layoutlm-captive-corp-7

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

• Loss: 0.7263
• Precision: 0.8352
• Recall: 0.8352
• F1: 0.8352
• Accuracy: 0.8812

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: 3e-05
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 150
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.3675	1.0	4	0.9283	0.7847	0.8052	0.7948	0.8423
0.3555	2.0	8	0.9010	0.7897	0.8015	0.7955	0.8510
0.3296	3.0	12	0.8976	0.7941	0.8090	0.8015	0.8402
0.3007	4.0	16	0.8562	0.7955	0.8015	0.7985	0.8488
0.2795	5.0	20	0.8642	0.7889	0.7978	0.7933	0.8467
0.2604	6.0	24	0.8300	0.7810	0.8015	0.7911	0.8531
0.2464	7.0	28	0.8305	0.7904	0.8052	0.7978	0.8488
0.2295	8.0	32	0.8205	0.7868	0.8015	0.7941	0.8423
0.2127	9.0	36	0.8065	0.7897	0.8015	0.7955	0.8488
0.1961	10.0	40	0.7998	0.8060	0.8090	0.8075	0.8531
0.185	11.0	44	0.7846	0.7941	0.8090	0.8015	0.8553
0.1756	12.0	48	0.7772	0.8015	0.8165	0.8089	0.8575
0.1636	13.0	52	0.7772	0.8118	0.8240	0.8178	0.8618
0.1567	14.0	56	0.7785	0.7978	0.8127	0.8052	0.8575
0.1474	15.0	60	0.7634	0.7941	0.8090	0.8015	0.8618
0.1383	16.0	64	0.7435	0.8007	0.8127	0.8067	0.8618
0.1313	17.0	68	0.7465	0.7978	0.8127	0.8052	0.8618
0.1231	18.0	72	0.7526	0.8051	0.8202	0.8126	0.8661
0.1197	19.0	76	0.7386	0.8015	0.8165	0.8089	0.8683
0.1143	20.0	80	0.7419	0.8044	0.8165	0.8104	0.8661
0.108	21.0	84	0.7442	0.8037	0.8127	0.8082	0.8575
0.1025	22.0	88	0.7415	0.8044	0.8165	0.8104	0.8618
0.0994	23.0	92	0.7392	0.8015	0.8165	0.8089	0.8639
0.0941	24.0	96	0.7372	0.7985	0.8165	0.8074	0.8639
0.0905	25.0	100	0.7424	0.8044	0.8165	0.8104	0.8639
0.0879	26.0	104	0.7349	0.8007	0.8127	0.8067	0.8618
0.0842	27.0	108	0.7296	0.7978	0.8127	0.8052	0.8639
0.0805	28.0	112	0.7339	0.7970	0.8090	0.8030	0.8596
0.0785	29.0	116	0.7405	0.8	0.8090	0.8045	0.8553
0.0759	30.0	120	0.7424	0.7970	0.8090	0.8030	0.8596
0.0726	31.0	124	0.7329	0.8044	0.8165	0.8104	0.8596
0.0703	32.0	128	0.7289	0.8015	0.8165	0.8089	0.8618
0.0681	33.0	132	0.7204	0.8022	0.8202	0.8111	0.8661
0.0663	34.0	136	0.7168	0.8015	0.8165	0.8089	0.8639
0.0619	35.0	140	0.7244	0.7978	0.8127	0.8052	0.8661
0.0614	36.0	144	0.7360	0.7978	0.8127	0.8052	0.8618
0.0594	37.0	148	0.7306	0.8044	0.8165	0.8104	0.8618
0.0586	38.0	152	0.7177	0.8081	0.8202	0.8141	0.8639
0.0562	39.0	156	0.7133	0.8088	0.8240	0.8163	0.8704
0.0557	40.0	160	0.7229	0.7978	0.8127	0.8052	0.8639
0.0558	41.0	164	0.7244	0.8044	0.8165	0.8104	0.8661
0.0513	42.0	168	0.7180	0.8044	0.8165	0.8104	0.8704
0.0515	43.0	172	0.7166	0.8007	0.8127	0.8067	0.8661
0.0496	44.0	176	0.7186	0.8104	0.8165	0.8134	0.8683
0.049	45.0	180	0.7165	0.8111	0.8202	0.8156	0.8661
0.0488	46.0	184	0.7139	0.8111	0.8202	0.8156	0.8683
0.0463	47.0	188	0.7199	0.8015	0.8165	0.8089	0.8639
0.0449	48.0	192	0.7257	0.8015	0.8165	0.8089	0.8618
0.0452	49.0	196	0.7231	0.8015	0.8165	0.8089	0.8618
0.0437	50.0	200	0.7161	0.8081	0.8202	0.8141	0.8683
0.0428	51.0	204	0.7125	0.8015	0.8165	0.8089	0.8683
0.0423	52.0	208	0.7170	0.8104	0.8165	0.8134	0.8661
0.0417	53.0	212	0.7242	0.8141	0.8202	0.8172	0.8661
0.0397	54.0	216	0.7269	0.8141	0.8202	0.8172	0.8661
0.0399	55.0	220	0.7239	0.8044	0.8165	0.8104	0.8661
0.0384	56.0	224	0.7246	0.8044	0.8165	0.8104	0.8704
0.0378	57.0	228	0.7237	0.8111	0.8202	0.8156	0.8726
0.0367	58.0	232	0.7200	0.8111	0.8202	0.8156	0.8726
0.037	59.0	236	0.7183	0.8111	0.8202	0.8156	0.8726
0.0356	60.0	240	0.7187	0.7978	0.8127	0.8052	0.8683
0.035	61.0	244	0.7161	0.8007	0.8127	0.8067	0.8661
0.0347	62.0	248	0.7148	0.8104	0.8165	0.8134	0.8683
0.0339	63.0	252	0.7195	0.8141	0.8202	0.8172	0.8704
0.0337	64.0	256	0.7236	0.8141	0.8202	0.8172	0.8704
0.033	65.0	260	0.7208	0.8246	0.8277	0.8262	0.8747
0.0324	66.0	264	0.7159	0.8178	0.8240	0.8209	0.8747
0.032	67.0	268	0.7122	0.8118	0.8240	0.8178	0.8747
0.0315	68.0	272	0.7116	0.8229	0.8352	0.8290	0.8790
0.0313	69.0	276	0.7154	0.8125	0.8277	0.8200	0.8747
0.0307	70.0	280	0.7197	0.8192	0.8315	0.8253	0.8769
0.0306	71.0	284	0.7214	0.8192	0.8315	0.8253	0.8769
0.0302	72.0	288	0.7223	0.8192	0.8315	0.8253	0.8769
0.0289	73.0	292	0.7191	0.8192	0.8315	0.8253	0.8769
0.0292	74.0	296	0.7170	0.8185	0.8277	0.8231	0.8769
0.0284	75.0	300	0.7172	0.8148	0.8240	0.8194	0.8747
0.0283	76.0	304	0.7196	0.8111	0.8202	0.8156	0.8726
0.0277	77.0	308	0.7212	0.8044	0.8165	0.8104	0.8704
0.0278	78.0	312	0.7230	0.8141	0.8202	0.8172	0.8726
0.027	79.0	316	0.7223	0.8178	0.8240	0.8209	0.8747
0.0269	80.0	320	0.7190	0.8253	0.8315	0.8284	0.8790
0.0267	81.0	324	0.7187	0.8253	0.8315	0.8284	0.8790
0.0262	82.0	328	0.7224	0.8216	0.8277	0.8246	0.8769
0.0261	83.0	332	0.7262	0.8216	0.8277	0.8246	0.8769
0.0259	84.0	336	0.7279	0.8141	0.8202	0.8172	0.8726
0.0257	85.0	340	0.7274	0.8141	0.8202	0.8172	0.8726
0.0251	86.0	344	0.7267	0.8141	0.8202	0.8172	0.8726
0.0252	87.0	348	0.7252	0.8253	0.8315	0.8284	0.8790
0.024	88.0	352	0.7251	0.8284	0.8315	0.8299	0.8790
0.0242	89.0	356	0.7263	0.8352	0.8352	0.8352	0.8812
0.0244	90.0	360	0.7268	0.8352	0.8352	0.8352	0.8812
0.0241	91.0	364	0.7275	0.8352	0.8352	0.8352	0.8812
0.0233	92.0	368	0.7279	0.8315	0.8315	0.8315	0.8790
0.0233	93.0	372	0.7301	0.8246	0.8277	0.8262	0.8769
0.0231	94.0	376	0.7315	0.8209	0.8240	0.8224	0.8747
0.0234	95.0	380	0.7322	0.8209	0.8240	0.8224	0.8747
0.023	96.0	384	0.7304	0.8209	0.8240	0.8224	0.8747
0.0222	97.0	388	0.7270	0.8209	0.8240	0.8224	0.8747
0.0221	98.0	392	0.7237	0.8246	0.8277	0.8262	0.8769
0.0223	99.0	396	0.7216	0.8246	0.8277	0.8262	0.8769
0.0222	100.0	400	0.7210	0.8246	0.8277	0.8262	0.8769
0.0218	101.0	404	0.7224	0.8284	0.8315	0.8299	0.8790
0.0223	102.0	408	0.7235	0.8284	0.8315	0.8299	0.8790
0.0216	103.0	412	0.7239	0.8284	0.8315	0.8299	0.8790
0.0213	104.0	416	0.7248	0.8284	0.8315	0.8299	0.8790
0.0213	105.0	420	0.7284	0.8284	0.8315	0.8299	0.8790
0.0212	106.0	424	0.7316	0.8246	0.8277	0.8262	0.8747
0.0207	107.0	428	0.7331	0.8246	0.8277	0.8262	0.8747
0.0206	108.0	432	0.7329	0.8315	0.8315	0.8315	0.8769
0.0208	109.0	436	0.7325	0.8352	0.8352	0.8352	0.8790
0.0208	110.0	440	0.7315	0.8315	0.8315	0.8315	0.8769
0.0205	111.0	444	0.7305	0.8352	0.8352	0.8352	0.8790
0.0202	112.0	448	0.7284	0.8352	0.8352	0.8352	0.8790
0.0202	113.0	452	0.7284	0.8253	0.8315	0.8284	0.8790
0.0201	114.0	456	0.7285	0.8284	0.8315	0.8299	0.8769
0.0199	115.0	460	0.7308	0.8284	0.8315	0.8299	0.8769
0.0196	116.0	464	0.7346	0.8284	0.8315	0.8299	0.8769
0.0197	117.0	468	0.7371	0.8284	0.8315	0.8299	0.8747
0.0195	118.0	472	0.7384	0.8284	0.8315	0.8299	0.8747
0.0197	119.0	476	0.7383	0.8284	0.8315	0.8299	0.8747
0.0198	120.0	480	0.7379	0.8284	0.8315	0.8299	0.8769
0.0194	121.0	484	0.7372	0.8216	0.8277	0.8246	0.8747
0.0192	122.0	488	0.7364	0.8216	0.8277	0.8246	0.8747
0.0192	123.0	492	0.7350	0.8216	0.8277	0.8246	0.8747
0.0192	124.0	496	0.7344	0.8216	0.8277	0.8246	0.8747
0.0189	125.0	500	0.7340	0.8216	0.8277	0.8246	0.8769
0.0185	126.0	504	0.7343	0.8216	0.8277	0.8246	0.8769
0.0184	127.0	508	0.7343	0.8284	0.8315	0.8299	0.8790
0.0188	128.0	512	0.7334	0.8284	0.8315	0.8299	0.8790
0.0188	129.0	516	0.7328	0.8284	0.8315	0.8299	0.8790
0.0187	130.0	520	0.7331	0.8284	0.8315	0.8299	0.8790
0.0188	131.0	524	0.7324	0.8284	0.8315	0.8299	0.8790
0.0188	132.0	528	0.7325	0.8284	0.8315	0.8299	0.8790
0.0187	133.0	532	0.7317	0.8284	0.8315	0.8299	0.8790
0.0186	134.0	536	0.7310	0.8284	0.8315	0.8299	0.8790
0.0184	135.0	540	0.7308	0.8284	0.8315	0.8299	0.8790
0.0182	136.0	544	0.7307	0.8284	0.8315	0.8299	0.8790
0.018	137.0	548	0.7312	0.8284	0.8315	0.8299	0.8790
0.0181	138.0	552	0.7324	0.8284	0.8315	0.8299	0.8790
0.0182	139.0	556	0.7333	0.8284	0.8315	0.8299	0.8790
0.0182	140.0	560	0.7344	0.8284	0.8315	0.8299	0.8790
0.0184	141.0	564	0.7351	0.8284	0.8315	0.8299	0.8790
0.0182	142.0	568	0.7358	0.8284	0.8315	0.8299	0.8790
0.0179	143.0	572	0.7358	0.8284	0.8315	0.8299	0.8790
0.0178	144.0	576	0.7357	0.8284	0.8315	0.8299	0.8790
0.0183	145.0	580	0.7356	0.8284	0.8315	0.8299	0.8790
0.0183	146.0	584	0.7355	0.8284	0.8315	0.8299	0.8790
0.0177	147.0	588	0.7354	0.8284	0.8315	0.8299	0.8790
0.0179	148.0	592	0.7355	0.8284	0.8315	0.8299	0.8790
0.0183	149.0	596	0.7355	0.8284	0.8315	0.8299	0.8790
0.0179	150.0	600	0.7355	0.8284	0.8315	0.8299	0.8790

Framework versions

• Transformers 4.47.0.dev0
• Pytorch 2.5.1+cu121
• Datasets 3.1.0
• Tokenizers 0.20.3