model_006943978_vit_tiny.py

Model Overview

A tiny-scale implementation of the vit architecture, built for retrieval tasks.

Architecture

  • Architecture: vit
  • Scale: tiny
  • Attention: flash
  • Fusion strategy: low rank
  • Task head: retrieval
  • Activation: gelu
  • Normalization: groupnorm
  • Initialization: kaiming

Training

  • Optimizer: adafactor
  • LR scheduler: onecycle

Files

  • model_006943978_vit_tiny.py — main artifact of this repository

License

See the license field above.

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support