[2026-01-06 15:20:32,536][fusion_bench.programs.fusion_program][INFO] - Running the model fusion program.
[2026-01-06 15:20:32,539][fusion_bench.programs.fusion_program][INFO] - loading model pool
[2026-01-06 15:20:34,422][fusion_bench.programs.fusion_program][INFO] - loading method
[2026-01-06 15:20:34,478][fusion_bench.method.classification.image_classification_finetune][INFO] - Training interval: step
[2026-01-06 15:20:34,479][fusion_bench.method.classification.image_classification_finetune][INFO] - Max epochs: -1, max steps: 4000
[2026-01-06 15:20:34,479][fusion_bench.programs.fusion_program][INFO] - loading task pool
[2026-01-06 15:20:35,984][fusion_bench.method.classification.image_classification_finetune][INFO] - Number of classes for dataset svhn: 10
[2026-01-06 15:20:35,995][datasets.load][WARNING] - Using the latest cached version of the dataset since svhn couldn't be found on the Hugging Face Hub (offline mode is enabled).
[2026-01-06 15:20:36,001][datasets.packaged_modules.cache.cache][WARNING] - Found the latest cached dataset configuration 'cropped_digits' at /data/dataset/datasets/svhn/cropped_digits/0.0.0/f9e1717d73324ebbc1ece84267ccd80d0aca0690 (last modified on Mon Jan  5 04:19:05 2026).
[2026-01-06 15:20:36,594][fusion_bench.method.classification.image_classification_finetune][INFO] - Training dataset size: 73257
[2026-01-06 15:20:36,620][datasets.load][WARNING] - Using the latest cached version of the dataset since svhn couldn't be found on the Hugging Face Hub (offline mode is enabled).
[2026-01-06 15:20:36,625][datasets.packaged_modules.cache.cache][WARNING] - Found the latest cached dataset configuration 'cropped_digits' at /data/dataset/datasets/svhn/cropped_digits/0.0.0/f9e1717d73324ebbc1ece84267ccd80d0aca0690 (last modified on Mon Jan  5 04:19:05 2026).
[2026-01-06 15:20:36,819][fusion_bench.method.classification.image_classification_finetune][INFO] - optimizer:
SGD (
Parameter Group 0
    dampening: 0
    differentiable: False
    foreach: None
    fused: None
    lr: 0.01
    maximize: False
    momentum: 0.9
    nesterov: False
    weight_decay: 0.0001
)
[2026-01-06 15:59:07,997][fusion_bench.method.classification.image_classification_finetune][INFO] - Saving the final model to outputs/convnext-base-224/svhn/batch_size=64,lr=0.01/raw_checkpoints/final