layoutlm-funsd-tf

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

Model description

More information needed

More information needed

More information needed

The following hyperparameters were used during training:

optimizer: {'name': 'AdamWeightDecay', 'learning_rate': 2e-05, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False, 'weight_decay_rate': 0.01}
training_precision: mixed_float16

Train Loss	Validation Loss	Train Overall Precision	Train Overall Recall	Train Overall F1	Train Overall Accuracy	Epoch
1.7401	1.5464	0.0930	0.1174	0.1038	0.3843	0
1.4833	1.3301	0.2414	0.3964	0.3000	0.4268	1
1.2693	1.2622	0.2985	0.4947	0.3724	0.4693	2
1.1369	1.0729	0.3617	0.4887	0.4157	0.5902	3
1.0364	1.1800	0.3293	0.5073	0.3994	0.5604	4
0.9327	1.2033	0.3938	0.5268	0.4507	0.5683	5
0.8211	1.0876	0.4192	0.5153	0.4623	0.6004	6
0.7265	1.0982	0.4480	0.5334	0.4869	0.6102	7
0.6561	1.1134	0.4490	0.5650	0.5003	0.6192	8
0.5783	1.0834	0.4764	0.5630	0.5161	0.6317	9
0.5160	1.1453	0.4504	0.5494	0.4950	0.6227	10
0.4714	1.1865	0.4873	0.5981	0.5371	0.6277	11
0.4340	1.2212	0.4972	0.5805	0.5356	0.6318	12
0.3990	1.2407	0.4913	0.6212	0.5486	0.6334	13
0.3743	1.2597	0.5173	0.5986	0.5550	0.6338	14
0.3454	1.2205	0.5157	0.6106	0.5592	0.6406	15
0.3276	1.3600	0.5186	0.6001	0.5564	0.6318	16
0.3013	1.6473	0.4805	0.5745	0.5233	0.5899	17
0.3093	1.2595	0.4957	0.5735	0.5318	0.6389	18
0.2577	1.4449	0.4772	0.5675	0.5185	0.6076	19
0.2301	1.4514	0.4790	0.5620	0.5172	0.6205	20
0.2118	1.4575	0.5255	0.5991	0.5599	0.6305	21
0.1845	1.4446	0.5270	0.6076	0.5644	0.6353	22
0.1698	1.4538	0.5428	0.6011	0.5705	0.6423	23
0.1606	1.4318	0.5131	0.5720	0.5409	0.6361	24
0.1538	1.4257	0.5310	0.6061	0.5661	0.6484	25
0.1403	1.5233	0.5232	0.6061	0.5616	0.6428	26
0.1229	1.4796	0.5547	0.6131	0.5825	0.6471	27
0.1225	1.5841	0.5239	0.5946	0.5570	0.6101	28
0.1085	1.5432	0.5253	0.6046	0.5622	0.6423	29
0.1025	1.5414	0.5176	0.5966	0.5543	0.6312	30
0.0961	1.5766	0.5302	0.6121	0.5682	0.6392	31