aoi_clip_high_resolution_concate_fusin_gpt_random_sampler

This model is a fine-tuned version of OFA-Sys/chinese-clip-vit-base-patch16 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 2.9963
• Accuracy: 0.0500

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: 40
• eval_batch_size: 20
• seed: 42
• gradient_accumulation_steps: 5
• total_train_batch_size: 200
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 100.0
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.0008	9.9872	3110	3.0008	0.0500
0.0007	19.9743	6220	2.9971	0.0501
0.0007	29.9615	9330	2.9971	0.0501
0.0007	39.9486	12440	2.9988	0.0500
0.0007	49.9358	15550	2.9968	0.0499
0.0007	59.9229	18660	2.9966	0.0502
0.0007	69.9101	21770	2.9961	0.0503
0.0007	79.8972	24880	2.9967	0.0503
0.0007	89.8844	27990	2.9966	0.0503
0.0007	99.8715	31100	2.9963	0.0502

Framework versions

• Transformers 4.42.3
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1