clip-finetuned-csu-p14-336-e4l59-l

This model is a fine-tuned version of openai/clip-vit-large-patch14-336 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 1.3460

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-09
• train_batch_size: 128
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 4.0

Training results

Training Loss	Epoch	Step	Validation Loss
0.3952	0.0921	500	1.4940
0.4562	0.1842	1000	1.4853
0.5131	0.2763	1500	1.4758
0.4481	0.3685	2000	1.4676
0.4839	0.4606	2500	1.4585
0.4377	0.5527	3000	1.4508
0.4231	0.6448	3500	1.4432
0.4369	0.7369	4000	1.4366
0.4082	0.8290	4500	1.4302
0.4234	0.9211	5000	1.4243
0.4266	1.0133	5500	1.4191
0.4438	1.1054	6000	1.4137
0.3814	1.1975	6500	1.4085
0.3327	1.2896	7000	1.4042
0.4045	1.3817	7500	1.3989
0.4038	1.4738	8000	1.3937
0.3659	1.5660	8500	1.3894
0.4282	1.6581	9000	1.3855
0.4173	1.7502	9500	1.3816
0.3758	1.8423	10000	1.3779
0.4105	1.9344	10500	1.3745
0.3765	2.0265	11000	1.3716
0.3746	2.1186	11500	1.3690
0.3783	2.2108	12000	1.3662
0.3832	2.3029	12500	1.3640
0.3984	2.3950	13000	1.3617
0.4124	2.4871	13500	1.3593
0.3363	2.5792	14000	1.3572
0.3274	2.6713	14500	1.3555
0.4039	2.7634	15000	1.3538
0.378	2.8556	15500	1.3524
0.3543	2.9477	16000	1.3511
0.3606	3.0398	16500	1.3501
0.4024	3.1319	17000	1.3491
0.3182	3.2240	17500	1.3482
0.3564	3.3161	18000	1.3475
0.3842	3.4083	18500	1.3470
0.352	3.5004	19000	1.3467
0.3828	3.5925	19500	1.3464
0.39	3.6846	20000	1.3462
0.3618	3.7767	20500	1.3461
0.3856	3.8688	21000	1.3461
0.3586	3.9609	21500	1.3460

Framework versions

• Transformers 4.45.0.dev0
• Pytorch 1.12.1
• Datasets 2.21.0
• Tokenizers 0.19.1