Documented ViT base
According to the model card, the model is fine-tuned from google/vit-base-patch16-224-in21k.
Open Source Model Profile · mo-thecreator
vit-Facial-Expression-Recognition
vit-Facial-Expression-Recognition is a ViT fine-tune from mo-thecreator for facial emotion recognition. According to the model card, it fine-tunes google/vit-base-patch16-224-in21k and reports 0.8434 accuracy.
- Publisher
- mo-thecreator
- Task
- image-classification
- Model type
- vit
- License
- Unknown
- Library
- transformers
- Publication status
- Accepted · not indexed
Model overview
vit-Facial-Expression-Recognition is published by mo-thecreator as an image-classification model. The captured configuration identifies ViTForImageClassification with a vit model type, and Safetensors metadata reports 85,804,039 parameters. According to the model card, it is a fine-tuned version of google/vit-base-patch16-224-in21k for facial emotion recognition.
Recorded capabilities
Three-dataset training
According to the model card, it was trained on FER 2013, MMI Facial Expression Database, and AffectNet datasets.
Reported 0.8434 accuracy
According to the model card, the evaluation set reports loss 0.4503 and accuracy 0.8434.
Documented hyperparameters
According to the model card, training used learning rate 3e-05, batch sizes 32, 3 epochs, cosine scheduling, and 1,000 warmup steps.
Use cases in the source record
- Facial emotion recognition over seven categories: Angry, Disgust, Fear, Happy, Sad, Surprise, and Neutral, according to the model card.
- Fine-tuned ViT experiments that reuse the documented resizing and augmentation setup for expression datasets, according to the model card.
Limitations and unknowns
- Reported loss and accuracy are publisher model-card claims and have not been independently verified by Ethen.
- No license value was extracted from this record.
- Provider state is historical snapshot data, not independently refreshed current availability.
Source and provenance
Source: mo-thecreator/vit-Facial-Expression-Recognition
Captured: Unknown. Processed: 2026-09-07T19:34:52.148205+00:00.
vit-Facial-Expression-Recognition This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the FER 2013 , MMI Facial Expression Database , and AffectNet dataset datasets. It achieves the following results on the evaluation set: Loss: 0.4503 Accuracy: 0.8434 Model description The vit-face-expression model is a Vision Transformer fine-tuned for the task of facial emotion recognition. It is trained on the FER2013, MMI facial Expression, and AffectNet datasets, which consist of facial images categorized into seven different emotions: Angry Disgust Fear Happy Sad Surprise Neutral Data Preprocessing The input images are…
F001F002F003F004F005F006F007F008F009F010F011F012F013F014F015