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

Loss: 1.7000
Precision: 0.9130
Recall: 0.9180
F1: 0.9155
Accuracy: 0.8387

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: 1e-05
train_batch_size: 2
eval_batch_size: 2
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
training_steps: 20000

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.33	100	0.6462	0.7683	0.8137	0.7903	0.7808
No log	2.67	200	0.4416	0.8248	0.8887	0.8556	0.8444
No log	4.0	300	0.5276	0.8563	0.9061	0.8805	0.8351
No log	5.33	400	0.5038	0.8319	0.8922	0.8610	0.8317
0.5539	6.67	500	0.5489	0.8864	0.9151	0.9005	0.8583
0.5539	8.0	600	0.5947	0.8646	0.9165	0.8898	0.8383
0.5539	9.33	700	0.7367	0.8796	0.8857	0.8827	0.8362
0.5539	10.67	800	0.8299	0.8911	0.9230	0.9068	0.8350
0.5539	12.0	900	0.6754	0.8934	0.9116	0.9024	0.8465
0.1203	13.33	1000	0.8242	0.8814	0.9011	0.8912	0.8420
0.1203	14.67	1100	0.9349	0.8835	0.8857	0.8846	0.8208
0.1203	16.0	1200	1.0205	0.8853	0.8783	0.8818	0.8131
0.1203	17.33	1300	0.8790	0.8865	0.8962	0.8913	0.8542
0.1203	18.67	1400	0.9262	0.8870	0.8967	0.8918	0.8482
0.0522	20.0	1500	0.9744	0.8979	0.8952	0.8965	0.8323
0.0522	21.33	1600	1.0198	0.8976	0.9146	0.9060	0.8401
0.0522	22.67	1700	1.0466	0.9114	0.9146	0.9130	0.8368
0.0522	24.0	1800	0.9874	0.8944	0.9086	0.9014	0.8462
0.0522	25.33	1900	1.1240	0.9051	0.9141	0.9095	0.8364
0.0208	26.67	2000	1.0658	0.9021	0.9205	0.9112	0.8399
0.0208	28.0	2100	1.2349	0.8934	0.9116	0.9024	0.8227
0.0208	29.33	2200	1.2906	0.8930	0.9081	0.9005	0.8111
0.0208	30.67	2300	1.2133	0.9020	0.9096	0.9058	0.8398
0.0208	32.0	2400	1.2202	0.9055	0.9096	0.9076	0.8375
0.0154	33.33	2500	1.2454	0.8909	0.9131	0.9019	0.8393
0.0154	34.67	2600	1.2065	0.9058	0.9126	0.9092	0.8364
0.0154	36.0	2700	1.2165	0.8998	0.9096	0.9046	0.8284
0.0154	37.33	2800	1.3195	0.8913	0.9160	0.9035	0.8272
0.0154	38.67	2900	1.3834	0.8957	0.9086	0.9021	0.8240
0.007	40.0	3000	1.3035	0.9001	0.9086	0.9043	0.8306
0.007	41.33	3100	1.3553	0.8991	0.9121	0.9055	0.8241
0.007	42.67	3200	1.3227	0.9054	0.9036	0.9045	0.8343
0.007	44.0	3300	1.2750	0.9262	0.9170	0.9216	0.8460
0.007	45.33	3400	1.2563	0.8980	0.9101	0.9040	0.8408
0.0035	46.67	3500	1.2013	0.9015	0.9136	0.9075	0.8383
0.0035	48.0	3600	1.2035	0.8997	0.9180	0.9088	0.8487
0.0035	49.33	3700	1.3997	0.9206	0.9041	0.9123	0.8312
0.0035	50.67	3800	1.3818	0.9117	0.9180	0.9149	0.8337
0.0035	52.0	3900	1.3568	0.9045	0.9086	0.9066	0.8410
0.0057	53.33	4000	1.3255	0.8935	0.9170	0.9051	0.8313
0.0057	54.67	4100	1.2954	0.8950	0.9151	0.9049	0.8343
0.0057	56.0	4200	1.4201	0.9028	0.9141	0.9084	0.8234
0.0057	57.33	4300	1.3858	0.9016	0.9061	0.9039	0.8238
0.0057	58.67	4400	1.3748	0.8976	0.9056	0.9016	0.8286
0.0155	60.0	4500	1.4346	0.9024	0.9051	0.9038	0.8274
0.0155	61.33	4600	1.4059	0.9073	0.9136	0.9104	0.8324
0.0155	62.67	4700	1.3779	0.9076	0.9121	0.9098	0.8358
0.0155	64.0	4800	1.3785	0.9064	0.9235	0.9149	0.8389
0.0155	65.33	4900	1.3995	0.9014	0.9220	0.9116	0.8190
0.0043	66.67	5000	1.2618	0.9087	0.9101	0.9094	0.8329
0.0043	68.0	5100	1.2878	0.9093	0.9165	0.9129	0.8526
0.0043	69.33	5200	1.4384	0.9063	0.9126	0.9094	0.8372
0.0043	70.67	5300	1.5029	0.9093	0.9165	0.9129	0.8347
0.0043	72.0	5400	1.4592	0.9107	0.9071	0.9089	0.8345
0.0045	73.33	5500	1.4604	0.9099	0.9136	0.9118	0.8263
0.0045	74.67	5600	1.5632	0.8933	0.9066	0.8999	0.8162
0.0045	76.0	5700	1.5839	0.9114	0.9096	0.9105	0.8335
0.0045	77.33	5800	1.5557	0.9099	0.9081	0.9090	0.8392
0.0045	78.67	5900	1.4348	0.9024	0.9051	0.9038	0.8266
0.0089	80.0	6000	1.2747	0.9026	0.9160	0.9093	0.8429
0.0089	81.33	6100	1.3560	0.8963	0.9230	0.9094	0.8376
0.0089	82.67	6200	1.2859	0.8987	0.9165	0.9075	0.8480
0.0089	84.0	6300	1.3389	0.9032	0.9220	0.9125	0.8315
0.0089	85.33	6400	1.3922	0.8922	0.9126	0.9023	0.8413
0.0045	86.67	6500	1.3723	0.9003	0.9066	0.9035	0.8353
0.0045	88.0	6600	1.3220	0.9011	0.9141	0.9075	0.8441
0.0045	89.33	6700	1.4433	0.9063	0.9126	0.9094	0.8213
0.0045	90.67	6800	1.5350	0.8977	0.9195	0.9085	0.8266
0.0045	92.0	6900	1.3681	0.9115	0.9215	0.9165	0.8467
0.002	93.33	7000	1.3141	0.8981	0.9240	0.9109	0.8398
0.002	94.67	7100	1.3836	0.9069	0.9146	0.9107	0.8310
0.002	96.0	7200	1.4456	0.8995	0.9155	0.9074	0.8310
0.002	97.33	7300	1.4218	0.9018	0.9031	0.9025	0.8304
0.002	98.67	7400	1.5428	0.8958	0.9225	0.9090	0.8252
0.0016	100.0	7500	1.4982	0.8974	0.9215	0.9093	0.8280
0.0016	101.33	7600	1.4787	0.9025	0.9240	0.9131	0.8330
0.0016	102.67	7700	1.5694	0.8982	0.9026	0.9004	0.8148
0.0016	104.0	7800	1.4361	0.8985	0.9146	0.9065	0.8244
0.0016	105.33	7900	1.5643	0.8912	0.9280	0.9092	0.8265
0.0065	106.67	8000	1.5890	0.9017	0.9155	0.9086	0.8269
0.0065	108.0	8100	1.5755	0.8901	0.9170	0.9034	0.8209
0.0065	109.33	8200	1.7716	0.9006	0.9051	0.9029	0.8105
0.0065	110.67	8300	1.6814	0.8973	0.9160	0.9066	0.8168
0.0065	112.0	8400	1.5316	0.9002	0.9230	0.9115	0.8199
0.0012	113.33	8500	1.5376	0.9041	0.9136	0.9088	0.8338
0.0012	114.67	8600	1.5773	0.9085	0.9175	0.9130	0.8334
0.0012	116.0	8700	1.5998	0.9050	0.9086	0.9068	0.8380
0.0012	117.33	8800	1.6401	0.8985	0.9101	0.9042	0.8298
0.0012	118.67	8900	1.6894	0.9055	0.8897	0.8975	0.8248
0.0011	120.0	9000	1.6945	0.9004	0.9031	0.9018	0.8275
0.0011	121.33	9100	1.5659	0.9038	0.9240	0.9138	0.8332
0.0011	122.67	9200	1.5270	0.8947	0.9240	0.9091	0.8358
0.0011	124.0	9300	1.5225	0.9081	0.9230	0.9155	0.8304
0.0011	125.33	9400	1.6064	0.8906	0.9141	0.9022	0.8232
0.0035	126.67	9500	1.5898	0.9034	0.9240	0.9136	0.8294
0.0035	128.0	9600	1.5404	0.8949	0.9225	0.9085	0.8336
0.0035	129.33	9700	1.4890	0.9074	0.9250	0.9161	0.8460
0.0035	130.67	9800	1.5620	0.9049	0.9175	0.9112	0.8315
0.0035	132.0	9900	1.5565	0.9050	0.9180	0.9115	0.8279
0.0014	133.33	10000	1.5553	0.8989	0.9230	0.9108	0.8424
0.0014	134.67	10100	1.5287	0.9060	0.9195	0.9127	0.8356
0.0014	136.0	10200	1.5282	0.9109	0.9146	0.9127	0.8398
0.0014	137.33	10300	1.5280	0.9073	0.9141	0.9107	0.8437
0.0014	138.67	10400	1.5719	0.9092	0.9151	0.9121	0.8387
0.0035	140.0	10500	1.5059	0.9074	0.9155	0.9115	0.8426
0.0035	141.33	10600	1.5702	0.9013	0.9250	0.9130	0.8355
0.0035	142.67	10700	1.5080	0.9035	0.9260	0.9146	0.8455
0.0035	144.0	10800	1.4643	0.9097	0.9255	0.9175	0.8467
0.0035	145.33	10900	1.5316	0.9037	0.9230	0.9132	0.8387
0.0011	146.67	11000	1.5314	0.9114	0.9195	0.9154	0.8392
0.0011	148.0	11100	1.4988	0.9114	0.9200	0.9157	0.8493
0.0011	149.33	11200	1.4546	0.9121	0.9275	0.9197	0.8538
0.0011	150.67	11300	1.5075	0.9062	0.9170	0.9116	0.8456
0.0011	152.0	11400	1.4556	0.8973	0.9076	0.9024	0.8393
0.0009	153.33	11500	1.5058	0.8911	0.9185	0.9046	0.8272
0.0009	154.67	11600	1.5903	0.9197	0.9101	0.9149	0.8318
0.0009	156.0	11700	1.5263	0.9164	0.9146	0.9155	0.8413
0.0009	157.33	11800	1.5729	0.9129	0.9160	0.9145	0.8386
0.0009	158.67	11900	1.5880	0.9086	0.9131	0.9108	0.8398
0.0009	160.0	12000	1.5907	0.9090	0.9126	0.9108	0.8399
0.0009	161.33	12100	1.5714	0.9111	0.9111	0.9111	0.8373
0.0009	162.67	12200	1.5848	0.9135	0.9126	0.9130	0.8378
0.0009	164.0	12300	1.5816	0.9112	0.9175	0.9144	0.8405
0.0009	165.33	12400	1.5425	0.9080	0.9121	0.9100	0.8386
0.0	166.67	12500	1.5837	0.9046	0.9136	0.9090	0.8362
0.0	168.0	12600	1.6781	0.9025	0.9195	0.9109	0.8290
0.0	169.33	12700	1.6219	0.9028	0.9185	0.9106	0.8326
0.0	170.67	12800	1.5786	0.9076	0.9126	0.9101	0.8380
0.0	172.0	12900	1.6212	0.9020	0.9146	0.9082	0.8322
0.0018	173.33	13000	1.6451	0.9086	0.9141	0.9113	0.8315
0.0018	174.67	13100	1.6730	0.9064	0.9185	0.9124	0.8293
0.0018	176.0	13200	1.6106	0.9026	0.9071	0.9049	0.8354
0.0018	177.33	13300	1.6403	0.9081	0.9180	0.9130	0.8402
0.0018	178.67	13400	1.6343	0.9043	0.9200	0.9121	0.8361
0.0012	180.0	13500	1.5853	0.9096	0.9195	0.9145	0.8431
0.0012	181.33	13600	1.5859	0.9101	0.9205	0.9153	0.8432
0.0012	182.67	13700	1.6137	0.9071	0.9215	0.9142	0.8394
0.0012	184.0	13800	1.6416	0.9002	0.9185	0.9093	0.8299
0.0012	185.33	13900	1.5497	0.9085	0.9126	0.9105	0.8457
0.0002	186.67	14000	1.6534	0.9015	0.9141	0.9077	0.8322
0.0002	188.0	14100	1.6003	0.9044	0.9116	0.9080	0.8290
0.0002	189.33	14200	1.5269	0.9046	0.9185	0.9115	0.8443
0.0002	190.67	14300	1.5977	0.9069	0.9141	0.9104	0.8360
0.0002	192.0	14400	1.5968	0.9090	0.9131	0.9110	0.8374
0.0004	193.33	14500	1.5945	0.9090	0.9131	0.9110	0.8376
0.0004	194.67	14600	1.6041	0.9117	0.9126	0.9121	0.8388
0.0004	196.0	14700	1.6038	0.9071	0.9121	0.9096	0.8362
0.0004	197.33	14800	1.6108	0.9059	0.9131	0.9095	0.8347
0.0004	198.67	14900	1.5873	0.9087	0.9151	0.9119	0.8393
0.0	200.0	15000	1.6047	0.9042	0.9146	0.9094	0.8417
0.0	201.33	15100	1.6125	0.9017	0.9155	0.9086	0.8348
0.0	202.67	15200	1.6191	0.9039	0.9155	0.9097	0.8360
0.0	204.0	15300	1.6626	0.9095	0.9131	0.9113	0.8320
0.0	205.33	15400	1.5967	0.9051	0.9190	0.9120	0.8427
0.0003	206.67	15500	1.5989	0.8982	0.9116	0.9048	0.8318
0.0003	208.0	15600	1.5990	0.8995	0.9116	0.9055	0.8307
0.0003	209.33	15700	1.6338	0.9043	0.9111	0.9077	0.8325
0.0003	210.67	15800	1.6390	0.9034	0.9101	0.9067	0.8329
0.0003	212.0	15900	1.6372	0.9015	0.9096	0.9055	0.8368
0.0001	213.33	16000	1.6020	0.9045	0.9131	0.9088	0.8378
0.0001	214.67	16100	1.5761	0.9071	0.9170	0.9121	0.8397
0.0001	216.0	16200	1.6536	0.9010	0.9131	0.9070	0.8293
0.0001	217.33	16300	1.6549	0.9023	0.9131	0.9077	0.8290
0.0001	218.67	16400	1.6737	0.8948	0.9081	0.9014	0.8292
0.0002	220.0	16500	1.6918	0.9106	0.9155	0.9131	0.8402
0.0002	221.33	16600	1.6726	0.9102	0.9165	0.9134	0.8379
0.0002	222.67	16700	1.6962	0.9121	0.9175	0.9148	0.8369
0.0002	224.0	16800	1.6974	0.9038	0.9146	0.9091	0.8367
0.0002	225.33	16900	1.7147	0.9126	0.9185	0.9156	0.8376
0.0006	226.67	17000	1.7000	0.9130	0.9180	0.9155	0.8387
0.0006	228.0	17100	1.6951	0.9083	0.9155	0.9119	0.8374
0.0006	229.33	17200	1.7014	0.9097	0.9160	0.9129	0.8369
0.0006	230.67	17300	1.7029	0.9102	0.9165	0.9134	0.8369
0.0006	232.0	17400	1.7039	0.9112	0.9170	0.9141	0.8374
0.0	233.33	17500	1.6516	0.9157	0.9116	0.9136	0.8355
0.0	234.67	17600	1.6536	0.9148	0.9126	0.9137	0.8348
0.0	236.0	17700	1.6548	0.9144	0.9131	0.9137	0.8348
0.0	237.33	17800	1.7110	0.9068	0.9185	0.9126	0.8360
0.0	238.67	17900	1.7115	0.9073	0.9185	0.9129	0.8362
0.0	240.0	18000	1.7124	0.9054	0.9180	0.9117	0.8362
0.0	241.33	18100	1.7146	0.9072	0.9175	0.9123	0.8376
0.0	242.67	18200	1.7217	0.9100	0.9141	0.9120	0.8317
0.0	244.0	18300	1.7225	0.9096	0.9146	0.9121	0.8315
0.0	245.33	18400	1.7159	0.9070	0.9155	0.9112	0.8323
0.0001	246.67	18500	1.7164	0.9074	0.9155	0.9115	0.8322
0.0001	248.0	18600	1.6927	0.9009	0.9165	0.9086	0.8326
0.0001	249.33	18700	1.6767	0.9034	0.9155	0.9094	0.8335
0.0001	250.67	18800	1.6773	0.9034	0.9155	0.9094	0.8335
0.0001	252.0	18900	1.6885	0.9029	0.9151	0.9090	0.8334
0.0002	253.33	19000	1.7032	0.9053	0.9165	0.9109	0.8312
0.0002	254.67	19100	1.7036	0.9057	0.9160	0.9108	0.8307
0.0002	256.0	19200	1.7041	0.9053	0.9160	0.9106	0.8310
0.0002	257.33	19300	1.7045	0.9053	0.9160	0.9106	0.8310
0.0002	258.67	19400	1.7049	0.9053	0.9160	0.9106	0.8310
0.0	260.0	19500	1.7069	0.9057	0.9165	0.9111	0.8310
0.0	261.33	19600	1.7062	0.9076	0.9170	0.9123	0.8312
0.0	262.67	19700	1.7071	0.9071	0.9170	0.9121	0.8312
0.0	264.0	19800	1.7083	0.9067	0.9170	0.9118	0.8313
0.0	265.33	19900	1.7084	0.9058	0.9170	0.9114	0.8316
0.0	266.67	20000	1.7086	0.9058	0.9170	0.9114	0.8316

Framework versions

Transformers 4.34.1
Pytorch 2.1.0+cu121
Datasets 2.14.5
Tokenizers 0.14.1

diegokauer/layoutlmv3-fintuned-funsd

layoutlmv3-fintuned-funsd

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

diegokauer/
layoutlmv3-fintuned-funsd