2024-01-04_one_stage_subgraphs_weighted_txt_vis_conc_2_5_9_11_gate

This model is a fine-tuned version of microsoft/layoutlmv3-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.0946
• Accuracy: 0.785
• Exit 0 Accuracy: 0.0525
• Exit 1 Accuracy: 0.0625
• Exit 2 Accuracy: 0.0725
• Exit 3 Accuracy: 0.0625
• Exit 4 Accuracy: 0.52

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: 2
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 24
• total_train_batch_size: 48
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 60

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Exit 0 Accuracy	Exit 1 Accuracy	Exit 2 Accuracy	Exit 3 Accuracy	Exit 4 Accuracy
No log	0.96	16	2.6478	0.1925	0.0625	0.0625	0.0625	0.08	0.0625
No log	1.98	33	2.4825	0.27	0.0675	0.0625	0.0625	0.0625	0.0625
No log	3.0	50	2.2710	0.2975	0.06	0.0625	0.0625	0.0625	0.0625
No log	3.96	66	2.0409	0.4125	0.05	0.0625	0.0625	0.0625	0.0625
No log	4.98	83	1.7483	0.55	0.0375	0.0625	0.0625	0.0625	0.0625
No log	6.0	100	1.4619	0.65	0.0325	0.0625	0.065	0.0625	0.0625
No log	6.96	116	1.2754	0.68	0.035	0.0625	0.0625	0.0625	0.0625
No log	7.98	133	1.1304	0.7125	0.0375	0.0625	0.0625	0.0625	0.0625
No log	9.0	150	1.0446	0.7175	0.0375	0.0625	0.06	0.0625	0.0625
No log	9.96	166	0.9698	0.745	0.0375	0.0625	0.0625	0.0625	0.065
No log	10.98	183	0.9085	0.75	0.04	0.0625	0.0625	0.0625	0.085
No log	12.0	200	0.8824	0.7475	0.0425	0.0625	0.0575	0.0625	0.09
No log	12.96	216	0.8442	0.7675	0.0375	0.0625	0.0675	0.0625	0.095
No log	13.98	233	0.8808	0.7725	0.0425	0.0625	0.065	0.0625	0.105
No log	15.0	250	0.8276	0.7725	0.0375	0.0625	0.0675	0.0625	0.1
No log	15.96	266	0.8572	0.775	0.0425	0.0625	0.06	0.0625	0.11
No log	16.98	283	0.8874	0.78	0.045	0.0625	0.0625	0.0625	0.1125
No log	18.0	300	1.0065	0.7575	0.0475	0.0625	0.0625	0.0625	0.14
No log	18.96	316	0.9279	0.775	0.0475	0.0625	0.07	0.0625	0.12
No log	19.98	333	0.9474	0.76	0.05	0.0625	0.075	0.0625	0.1475
No log	21.0	350	0.9407	0.775	0.0375	0.0625	0.07	0.0625	0.1475
No log	21.96	366	0.9644	0.78	0.04	0.0625	0.065	0.0625	0.18
No log	22.98	383	0.9690	0.7825	0.04	0.0625	0.07	0.0625	0.1925
No log	24.0	400	0.9678	0.7825	0.045	0.0625	0.07	0.0625	0.2325
No log	24.96	416	0.9804	0.785	0.0425	0.0625	0.07	0.0625	0.2625
No log	25.98	433	0.9877	0.785	0.0475	0.0625	0.075	0.0625	0.3225
No log	27.0	450	0.9941	0.79	0.0475	0.0625	0.0725	0.0625	0.3825
No log	27.96	466	1.0016	0.7875	0.0475	0.0625	0.075	0.0625	0.4425
No log	28.98	483	1.0051	0.7875	0.0475	0.0625	0.0725	0.0625	0.4825
0.3531	30.0	500	0.9991	0.7875	0.05	0.0625	0.0725	0.0625	0.4975
0.3531	30.96	516	1.0213	0.785	0.0525	0.0625	0.0675	0.0625	0.5075
0.3531	31.98	533	1.0350	0.785	0.0525	0.0625	0.0675	0.0625	0.51
0.3531	33.0	550	1.0285	0.7825	0.0525	0.0625	0.07	0.0625	0.5125
0.3531	33.96	566	1.0311	0.79	0.0525	0.0625	0.07	0.0625	0.505
0.3531	34.98	583	1.0463	0.7875	0.0525	0.0625	0.07	0.0625	0.5125
0.3531	36.0	600	1.0501	0.785	0.0525	0.0625	0.07	0.0625	0.515
0.3531	36.96	616	1.0494	0.7825	0.0525	0.0625	0.07	0.0625	0.5225
0.3531	37.98	633	1.0574	0.7875	0.0525	0.0625	0.07	0.0625	0.515
0.3531	39.0	650	1.0625	0.78	0.0525	0.0625	0.0675	0.0625	0.51
0.3531	39.96	666	1.0643	0.78	0.0525	0.0625	0.07	0.0625	0.5075
0.3531	40.98	683	1.0679	0.7825	0.0525	0.0625	0.07	0.0625	0.515
0.3531	42.0	700	1.0690	0.7825	0.0525	0.0625	0.07	0.0625	0.5175
0.3531	42.96	716	1.0682	0.7825	0.0525	0.0625	0.07	0.0625	0.5225
0.3531	43.98	733	1.0720	0.785	0.0525	0.0625	0.07	0.0625	0.5175
0.3531	45.0	750	1.0744	0.7875	0.0525	0.0625	0.07	0.0625	0.5225
0.3531	45.96	766	1.0801	0.7875	0.0525	0.0625	0.07	0.0625	0.52
0.3531	46.98	783	1.0864	0.78	0.0525	0.0625	0.07	0.0625	0.52
0.3531	48.0	800	1.0831	0.7825	0.0525	0.0625	0.075	0.0625	0.5175
0.3531	48.96	816	1.0800	0.785	0.0525	0.0625	0.0775	0.0625	0.5275
0.3531	49.98	833	1.0840	0.78	0.0525	0.0625	0.0775	0.0625	0.5225
0.3531	51.0	850	1.0863	0.78	0.0525	0.0625	0.075	0.0625	0.5225
0.3531	51.96	866	1.0887	0.78	0.0525	0.0625	0.0725	0.0625	0.5275
0.3531	52.98	883	1.0904	0.785	0.0525	0.0625	0.0725	0.0625	0.525
0.3531	54.0	900	1.0910	0.785	0.0525	0.0625	0.0725	0.0625	0.525
0.3531	54.96	916	1.0922	0.785	0.0525	0.0625	0.0725	0.0625	0.52
0.3531	55.98	933	1.0937	0.785	0.0525	0.0625	0.07	0.0625	0.52
0.3531	57.0	950	1.0946	0.785	0.0525	0.0625	0.0725	0.0625	0.52
0.3531	57.6	960	1.0946	0.785	0.0525	0.0625	0.0725	0.0625	0.52

Framework versions

• Transformers 4.31.0
• Pytorch 2.0.1+cu117
• Datasets 2.13.1
• Tokenizers 0.13.3