This model is a small densenet121 trained on cifar100.
Use the code below to get started with the model.
1import detectors
2import timm
3
4model = timm.create_model("densenet121_cifar100", pretrained=True)
Training data is cifar100.
-
config: scripts/train_configs/cifar100.json
-
model: densenet121_cifar100
-
dataset: cifar100
-
batch_size: 128
-
epochs: 300
-
validation_frequency: 5
-
seed: 1
-
criterion: CrossEntropyLoss
-
criterion_kwargs: {}
-
optimizer: SGD
-
lr: 0.1
-
optimizer_kwargs: {'momentum': 0.9, 'weight_decay': 0.0005}
-
scheduler: CosineAnnealingLR
-
scheduler_kwargs: {'T_max': 280}
-
debug: False
Testing data is cifar100.
This model card was created by Eduardo Dadalto.
