Publisher-reported 0.9708 accuracy
According to the model card, the model reports 0.970833 accuracy on a 240-image evaluation set with 120 male and 120 female images.
Open Source Model Profile · hungdang1610
gender
gender is an 85.8M-parameter ViT gender classifier from hungdang1610. According to the model card, it reports 0.970833 accuracy on a 240-image evaluation set.
- Publisher
- hungdang1610
- Task
- image-classification
- Model type
- vit
- License
- apache-2.0
- Library
- transformers
- Publication status
- Accepted · not indexed
Model overview
gender is published by hungdang1610 for image classification. The captured configuration identifies ViTForImageClassification with model type vit, and Safetensors metadata reports 85,800,194 parameters with Apache-2.0 licensing. According to the model card, it was fine-tuned on 1,827 images, building on rizvandwiki/gender-classification.
Recorded capabilities
Fine-tuned from a sibling classifier
According to the model card, the model was fine-tuned on 1,827 images, building on rizvandwiki/gender-classification.
Documented optimizer schedule
According to the model card, training used the AdamW optimizer with weight decay 0.05, a CosineAnnealingLR scheduler, learning rate 5e-6, and 20 epochs.
Captured 85.8M ViT weights
Safetensors metadata reports 85,800,194 parameters, and the hub record lists Transformers support.
Use cases in the source record
- Gender image classification over male and female labels within the publisher-described 240-image evaluation setup.
- Small-data ViT fine-tuning experiments following the publisher-documented AdamW and cosine-schedule workflow.
Limitations and unknowns
- Accuracy figures are publisher-reported values from the model card and were not independently verified by Ethen.
- No context-window value applies to this image-classification record beyond the model configuration.
- Provider state is historical snapshot data and should be refreshed before being presented as current.
Source and provenance
Source: hungdang1610/gender
Captured: Unknown. Processed: 2026-09-07T19:34:46.598154+00:00.
tags: image-classification pytorch metrics: accuracy model-index: - name: gender-classification results: - task: name: Image Classification type: image-classification metrics: - name: Accuracy type: accuracy value: 0.970833 gender-classification Evaluate set: 240 unseen images, from 27/05-29/05 from ShotX, no duplicate, clean, 120 male and 120 female. Loss function is CrossEntropy. Model finetuning on 1827 images from 15/05-21/05 from ShotX, base on rizvandwiki/gender-classification . . Using AdamW optimizer, weight_decay=0.05, CosineAnnealingLR scheduler, learning rate 5e-6, 20 epochs. accuracy loss 0.970833 0.102212 Example Images…
F001F002F003F004F005F006F007F008F010F011F013F014