Xpos Vit - IMAGENET100

This model was trained using the vit-analysis framework.

Model Details

  • Model Type: XPOS Vision Transformer
  • Dataset: imagenet100
  • Best Accuracy: 72.30%
  • Image Size: 224
  • Patch Size: 16
  • Hidden Dim: 192
  • Depth: 12
  • Num Heads: 3
  • MLP Dim: 768
  • Num Classes: 100

Training Configuration

  • Epochs: 120
  • Batch Size: 512
  • Learning Rate: 0.004
  • Weight Decay: 0.05
  • Label Smoothing: 0.1

Usage

import torch
from models import XPOSSimpleVisionTransformer

# Load checkpoint
checkpoint = torch.load('xpos_vit_imagenet100_best.pth')
model = ...  # Initialize model with same config
model.load_state_dict(checkpoint['state_dict'])
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. ๐Ÿ™‹ Ask for provider support