Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/yeonghyeon/ghostnet
TensorFlow implementation of GhostNet: More Features from Cheap Operations.
https://github.com/yeonghyeon/ghostnet
convolutional-neural-network convolutional-neural-networks deep-learning deep-neural-networks mnist mnist-classification mnist-dataset neural-network neural-networks
Last synced: about 2 months ago
JSON representation
TensorFlow implementation of GhostNet: More Features from Cheap Operations.
- Host: GitHub
- URL: https://github.com/yeonghyeon/ghostnet
- Owner: YeongHyeon
- License: mit
- Created: 2020-02-04T08:26:07.000Z (almost 5 years ago)
- Default Branch: master
- Last Pushed: 2020-02-06T01:03:08.000Z (almost 5 years ago)
- Last Synced: 2023-10-20T23:55:49.290Z (about 1 year ago)
- Topics: convolutional-neural-network, convolutional-neural-networks, deep-learning, deep-neural-networks, mnist, mnist-classification, mnist-dataset, neural-network, neural-networks
- Language: Python
- Homepage:
- Size: 180 KB
- Stars: 10
- Watchers: 3
- Forks: 4
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
GhostNet: More Features from Cheap Operations
=====
TensorFlow implementation of GhostNet: More Features from Cheap Operations.
## Performance
The performance is measured using below two CNN architectures.
Two Convolutional Neural Networks for experiment.
| |ConvNet8|GhostNet|
|:---|:---:|:---:|
|Accuracy|0.99340|0.99370|
|Precision|0.99339|0.99373|
|Recall|0.99329|0.99357|
|F1-Score|0.99334|0.99364|
### ConvNet8
```
Confusion Matrix
[[ 979 0 0 0 0 0 0 1 0 0]
[ 0 1132 0 1 0 0 1 1 0 0]
[ 0 0 1029 0 0 0 0 3 0 0]
[ 0 0 1 1006 0 3 0 0 0 0]
[ 0 0 1 0 975 0 2 0 0 4]
[ 1 0 0 7 0 882 1 0 0 1]
[ 4 2 0 0 0 1 950 0 1 0]
[ 1 3 3 2 0 0 0 1018 1 0]
[ 3 0 1 1 0 1 0 0 966 2]
[ 0 0 0 1 6 2 0 3 0 997]]
Class-0 | Precision: 0.99089, Recall: 0.99898, F1-Score: 0.99492
Class-1 | Precision: 0.99560, Recall: 0.99736, F1-Score: 0.99648
Class-2 | Precision: 0.99420, Recall: 0.99709, F1-Score: 0.99565
Class-3 | Precision: 0.98821, Recall: 0.99604, F1-Score: 0.99211
Class-4 | Precision: 0.99388, Recall: 0.99287, F1-Score: 0.99338
Class-5 | Precision: 0.99213, Recall: 0.98879, F1-Score: 0.99045
Class-6 | Precision: 0.99581, Recall: 0.99165, F1-Score: 0.99372
Class-7 | Precision: 0.99220, Recall: 0.99027, F1-Score: 0.99124
Class-8 | Precision: 0.99793, Recall: 0.99179, F1-Score: 0.99485
Class-9 | Precision: 0.99303, Recall: 0.98811, F1-Score: 0.99056
Total | Accuracy: 0.99340, Precision: 0.99339, Recall: 0.99329, F1-Score: 0.99334
```
### GhostNet
```
Confusion Matrix
[[ 977 0 0 0 0 0 2 1 0 0]
[ 0 1131 1 1 0 0 1 1 0 0]
[ 1 1 1028 1 0 0 0 1 0 0]
[ 0 0 0 1008 0 1 0 1 0 0]
[ 0 0 2 0 972 0 4 0 1 3]
[ 1 0 0 7 0 882 1 0 0 1]
[ 4 0 3 1 0 1 947 0 2 0]
[ 0 2 3 0 0 0 0 1022 0 1]
[ 1 0 2 1 0 0 0 1 968 1]
[ 0 0 0 1 5 0 0 1 0 1002]]
Class-0 | Precision: 0.99289, Recall: 0.99694, F1-Score: 0.99491
Class-1 | Precision: 0.99735, Recall: 0.99648, F1-Score: 0.99691
Class-2 | Precision: 0.98941, Recall: 0.99612, F1-Score: 0.99276
Class-3 | Precision: 0.98824, Recall: 0.99802, F1-Score: 0.99310
Class-4 | Precision: 0.99488, Recall: 0.98982, F1-Score: 0.99234
Class-5 | Precision: 0.99774, Recall: 0.98879, F1-Score: 0.99324
Class-6 | Precision: 0.99162, Recall: 0.98852, F1-Score: 0.99007
Class-7 | Precision: 0.99416, Recall: 0.99416, F1-Score: 0.99416
Class-8 | Precision: 0.99691, Recall: 0.99384, F1-Score: 0.99537
Class-9 | Precision: 0.99405, Recall: 0.99306, F1-Score: 0.99355
Total | Accuracy: 0.99370, Precision: 0.99373, Recall: 0.99357, F1-Score: 0.99364
```
## Requirements
* Python 3.6.8
* Tensorflow 1.14.0
* Numpy 1.17.1
* Matplotlib 3.1.1
## Reference
[1] Kai Han et al. GhostNet: More Features from Cheap Operations
. arXiv preprint arXiv:1911.119075 (2019).