SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2109.03508 · 2021

RepNAS: Searching for Efficient Re-parameterizing Blocks

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 13 functions out of this paper's own repositories and ran 3 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
bestfleer/RepNAS canonical 3 of 13
FunctionStatusWhere it lives
mixup_target Ran bestfleer/RepNAS/utils/Mixup.py
code served (permissive licence) · get_code("8733e0d87f232d11")
one_hot Ran bestfleer/RepNAS/utils/Mixup.py
code served (permissive licence) · get_code("4728dc2025227bba")
shear_y Ran bestfleer/RepNAS/utils/auto_augment.py
code served (permissive licence) · get_code("2d7b973a4b160d73")
accuracy Not yet run bestfleer/RepNAS/cnn/utils.py
code served (permissive licence) · get_code("e93d9ddbfeb73ced")
dist_reduce_tensor Not yet run bestfleer/RepNAS/distributed.py
code served (permissive licence) · get_code("4be37bd74adfab6e")
init_dist Not yet run bestfleer/RepNAS/distributed.py
code served (permissive licence) · get_code("c94c1dbbf94b5928")
master_only Not yet run bestfleer/RepNAS/distributed.py
code served (permissive licence) · get_code("028e1a08f69df3e7")
rand_bbox Not yet run bestfleer/RepNAS/utils/Mixup.py
code served (permissive licence) · get_code("45e8fbb4a42814eb")
shear_x Not yet run bestfleer/RepNAS/utils/auto_augment.py
code served (permissive licence) · get_code("b31e84df5541213a")
transIII_1x1_kxk Not yet run bestfleer/RepNAS/cnn/ddb_transforms.py
code served (permissive licence) · get_code("a1bfd1b41b138e66")
transII_addbranch Not yet run bestfleer/RepNAS/cnn/ddb_transforms.py
code served (permissive licence) · get_code("788ed0b33aadf578")
transI_fusebn Not yet run bestfleer/RepNAS/cnn/ddb_transforms.py
code served (permissive licence) · get_code("079d0a85a0bfbb47")
translate_x_rel Not yet run bestfleer/RepNAS/utils/auto_augment.py
code served (permissive licence) · get_code("61f2a3e08a11e1f5")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

In the past years, significant improvements in the field of neural architecture search(NAS) have been made. However, it is still challenging to search for efficient networks due to the gap between the searched constraint and real inference time exists. To search for a high-performance network with low inference time, several previous works set a computational complexity constraint for the search algorithm. However, many factors affect the speed of inference(e.g., FLOPs, MACs). The correlation between a single indicator and the latency is not strong. Currently, some re-parameterization(Rep) techniques are proposed to convert multi-branch to single-path architecture which is inference-friendly. Nevertheless, multi-branch architectures are still human-defined and inefficient. In this work, we propose a new search space that is suitable for structural re-parameterization techniques. RepNAS, a one-stage NAS approach, is present to efficiently search the optimal diverse branch block(ODBB) for each layer under the branch number constraint. Our experimental results show the searched ODBB can easily surpass the manual diverse branch block(DBB) with efficient training.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2109.03508")
get_code_for_paper("2109.03508")
have("2109.03508")

Connect an agent — have() is free.