logo-matching-base

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

• Adjusted Rand Score: 0.1048
• Adjusted Mutual Info Score: 0.1103
• Homogeneity Score: 0.1072
• Completeness Score: 0.6399
• Fowlkes Mallows Score: 0.4934
• Pair Confusion Matrix: [[13442, 34208], [1176, 12430]]
• Loss: 0.1048

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: 32
• eval_batch_size: 16
• seed: 42
• optimizer: Use OptimizerNames.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: 20

Training results

Training Loss	Epoch	Step	Adjusted Rand Score	Adjusted Mutual Info Score	Homogeneity Score	Completeness Score	Fowlkes Mallows Score	Pair Confusion Matrix	Validation Loss
0.0988	1.0	34	0.0667	0.1257	0.1492	0.5530	0.4345	[[17354, 30296], [3408, 10198]]	0.0667
0.0918	2.0	68	0.0512	0.0801	0.1155	0.4762	0.4303	[[15840, 31810], [3306, 10300]]	0.0512
0.0972	3.0	102	0.0494	0.1485	0.2067	0.4977	0.3496	[[27372, 20278], [6908, 6698]]	0.0494
0.1027	4.0	136	0.1550	0.1639	0.2052	0.5160	0.4118	[[30918, 16732], [6132, 7474]]	0.1550
0.0944	5.0	170	0.1798	0.1948	0.2833	0.5017	0.3910	[[35784, 11866], [7490, 6116]]	0.1798
0.09	6.0	204	0.1186	0.1355	0.1836	0.5103	0.4414	[[23042, 24608], [4096, 9510]]	0.1186
0.0858	7.0	238	0.0756	0.1533	0.2163	0.4937	0.3570	[[29292, 18358], [7030, 6576]]	0.0756
0.0735	8.0	272	0.1107	0.1555	0.2280	0.4815	0.3651	[[31962, 15688], [7288, 6318]]	0.1107
0.0868	9.0	306	0.2523	0.2273	0.3022	0.5371	0.4428	[[36956, 10694], [6766, 6840]]	0.2523
0.0932	10.0	340	0.2318	0.2251	0.3703	0.4901	0.3838	[[41708, 5942], [9010, 4596]]	0.2318
0.0642	11.0	374	0.1840	0.1915	0.2284	0.5528	0.4331	[[31348, 16302], [5754, 7852]]	0.1840
0.0918	12.0	408	0.1472	0.1783	0.2400	0.5135	0.3946	[[32168, 15482], [6722, 6884]]	0.1472
0.0818	13.0	442	0.2861	0.2445	0.3090	0.5556	0.4658	[[37650, 10000], [6500, 7106]]	0.2861
0.0877	14.0	476	0.1159	0.1723	0.2474	0.4934	0.3621	[[32834, 14816], [7496, 6110]]	0.1159
0.0849	15.0	510	0.1053	0.1092	0.1086	0.6293	0.4935	[[13498, 34152], [1182, 12424]]	0.1053
0.0927	16.0	544	0.1119	0.1599	0.2309	0.4880	0.3674	[[31812, 15838], [7216, 6390]]	0.1119
0.0794	17.0	578	0.2050	0.2036	0.2405	0.5662	0.4462	[[31942, 15708], [5590, 8016]]	0.2050
0.0888	18.0	612	0.2215	0.2213	0.3139	0.5143	0.4002	[[38766, 8884], [7984, 5622]]	0.2215
0.0815	19.0	646	0.1725	0.1935	0.2815	0.4997	0.3833	[[35946, 11704], [7668, 5938]]	0.1725
0.0796	20.0	680	0.1048	0.1103	0.1072	0.6399	0.4934	[[13442, 34208], [1176, 12430]]	0.1048

Framework versions

• Transformers 4.47.1
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0