fusion-bench/resnet152-cifar100-batch_size_128_lr_0.001_training_data_ratio_0.5-2000

Deep Model Fusion

Fine-tuned ResNet model on dataset cifar100.

Models Merged

This is a merged model created using fusion-bench.

The following models were included in the merge:

Configuration

The following YAML configuration was used to produce this model:

Algorithm Configuration

_recursive_: false
_target_: fusion_bench.method.classification.image_classification_finetune.ImageClassificationFineTuning
_usage_: null
_version_: 0.2.25.dev0
dataloader_kwargs:
  batch_size: 128
  num_workers: 8
  pin_memory: true
label_smoothing: 0
lr_scheduler: null
max_epochs: -1
max_steps: 4000
optimizer:
  _target_: torch.optim.SGD
  lr: 0.001
  momentum: 0.9
  weight_decay: 0.0001
save_interval: 1000
save_on_train_epoch_end: false
save_top_k: -1
training_data_ratio: 0.5

Model Pool Configuration

_recursive_: false
_target_: fusion_bench.modelpool.resnet_for_image_classification.ResNetForImageClassificationPool
_usage_: null
_version_: 0.2.25.dev0
models:
  _pretrained_:
    config_path: microsoft/resnet-152
    dataset_name: cifar100
    pretrained: true
test_datasets: null
train_datasets:
  cifar100:
    _target_: datasets.load_dataset
    path: tanganke/cifar100
    split: train
type: transformers
val_datasets:
  cifar100:
    _target_: datasets.load_dataset
    path: tanganke/cifar100
    split: test