ViT fine-tune design
The captured configuration identifies ViTForImageClassification with model type vit, and hub tags list fine-tuning from google/vit-base-patch16-224-in21k.
Open Source Model Profile · dima806
man_woman_face_image_detection
man_woman_face_image_detection is an 85.80M-parameter ViT image classifier from dima806. According to the model card, it reports about 98.7% face-image accuracy.
- Publisher
- dima806
- Task
- image-classification
- Model type
- vit
- License
- apache-2.0
- Library
- transformers
- Publication status
- Accepted · not indexed
Model overview
man_woman_face_image_detection is published by dima806 as an image-classification model. The captured configuration identifies ViTForImageClassification with model type vit, and Safetensors metadata reports 85,800,194 parameters. According to the model card, it classifies face images with reported 98.7% accuracy.
Recorded capabilities
Publisher-reported accuracy detail
According to the model card, the publisher reports about 98.7% accuracy and provides a classification report covering precision, recall, F1-score, and support.
Transformers-compatible release
Hub metadata identifies transformers library support with pytorch, safetensors, vit, and endpoints compatibility.
Linked training notebook
According to the model card, further detail is linked through the publisher's Kaggle notebook for the ViT detection workflow.
Use cases in the source record
- Face-image classification research using the publisher's reported 98.7% accuracy workflow.
- Classifier evaluation study referencing the card's precision, recall, F1-score, and support report structure.
Limitations and unknowns
- No independently measured evaluation results were extracted; the 98.7% figure is a publisher-reported card claim.
- No training data composition, compute, or threshold detail was extracted from this record.
- Provider state is historical snapshot data, not independently refreshed current availability.
Source and provenance
Source: dima806/man_woman_face_image_detection
Captured: Unknown. Processed: 2026-09-07T19:34:43.616185+00:00.
Returns with about 98.7% accuracy whether the face belongs to man or woman based on face image. See https://www.kaggle.com/code/dima806/man-woman-face-image-detection-vit for more details. Classification report: precision recall f1-score support man 0.9885 0.9857 0.9871 51062 woman 0.9857 0.9885 0.9871 51062 accuracy 0.9871 102124 macro avg 0.9871 0.9871 0.9871 102124 weighted avg 0.9871 0.9871 0.9871 102124
F001F002F003F004F005F006F007F008F009F010F011F012F013F014