Datasetpublic

happy8825/siglip_classification_result_bestcheckpoint_20251217_003311

SigLIP Classification Eval Model: /hub_data3/seohyun/outputs/siglip_ecva/best Test repo: happy8825/siglip_test (split: train) Samples visualized: 50 of 924 Metrics accuracy: 0.6732 f1: 0.7051 total: 924, correct: 622 Confusion Matrix

sourceHugging Faceupdated 10mo agoView on Hugging Face

0likes5downloads

Dataset Card

SigLIP Classification Eval

—Model: /hubdata3/seohyun/outputs/siglipecva/best

—Test repo: happy8825/siglip_test (split: train)

—Samples visualized: 50 of 924

Metrics

—accuracy: 0.6732

—f1: 0.7051

—total: 924, correct: 622

Confusion Matrix

[image]