64	0.5011	0.6016	0.5468	374	0.5179	0.6104	0.5604	0.9328
No log	5.0	425	0.1929	0.7778	0.8505	0.8125	107	0.84	0.5915	0.6942	71	0.44	0.5156	0.4748	64	0.5043	0.6257	0.5585	374	0.5666	0.6494	0.6051	0.9440
0.489	6.0	510	0.2214	0.7227	0.8037	0.7611	107	0.7538	0.6901	0.7206	71	0.4203	0.4531	0.4361	64	0.5683	0.6337	0.5992	374	0.5985	0.6510	0.6236	0.9447
0.489	7.0	595	0.2452	0.8598	0.8598	0.8598	107	0.7759	0.6338	0.6977	71	0.4853	0.5156	0.5	64	0.6460	0.6684	0.6570	374	0.6774	0.6818	0.6796	0.9469
0.489	8.0	680	0.2409	0.9245	0.9159	0.9202	107	0.7625	0.8592	0.8079	71	0.4321	0.5469	0.4828	64	0.6614	0.6738	0.6675	374	0.6883	0.7240	0.7057	0.9485
0.489	9.0	765	0.2760	0.8739	0.9065	0.8899	107	0.8529	0.8169	0.8345	71	0.5	0.5312	0.5152	64	0.6014	0.6898	0.6426	374	0.6612	0.7256	0.6920	0.9473
0.489	10.0	850	0.2676	0.8739	0.9065	0.8899	107	0.8254	0.7324	0.7761	71	0.5077	0.5156	0.5116	64	0.6460	0.6979	0.6710	374	0.6890	0.7192	0.7037	0.9504

Framework versions

• Transformers 4.44.0
• Pytorch 2.4.0
• Datasets 2.21.0
• Tokenizers 0.19.1