receipt-core-model

This model is a fine-tuned version of DanSarm/receipt-core-model on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.2194

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: 0.0001
• train_batch_size: 4
• eval_batch_size: 4
• 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: 1000

Training results

Training Loss	Epoch	Step	Validation Loss
0.1872	1.0	36	0.8766
0.1235	2.0	72	0.9059
0.0904	3.0	108	0.9360
0.0762	4.0	144	0.8768
0.0652	5.0	180	0.9361
0.054	6.0	216	0.9305
0.047	7.0	252	0.9453
0.0427	8.0	288	1.0083
0.0375	9.0	324	1.0142
0.0317	10.0	360	1.0458
0.0303	11.0	396	1.0515
0.0283	12.0	432	1.0791
0.0259	13.0	468	1.0594
0.0236	14.0	504	1.1078
0.0213	15.0	540	1.0250
0.0194	16.0	576	1.0492
0.0158	17.0	612	1.0782
0.016	18.0	648	1.1181
0.0135	19.0	684	1.1222
0.0138	20.0	720	1.1314
0.013	21.0	756	1.1197
0.0106	22.0	792	1.1216
0.0106	23.0	828	1.1382
0.0105	24.0	864	1.1542
0.0084	25.0	900	1.1758
0.0078	26.0	936	1.1630
0.0071	27.0	972	1.1524
0.007	28.0	1008	1.1615
0.0049	29.0	1044	1.1673
0.0062	30.0	1080	1.1623
0.0057	31.0	1116	1.1709
0.0046	32.0	1152	1.1976
0.0043	33.0	1188	1.2217
0.0035	34.0	1224	1.1863
0.0051	35.0	1260	1.2208
0.006	36.0	1296	1.1681
0.0044	37.0	1332	1.1783
0.0053	38.0	1368	1.1821
0.0049	39.0	1404	1.1724
0.0042	40.0	1440	1.1936
0.0031	41.0	1476	1.2066
0.0031	42.0	1512	1.2156
0.0039	43.0	1548	1.2054
0.0026	44.0	1584	1.2000
0.0028	45.0	1620	1.2259
0.0021	46.0	1656	1.2244
0.0026	47.0	1692	1.2218
0.0037	48.0	1728	1.2165
0.003	49.0	1764	1.2012
0.0021	50.0	1800	1.1950
0.0026	51.0	1836	1.2444
0.0024	52.0	1872	1.2066
0.0023	53.0	1908	1.2075
0.002	54.0	1944	1.2476
0.0016	55.0	1980	1.2365
0.0016	56.0	2016	1.2422
0.0014	57.0	2052	1.2420
0.0013	58.0	2088	1.2246
0.002	59.0	2124	1.2482
0.0014	60.0	2160	1.2752
0.0014	61.0	2196	1.2494
0.0013	62.0	2232	1.2648
0.0018	63.0	2268	1.2743
0.0027	64.0	2304	1.2162
0.0019	65.0	2340	1.2315
0.0016	66.0	2376	1.2573
0.0012	67.0	2412	1.2511
0.0018	68.0	2448	1.2632
0.0022	69.0	2484	1.2582
0.0015	70.0	2520	1.2676
0.0011	71.0	2556	1.2798
0.002	72.0	2592	1.2352
0.0012	73.0	2628	1.2430
0.0012	74.0	2664	1.2731
0.001	75.0	2700	1.2773
0.0009	76.0	2736	1.2506
0.001	77.0	2772	1.2479
0.0008	78.0	2808	1.2521
0.0008	79.0	2844	1.2630
0.0005	80.0	2880	1.2725
0.0009	81.0	2916	1.2539
0.0005	82.0	2952	1.2643
0.0007	83.0	2988	1.2722
0.001	84.0	3024	1.2690
0.0007	85.0	3060	1.2914
0.0006	86.0	3096	1.2911
0.0007	87.0	3132	1.2977
0.0007	88.0	3168	1.3432
0.0008	89.0	3204	1.3392
0.001	90.0	3240	1.2964
0.0023	91.0	3276	1.2660
0.0019	92.0	3312	1.2739
0.0017	93.0	3348	1.2968
0.0017	94.0	3384	1.3048
0.0014	95.0	3420	1.3139
0.0017	96.0	3456	1.3031
0.0012	97.0	3492	1.2952
0.0014	98.0	3528	1.3281
0.0021	99.0	3564	1.3087
0.0024	100.0	3600	1.2122
0.0028	101.0	3636	1.2194

Framework versions

• Transformers 4.49.0
• Pytorch 2.6.0+cu124
• Datasets 3.3.1
• Tokenizers 0.21.0