initial_model

This model is a fine-tuned version of docketanalyzer/docket-lm-xs on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0187
• F1: 0.9938

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: 8
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 30
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss	F1
0.2297	0.0533	60	0.1805	0.9693
0.1411	0.1067	120	0.0593	0.9850
0.0099	0.16	180	0.0447	0.9908
0.0348	0.2133	240	0.0474	0.9892
0.0046	0.2667	300	0.0379	0.9923
0.0031	0.32	360	0.0334	0.9938
0.127	0.3733	420	0.0325	0.9933
0.1795	0.4267	480	0.0325	0.9928
0.0023	0.48	540	0.0364	0.9933
0.0028	0.5333	600	0.0353	0.9923
0.0043	0.5867	660	0.0290	0.9933
0.1299	0.64	720	0.0252	0.9938
0.188	0.6933	780	0.0235	0.9933
0.0019	0.7467	840	0.0208	0.9938
0.002	0.8	900	0.0199	0.9938
0.0525	0.8533	960	0.0192	0.9938
0.008	0.9067	1020	0.0190	0.9938
0.0013	0.96	1080	0.0193	0.9938

Framework versions

• Transformers 4.41.1
• Pytorch 2.3.0+cu121
• Datasets 2.14.4
• Tokenizers 0.19.1