SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1811.12814 · 2018

Graph-Based Global Reasoning Networks

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 15 functions out of this paper's own repositories and ran 11 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
facebookresearch/GloRe canonical 1 of 1
yangyucheng000/glore_res200 pwc_unofficial 6 of 10
ChriXiang/GloRe_pytorch reimplementation 4 of 4
FunctionStatusWhere it lives
ResNet18 Ran ChriXiang/GloRe_pytorch/resnet_example.py
pointer only (licence: NONE) · get_code("fb183196c497a93d")
ResNet34 Ran ChriXiang/GloRe_pytorch/resnet_example.py
pointer only (licence: NONE) · get_code("e39cd693a1b17d14")
ResNet50 Ran ChriXiang/GloRe_pytorch/resnet_example.py
pointer only (licence: NONE) · get_code("5dc8f93be8ef2d48")
autofill Ran facebookresearch/GloRe/test/test-single-clip.py
code served (permissive licence) · get_code("53974ebeab9f0b81")
calcul_acc Ran yangyucheng000/glore_res200/postprocess.py
code served (permissive licence) · get_code("5af3c724a0f8260c")
get_lr Ran yangyucheng000/glore_res200/src/lr_generator.py
code served (permissive licence) · get_code("ccb82a7bb4dc4e76")
glore Ran ChriXiang/GloRe_pytorch/glore.py
pointer only (licence: NONE) · get_code("1f6faceba55c7873")
linear_warmup_lr Ran yangyucheng000/glore_res200/src/lr_generator.py
code served (permissive licence) · get_code("9da7260d717664f9")
shear_x Ran yangyucheng000/glore_res200/src/transform_utils.py
code served (permissive licence) · get_code("8eaf53b48155177c")
shear_y Ran yangyucheng000/glore_res200/src/transform_utils.py
code served (permissive licence) · get_code("523681f17bc13539")
translate_x_rel Ran yangyucheng000/glore_res200/src/transform_utils.py
code served (permissive licence) · get_code("f38706e32849458d")
cre_groundtruth_dict_fromjson Not yet run yangyucheng000/glore_res200/infer/util/eval_by_sdk.py
code served (permissive licence) · get_code("91090d4b24121f00")
cre_groundtruth_dict_fromtxt Not yet run yangyucheng000/glore_res200/infer/util/eval_by_sdk.py
code served (permissive licence) · get_code("cc13de83d0285dde")
gen_file_name Not yet run yangyucheng000/glore_res200/infer/util/eval_by_sdk.py
code served (permissive licence) · get_code("8e4b6f31a18e273d")
warmup_cosine_annealing_lr Not yet run yangyucheng000/glore_res200/src/lr_generator.py
code served (permissive licence) · get_code("d4a0bdf1190319c4")

Repositories linked to this paper

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

Abstract

Globally modeling and reasoning over relations between regions can be beneficial for many computer vision tasks on both images and videos. Convolutional Neural Networks (CNNs) excel at modeling local relations by convolution operations, but they are typically inefficient at capturing global relations between distant regions and require stacking multiple convolution layers. In this work, we propose a new approach for reasoning globally in which a set of features are globally aggregated over the coordinate space and then projected to an interaction space where relational reasoning can be efficiently computed. After reasoning, relation-aware features are distributed back to the original coordinate space for down-stream tasks. We further present a highly efficient instantiation of the proposed approach and introduce the Global Reasoning unit (GloRe unit) that implements the coordinate-interaction space mapping by weighted global pooling and weighted broadcasting, and the relation reasoning via graph convolution on a small graph in interaction space. The proposed GloRe unit is lightweight, end-to-end trainable and can be easily plugged into existing CNNs for a wide range of tasks. Extensive experiments show our GloRe unit can consistently boost the performance of state-of-the-art backbone architectures, including ResNet, ResNeXt, SE-Net and DPN, for both 2D and 3D CNNs, on image classification, semantic segmentation and video action recognition task.

For agents

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

get_harvested_code_for_paper("1811.12814")
get_code_for_paper("1811.12814")
have("1811.12814")

Connect an agent — have() is free.