EElayoutlmv3_jordyvl_rvl_cdip_100_examples_per_class_2023-08-22_text_vision_enc_5_6_7_8_ramp

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: 1.5033
Accuracy: 0.76
Exit 0 Accuracy: 0.075
Exit 1 Accuracy: 0.095
Exit 2 Accuracy: 0.73
Exit 3 Accuracy: 0.74
Exit 4 Accuracy: 0.7425
Exit 5 Accuracy: 0.7525

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	Exit 5 Accuracy
No log	0.96	16	2.6757	0.19	0.0775	0.0925	0.115	0.0675	0.0625	0.0625
No log	1.98	33	2.4683	0.275	0.075	0.0925	0.1075	0.07	0.0625	0.0625
No log	3.0	50	2.2534	0.3525	0.0775	0.095	0.1175	0.0725	0.0625	0.0625
No log	3.96	66	1.9728	0.4225	0.0775	0.095	0.1375	0.1325	0.0675	0.0625
No log	4.98	83	1.6809	0.565	0.08	0.095	0.1625	0.1475	0.0775	0.0625
No log	6.0	100	1.5205	0.6075	0.0775	0.095	0.1725	0.155	0.085	0.0625
No log	6.96	116	1.3038	0.6825	0.0775	0.095	0.1775	0.195	0.11	0.0625
No log	7.98	133	1.1279	0.7125	0.075	0.095	0.1925	0.2175	0.2275	0.0625
No log	9.0	150	1.0389	0.7175	0.0775	0.095	0.1975	0.25	0.385	0.0625
No log	9.96	166	0.9649	0.7525	0.075	0.095	0.19	0.2975	0.42	0.0625
No log	10.98	183	0.9956	0.725	0.0775	0.095	0.2325	0.3425	0.52	0.0625
No log	12.0	200	0.8946	0.7575	0.075	0.095	0.23	0.3675	0.555	0.0875
No log	12.96	216	0.9136	0.735	0.0775	0.095	0.275	0.4325	0.57	0.0625
No log	13.98	233	0.8954	0.7725	0.0775	0.095	0.3125	0.4625	0.57	0.1075
No log	15.0	250	0.9159	0.7575	0.075	0.095	0.3325	0.4925	0.5925	0.235
No log	15.96	266	0.9258	0.7775	0.08	0.095	0.3675	0.5175	0.6175	0.34
No log	16.98	283	0.9848	0.7625	0.0775	0.095	0.46	0.5525	0.6375	0.5375
No log	18.0	300	1.0248	0.7575	0.0775	0.095	0.505	0.5925	0.65	0.6275
No log	18.96	316	1.0362	0.7675	0.0775	0.095	0.5325	0.6275	0.6525	0.675
No log	19.98	333	1.0724	0.7475	0.0775	0.095	0.58	0.6625	0.6575	0.6975
No log	21.0	350	1.1489	0.7425	0.0775	0.095	0.6	0.675	0.6925	0.715
No log	21.96	366	1.1769	0.755	0.0775	0.095	0.5725	0.6675	0.68	0.7025
No log	22.98	383	1.1129	0.775	0.08	0.095	0.6125	0.685	0.695	0.7375
No log	24.0	400	1.2476	0.75	0.0775	0.095	0.635	0.705	0.72	0.73
No log	24.96	416	1.1019	0.7825	0.0775	0.095	0.6375	0.7275	0.7225	0.755
No log	25.98	433	1.2046	0.78	0.0775	0.095	0.6425	0.7175	0.725	0.7425
No log	27.0	450	1.2260	0.7825	0.0775	0.095	0.66	0.71	0.74	0.7475
No log	27.96	466	1.2632	0.765	0.0775	0.095	0.6675	0.6975	0.7275	0.7475
No log	28.98	483	1.2968	0.775	0.0775	0.095	0.665	0.72	0.7275	0.7325
1.4887	30.0	500	1.3257	0.7675	0.0825	0.095	0.6975	0.705	0.7375	0.7475
1.4887	30.96	516	1.2935	0.7725	0.0775	0.095	0.69	0.7175	0.7525	0.76
1.4887	31.98	533	1.3605	0.7675	0.08	0.095	0.7	0.7275	0.755	0.7525
1.4887	33.0	550	1.3775	0.765	0.08	0.095	0.7	0.7275	0.755	0.74
1.4887	33.96	566	1.3881	0.765	0.0775	0.095	0.6975	0.7225	0.7475	0.755
1.4887	34.98	583	1.3496	0.77	0.075	0.095	0.715	0.735	0.7575	0.76
1.4887	36.0	600	1.4044	0.7725	0.075	0.095	0.7075	0.7375	0.75	0.75
1.4887	36.96	616	1.4085	0.775	0.075	0.095	0.72	0.7325	0.745	0.755
1.4887	37.98	633	1.4755	0.7675	0.075	0.095	0.7225	0.7375	0.75	0.7475
1.4887	39.0	650	1.4363	0.7775	0.0775	0.095	0.725	0.7425	0.755	0.7475
1.4887	39.96	666	1.4532	0.775	0.08	0.095	0.72	0.74	0.745	0.7525
1.4887	40.98	683	1.4626	0.7725	0.075	0.095	0.73	0.7325	0.75	0.7525
1.4887	42.0	700	1.4340	0.785	0.0775	0.095	0.725	0.7425	0.7525	0.765
1.4887	42.96	716	1.4723	0.7725	0.0775	0.095	0.7325	0.735	0.75	0.755
1.4887	43.98	733	1.4687	0.7725	0.0775	0.095	0.7275	0.7475	0.75	0.7625
1.4887	45.0	750	1.4757	0.7625	0.0775	0.095	0.7225	0.7325	0.7375	0.74
1.4887	45.96	766	1.4801	0.77	0.08	0.095	0.7225	0.745	0.7525	0.76
1.4887	46.98	783	1.5045	0.76	0.0775	0.095	0.725	0.735	0.7425	0.75
1.4887	48.0	800	1.5106	0.76	0.0775	0.095	0.7275	0.74	0.7525	0.745
1.4887	48.96	816	1.4946	0.765	0.0775	0.095	0.73	0.745	0.745	0.755
1.4887	49.98	833	1.4739	0.765	0.0775	0.095	0.7325	0.745	0.7525	0.755
1.4887	51.0	850	1.4943	0.765	0.0775	0.095	0.7325	0.74	0.745	0.755
1.4887	51.96	866	1.5010	0.7625	0.075	0.095	0.7275	0.74	0.755	0.76
1.4887	52.98	883	1.4976	0.765	0.075	0.095	0.73	0.74	0.745	0.76
1.4887	54.0	900	1.5025	0.76	0.075	0.095	0.725	0.7375	0.74	0.7525
1.4887	54.96	916	1.5016	0.76	0.075	0.095	0.73	0.7375	0.745	0.75
1.4887	55.98	933	1.5024	0.76	0.075	0.095	0.7275	0.7375	0.7425	0.7525
1.4887	57.0	950	1.5025	0.76	0.075	0.095	0.73	0.74	0.7425	0.75
1.4887	57.6	960	1.5033	0.76	0.075	0.095	0.73	0.74	0.7425	0.7525

Framework versions

Transformers 4.31.0
Pytorch 2.0.1+cu117
Datasets 2.13.1
Tokenizers 0.13.3

EElayoutlmv3_jordyvl_rvl_cdip_100_examples_per_class_2023-08-22_text_vision_enc_5_6_7_8_ramp

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

NSDT 3DConvert

UnrealSynth

DreamTexture.js