Downloads · 30 days
3
16% of all-time downloads
domainbed-dann-pacs-best
liangshuxin/domainbed-dann-pacs-best
domainbed-dann-pacs-best is a machine learning model from liangshuxin. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
This is the best performing DANN model selected based on validation accuracy.
All-time downloads
19
Public
Repo size
96.8 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin96.8 MB · 100%
At a glance
- Model type
- domainbed
- Access
- Public
- Created
- Dec 2, 2024
- Updated
- Dec 2, 2024
- SHA
- 9034b2ac
Model card
From the Hugging Face model README
Best DomainBed DANN Model for PACS
This is the best performing DANN model selected based on validation accuracy.
Model Details
- Validation Accuracy: 0.98%
- Test Environments: [1, 3]
- Input Shape: (3, 224, 224)
- Number of Classes: 7
- Number of Domains: 2
Hyperparameters
{
"data_augmentation": true,
"resnet18": false,
"resnet50_augmix": true,
"dinov2": false,
"vit": false,
"vit_attn_tune": false,
"freeze_bn": false,
"lars": false,
"linear_steps": 500,
"resnet_dropout": 0.0,
"vit_dropout": 0.0,
"class_balanced": false,
"nonlinear_classifier": false,
"lambda": 1.0,
"weight_decay_d": 0.0,
"d_steps_per_g_step": 1,
"grad_penalty": 0.0,
"beta1": 0.5,
"mlp_width": 256,
"mlp_depth": 3,
"mlp_dropout": 0.0,
"lr": 5e-05,
"weight_decay": 0.0,
"batch_size": 32,
"lr_g": 5e-05,
"lr_d": 5e-05,
"weight_decay_g": 0.0
}
- Type
- domainbed
- Created
- Dec 2, 2024
- Updated
- Dec 2, 2024