Replication Data for: Towards Efficient Convolutional Neural Network for Embedded Hardware via Multi-Dimensional Pruning (doi:10.21979/N9/BTNOJN)

View:

Part 1: Document Description
Part 2: Study Description
Part 3: Data Files Description
Part 4: Variable Description
Part 5: Other Study-Related Materials
Entire Codebook

Replication Data for: Towards Efficient Convolutional Neural Network for Embedded Hardware via Multi-Dimensional Pruning

doi:10.21979/N9/BTNOJN

DR-NTU (Data)

2023-12-14

1

Kong, Hao; Liu, Weichen, 2023, "Replication Data for: Towards Efficient Convolutional Neural Network for Embedded Hardware via Multi-Dimensional Pruning", https://doi.org/10.21979/N9/BTNOJN, DR-NTU (Data), V1, UNF:6:YowCvL3Bn3yfomoWb2+WPg== [fileUNF]

Replication Data for: Towards Efficient Convolutional Neural Network for Embedded Hardware via Multi-Dimensional Pruning

doi:10.21979/N9/BTNOJN

Kong, Hao (Nanyang Technological University)

Liu, Weichen (Nanyang Technological University)

Pycharm

Academic Research Fund Tier 2 (MOE2019-T2-1-071)

NAP (M4082282)

DR-NTU (Data)

Kong, Hao

Kong, Hao

2023-05-13

https://doi.org/10.21979/N9/BTNOJN

Computer and Information Science, Computer and Information Science, Multi-dimensional model compression, Deep learning acceleration

This dataset is created to store the related data of the following published paper: Hao Kong, Di Liu, Xiangzhong Luo, Shuo Huai, Ravi Subramaniam, Christian Makaya, Qian Lin, Weichen Liu*, “Towards Efficient Convolutional Neural Network for Embedded Hardware via Multi-Dimensional Pruning”, ACM/IEEE Design Automation Conference (DAC), 2023.

Source code

MIT License<br> Copyright (c) 2022 Hao Kong<br><br> Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:<br><br> The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.<br><br> THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

10.1109/DAC56929.2023.10247965

Kong, H., Liu, D., Luo, X., Huai, S., Subramaniam, R., Makaya, C., ... & Liu, W. (2023, July). Towards Efficient Convolutional Neural Network for Embedded Hardware via Multi-Dimensional Pruning. In 2023 60th ACM/IEEE Design Automation Conference (DAC) (pp. 1-6). IEEE.

10356/167489

Kong, H., Liu, D., Luo, X., Huai, S., Subramaniam, R., Makaya, C., ... & Liu, W. (2023, July). Towards Efficient Convolutional Neural Network for Embedded Hardware via Multi-Dimensional Pruning. In 2023 60th ACM/IEEE Design Automation Conference (DAC) (pp. 1-6). IEEE.

• Number of cases: 100

• No. of variables per record: 11

• Type of File: text/tab-separated-values

UNF:6:MiWG8Ugc8u6XggEAZMqiPg==

• Number of cases: 43

• No. of variables per record: 11

• Type of File: text/tab-separated-values

UNF:6:6TW8Cfq+ncf4EPDTFiTXhw==

• Number of cases: 100

• No. of variables per record: 11

• Type of File: text/tab-separated-values

UNF:6:/yqxOdOsdrrzyEMQEXidNA==

• dim - dim
• l1 - l1
• l2 - l2
• l3 - l3
• l4 - l4
• w1 - w1
• w2 - w2
• w3 - w3
• w4 - w4
• res - res
• res_mult - res_mult
• dim - dim
• l1 - l1
• l2 - l2
• l3 - l3
• l4 - l4
• w1 - w1
• w2 - w2
• w3 - w3
• w4 - w4
• res - res
• res_mult - res_mult
• dim - dim
• l1 - l1
• l2 - l2
• l3 - l3
• l4 - l4
• w1 - w1
• w2 - w2
• w3 - w3
• w4 - w4
• res - res
• res_mult - res_mult

f112245 Location:

Variable Format: character

Notes: UNF:6:9Syx2Tp+fO5ow94nLjhC5Q==

f112245 Location:

Summary Statistics: Max. 3.0; Min. 1.0; Mean 1.9300000000000002; StDev 0.8439325934114774; Valid 100.0;

Variable Format: numeric

Notes: UNF:6:4rrHVYRLr5fAbRoBA6kj5w==

f112245 Location:

Summary Statistics: Min. 1.0; Mean 2.5599999999999996; Max. 4.0; StDev 1.1574964275530097; Valid 100.0;

Variable Format: numeric

Notes: UNF:6:uD9UAQixvFs0Zbs4p8FtTg==

f112245 Location:

Summary Statistics: Mean 3.57; Min. 1.0; Max. 6.0; Valid 100.0; StDev 1.8436623630944222;

Variable Format: numeric

Notes: UNF:6:6eVGGNoZpN7QcMbLH5fRkg==

f112245 Location:

Summary Statistics: Mean 2.03; Valid 100.0; Max. 3.0; StDev 0.79715402868865; Min. 1.0

Variable Format: numeric

Notes: UNF:6:bZtsCNX+Wdhm8Tki4kicPw==

f112245 Location:

Summary Statistics: Max. 64.0; StDev 0.0; Valid 100.0; Mean 64.0; Min. 64.0

Variable Format: numeric

Notes: UNF:6:Dg4TLVpyjbGMxr+0EujS3g==

f112245 Location:

Summary Statistics: Valid 100.0; Mean 128.0; Min. 128.0; StDev 0.0; Max. 128.0

Variable Format: numeric

Notes: UNF:6:ywRIUTqRFLRC+8YXDsrFOw==

f112245 Location:

Summary Statistics: Valid 100.0; StDev 0.0; Min. 256.0; Mean 256.0; Max. 256.0

Variable Format: numeric

Notes: UNF:6:ncmRIfUrResUaEe9hCiPVw==

f112245 Location:

Summary Statistics: Valid 100.0; Max. 512.0; Mean 512.0; Min. 512.0; StDev 0.0

Variable Format: numeric

Notes: UNF:6:r3Np3AKiEheMbvkZ+rg5qw==

f112245 Location:

Summary Statistics: Max. 224.0; Min. 224.0; StDev 0.0; Valid 100.0; Mean 224.0;

Variable Format: numeric

Notes: UNF:6:gl0JF7bOWB1M68r1rl906w==

f112245 Location:

Summary Statistics: StDev 0.0; Mean 1.0; Max. 1.0; Min. 1.0; Valid 100.0

Variable Format: numeric

Notes: UNF:6:uiwq51K46f3Bh7o/C/ucAw==

f112211 Location:

Variable Format: character

Notes: UNF:6:WHen03lmMel7y2ruKC8F+Q==

f112211 Location:

Summary Statistics: Valid 43.0; Max. 3.0; Mean 3.0; StDev 0.0; Min. 3.0

Variable Format: numeric

Notes: UNF:6:aVuhy3zp/uIHkTkwB1UQxw==

f112211 Location:

Summary Statistics: Min. 4.0; Mean 4.0; Valid 43.0; Max. 4.0; StDev 0.0

Variable Format: numeric

Notes: UNF:6:1rapvoyFAjKKzP5SFDNtGQ==

f112211 Location:

Summary Statistics: StDev 0.0; Max. 6.0; Valid 43.0; Min. 6.0; Mean 6.0;

Variable Format: numeric

Notes: UNF:6:K77+cHcNmRCIohsopfy4wA==

f112211 Location:

Summary Statistics: Mean 3.0; Max. 3.0; StDev 0.0; Min. 3.0; Valid 43.0;

Variable Format: numeric

Notes: UNF:6:aVuhy3zp/uIHkTkwB1UQxw==

f112211 Location:

Summary Statistics: Mean 64.0; StDev 0.0; Max. 64.0; Valid 43.0; Min. 64.0;

Variable Format: numeric

Notes: UNF:6:K2p1+z9tb8wOACkki8ftPQ==

f112211 Location:

Summary Statistics: Min. 128.0; Mean 128.0; Valid 43.0; Max. 128.0; StDev 0.0

Variable Format: numeric

Notes: UNF:6:1jwm4QScGr60VEhyqZq7Ag==

f112211 Location:

Summary Statistics: Max. 256.0; Valid 43.0; StDev 0.0; Mean 256.0; Min. 256.0

Variable Format: numeric

Notes: UNF:6:811aR8KfZAUTftLOJpxPDg==

f112211 Location:

Summary Statistics: Min. 512.0; StDev 0.0; Valid 43.0; Max. 512.0; Mean 512.0;

Variable Format: numeric

Notes: UNF:6:bIpegQ01bEvnfb49zMXmEQ==

f112211 Location:

Summary Statistics: Valid 43.0; Mean 140.0; Max. 224.0; Min. 56.0; StDev 50.226155204899634

Variable Format: numeric

Notes: UNF:6:sW5iqfxOZ3aYPa73aR5WGA==

f112211 Location:

Summary Statistics: Min. 0.25; StDev 0.22422390716473048; Valid 43.0; Max. 1.0; Mean 0.625

Variable Format: numeric

Notes: UNF:6:/ON6XU+WwOpGxigWS5G34Q==

f112208 Location:

Variable Format: character

Notes: UNF:6:mpXO9aO6x/1+RB46FABFRQ==

f112208 Location:

Summary Statistics: Valid 100.0; StDev 0.0; Max. 3.0; Min. 3.0; Mean 3.0

Variable Format: numeric

Notes: UNF:6:/YpJspVrfSzcYIIeQkPv4Q==

f112208 Location:

Summary Statistics: Max. 4.0; Min. 4.0; Mean 4.0; StDev 0.0; Valid 100.0

Variable Format: numeric

Notes: UNF:6:J+hMvZsgeqBcRmdTo2QsFQ==

f112208 Location:

Summary Statistics: Max. 6.0; Mean 6.0; Valid 100.0; StDev 0.0; Min. 6.0

Variable Format: numeric

Notes: UNF:6:/aS/jYRv58kV21yerrqSeA==

f112208 Location:

Summary Statistics: Min. 3.0; Valid 100.0; StDev 0.0; Max. 3.0; Mean 3.0

Variable Format: numeric

Notes: UNF:6:/YpJspVrfSzcYIIeQkPv4Q==

f112208 Location:

Summary Statistics: Mean 40.8; Max. 64.0; Min. 16.0; StDev 17.225773501566568; Valid 100.0

Variable Format: numeric

Notes: UNF:6:KPSL62T/34pQVZ2/+vV8ng==

f112208 Location:

Summary Statistics: Min. 32.0; Valid 100.0; Max. 128.0; Mean 87.03999999999999; StDev 35.254319887172294

Variable Format: numeric

Notes: UNF:6:V4XYzlJlp8S30H+rBe04KA==

f112208 Location:

Summary Statistics: Mean 154.88; Min. 64.0; StDev 77.81683752260948; Max. 256.0; Valid 100.0

Variable Format: numeric

Notes: UNF:6:GjEDLzFmK6YjyB8rjdeeAQ==

f112208 Location:

Summary Statistics: Mean 309.76; Min. 128.0; Max. 512.0; StDev 155.63367504521895; Valid 100.0

Variable Format: numeric

Notes: UNF:6:GvQqGrU56lfI5Nj3DGf1Vg==

f112208 Location:

Summary Statistics: Valid 100.0; Min. 224.0; StDev 0.0; Mean 224.0; Max. 224.0

Variable Format: numeric

Notes: UNF:6:gl0JF7bOWB1M68r1rl906w==

f112208 Location:

Summary Statistics: Min. 1.0; StDev 0.0; Mean 1.0; Max. 1.0; Valid 100.0;

Variable Format: numeric

Notes: UNF:6:uiwq51K46f3Bh7o/C/ucAw==

Label:

augment.py

text/x-python

Label:

benchmark.py

text/x-python

Label:

benchmark_resnet50.csv

text/csv

Label:

benchmark_resnet50.yaml

application/octet-stream

Label:

blocks.py

text/x-python

Label:

checkpoint.py

text/x-python

Label:

cifar.py

text/x-python

Label:

config.py

text/x-python

Label:

dataloader.py

text/x-python

Label:

depth_pruner_resnet50.yaml

application/octet-stream

Label:

depth_resnet50.yaml

application/octet-stream

Label:

distributed.py

text/x-python

Label:

eval_resnet50.yaml

application/octet-stream

Label:

imagenet.py

text/x-python

Label:

image.py

text/x-python

Label:

ines_wo-ites.yaml

application/octet-stream

Label:

joint_pruner_resnet110_v2.yaml

application/octet-stream

Label:

joint_pruner_resnet110_v3.yaml

application/octet-stream

Label:

joint_pruner_resnet110.yaml

application/octet-stream

Label:

joint_pruner_resnet50.yaml

application/octet-stream

Label:

LICENSE

text/plain; charset=US-ASCII

Label:

logging.py

text/x-python

Label:

main.py

text/x-python

Label:

main_worker.py

text/x-python

Label:

meters.py

text/x-python

Label:

mnade-l_resnet50_v2.yaml

application/octet-stream

Label:

mnade-l_resnet50.yaml

application/octet-stream

Label:

mnade-m_resnet50_v2.yaml

application/octet-stream

Label:

mnade-m_resnet50.yaml

application/octet-stream

Label:

mnade_resnet110_mix-res.yaml

application/octet-stream

Label:

mnade_resnet110.yaml

application/octet-stream

Label:

mnade_resnet50-1.yaml

application/octet-stream

Label:

mnade_resnet50_2.yaml

application/octet-stream

Label:

mnade_resnet50_3.yaml

application/octet-stream

Label:

mnade_resnet50_4.yaml

application/octet-stream

Label:

mnade_resnet50_5.yaml

application/octet-stream

Label:

mnade_resnet50_6.yaml

application/octet-stream

Label:

model_samples.ipynb

application/x-ipynb+json

Label:

models.py

text/x-python

Label:

net.py

text/x-python

Label:

optimizer.py

text/x-python

Label:

prune_resnet50.yaml

application/octet-stream

Label:

pruner.py

text/x-python

Label:

README.md

text/markdown

Label:

resnet110_mix-res.yaml

application/octet-stream

Label:

resnet110.yaml

application/octet-stream

Label:

resnet44.yaml

application/octet-stream

Label:

resnet50.yaml

application/octet-stream

Label:

resnet.py

text/x-python

Label:

resolution_resnet50.yaml

application/octet-stream

Label:

trainer.py

text/x-python

Label:

transforms.py

text/x-python

Label:

utils.py

text/x-python

Label:

validation_samples-checkpoint.ipynb

application/x-ipynb+json

Label:

validation_samples.ipynb

application/x-ipynb+json

Label:

verify_model-checkpoint.ipynb

application/x-ipynb+json

Label:

verify_model.ipynb

application/x-ipynb+json

Label:

width_pruner_resnet50.yaml

application/octet-stream

Label:

width_resnet50.yaml

application/octet-stream

Label:

wo-ines_ites.yaml

application/octet-stream

Label:

wo-ines_wo-ites.yaml

application/octet-stream