dit-base-finetuned-rvlcdip-finetuned-mobile-eye-tracking-dataset-v2

This model is a fine-tuned version of microsoft/dit-base-finetuned-rvlcdip on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.4029
• Accuracy: 0.8333

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: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	0.86	3	0.6939	0.4792
No log	2.0	7	0.6927	0.5417
0.6938	2.86	10	0.6986	0.4167
0.6938	4.0	14	0.6959	0.4583
0.6938	4.86	17	0.6798	0.6875
0.684	6.0	21	0.6890	0.5208
0.684	6.86	24	0.6768	0.5833
0.684	8.0	28	0.6637	0.6667
0.6772	8.86	31	0.7099	0.5
0.6772	10.0	35	0.6467	0.6042
0.6772	10.86	38	0.6059	0.6875
0.6283	12.0	42	0.5860	0.7292
0.6283	12.86	45	0.5670	0.7708
0.6283	14.0	49	0.5677	0.6875
0.5712	14.86	52	0.5453	0.6875
0.5712	16.0	56	0.4991	0.7708
0.5712	16.86	59	0.4830	0.75
0.5485	18.0	63	0.4485	0.7917
0.5485	18.86	66	0.4268	0.8542
0.506	20.0	70	0.4588	0.7708
0.506	20.86	73	0.4412	0.75
0.506	22.0	77	0.4099	0.8333
0.4636	22.86	80	0.4005	0.875
0.4636	24.0	84	0.4039	0.8333
0.4636	24.86	87	0.4040	0.8333
0.4453	25.71	90	0.4029	0.8333

Framework versions

• Transformers 4.33.0
• Pytorch 2.6.0+cu124
• Datasets 3.5.0
• Tokenizers 0.13.3