vilt-b32-finetuned-vqa-willowmore

This model is a fine-tuned version of dandelin/vilt-b32-finetuned-vqa on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.2704

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: 5e-05
• train_batch_size: 8
• eval_batch_size: 16
• 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: 100

Training results

Training Loss	Epoch	Step	Validation Loss
0.8442	1.0	74	0.3420
0.4009	2.0	148	0.2322
0.2016	3.0	222	0.0716
0.0257	4.0	296	0.4162
0.0082	5.0	370	0.0805
0.0022	6.0	444	0.0889
0.0013	7.0	518	0.1099
0.0011	8.0	592	0.1282
0.0008	9.0	666	0.1389
0.0007	10.0	740	0.1484
0.0006	11.0	814	0.1567
0.0005	12.0	888	0.1632
0.0004	13.0	962	0.1687
0.0004	14.0	1036	0.1735
0.0003	15.0	1110	0.1785
0.0003	16.0	1184	0.1818
0.0003	17.0	1258	0.1855
0.0002	18.0	1332	0.1888
0.0002	19.0	1406	0.1917
0.0002	20.0	1480	0.1947
0.0002	21.0	1554	0.1975
0.0002	22.0	1628	0.2001
0.0002	23.0	1702	0.2022
0.0002	24.0	1776	0.2045
0.0001	25.0	1850	0.2066
0.0001	26.0	1924	0.2085
0.0001	27.0	1998	0.2105
0.0001	28.0	2072	0.2124
0.0001	29.0	2146	0.2141
0.0001	30.0	2220	0.2158
0.0001	31.0	2294	0.2175
0.0001	32.0	2368	0.2190
0.0001	33.0	2442	0.2203
0.0001	34.0	2516	0.2218
0.0001	35.0	2590	0.2234
0.0001	36.0	2664	0.2247
0.0001	37.0	2738	0.2261
0.0001	38.0	2812	0.2274
0.0001	39.0	2886	0.2287
0.0001	40.0	2960	0.2300
0.0001	41.0	3034	0.2312
0.0001	42.0	3108	0.2324
0.0001	43.0	3182	0.2337
0.0	44.0	3256	0.2348
0.0	45.0	3330	0.2359
0.0	46.0	3404	0.2371
0.0	47.0	3478	0.2382
0.0	48.0	3552	0.2393
0.0	49.0	3626	0.2404
0.0	50.0	3700	0.2414
0.0	51.0	3774	0.2424
0.0	52.0	3848	0.2434
0.0	53.0	3922	0.2444
0.0	54.0	3996	0.2454
0.0	55.0	4070	0.2463
0.0	56.0	4144	0.2473
0.0	57.0	4218	0.2482
0.0	58.0	4292	0.2491
0.0	59.0	4366	0.2500
0.0	60.0	4440	0.2509
0.0	61.0	4514	0.2517
0.0	62.0	4588	0.2525
0.0	63.0	4662	0.2533
0.0	64.0	4736	0.2541
0.0	65.0	4810	0.2549
0.0	66.0	4884	0.2556
0.0	67.0	4958	0.2564
0.0	68.0	5032	0.2571
0.0	69.0	5106	0.2578
0.0	70.0	5180	0.2585
0.0	71.0	5254	0.2592
0.0	72.0	5328	0.2598
0.0	73.0	5402	0.2605
0.0	74.0	5476	0.2611
0.0	75.0	5550	0.2617
0.0	76.0	5624	0.2623
0.0	77.0	5698	0.2629
0.0	78.0	5772	0.2634
0.0	79.0	5846	0.2640
0.0	80.0	5920	0.2645
0.0	81.0	5994	0.2650
0.0	82.0	6068	0.2655
0.0	83.0	6142	0.2660
0.0	84.0	6216	0.2664
0.0	85.0	6290	0.2669
0.0	86.0	6364	0.2673
0.0	87.0	6438	0.2677
0.0	88.0	6512	0.2680
0.0	89.0	6586	0.2684
0.0	90.0	6660	0.2687
0.0	91.0	6734	0.2690
0.0	92.0	6808	0.2693
0.0	93.0	6882	0.2695
0.0	94.0	6956	0.2697
0.0	95.0	7030	0.2699
0.0	96.0	7104	0.2701
0.0	97.0	7178	0.2702
0.0	98.0	7252	0.2703
0.0	99.0	7326	0.2704
0.0	100.0	7400	0.2704

Framework versions

• Transformers 4.48.1
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0