Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/Shathe/MNasNet-Keras-Tensorflow
A Tensorflow Keras implementation (Graph and eager execution) of Mnasnet: MnasNet: Platform-Aware Neural Architecture Search for Mobile.
https://github.com/Shathe/MNasNet-Keras-Tensorflow
architecture-search deep-learning eager-execution keras mnasnet tensorflow
Last synced: about 1 month ago
JSON representation
A Tensorflow Keras implementation (Graph and eager execution) of Mnasnet: MnasNet: Platform-Aware Neural Architecture Search for Mobile.
- Host: GitHub
- URL: https://github.com/Shathe/MNasNet-Keras-Tensorflow
- Owner: Shathe
- Created: 2018-08-08T18:32:26.000Z (over 6 years ago)
- Default Branch: master
- Last Pushed: 2020-03-05T14:54:38.000Z (almost 5 years ago)
- Last Synced: 2024-08-01T22:50:09.215Z (4 months ago)
- Topics: architecture-search, deep-learning, eager-execution, keras, mnasnet, tensorflow
- Language: Python
- Homepage:
- Size: 219 KB
- Stars: 98
- Watchers: 10
- Forks: 29
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
- awesome-image-classification - unofficial-keras : https://github.com/Shathe/MNasNet-Keras-Tensorflow
- awesome-image-classification - unofficial-keras : https://github.com/Shathe/MNasNet-Keras-Tensorflow
README
# MNasNet
[Keras (Tensorflow) Implementation](https://github.com/Shathe/MNasNet-Keras-Tensorflow/blob/master/Mnasnet.py) of MNasNet and an example for training and evaluating it on the MNIST dataset.
Check also the [eager execution implementation](https://github.com/Shathe/MNasNet-Keras-Tensorflow/blob/master/MnasnetEager.py)
According to the paper: [MnasNet: Platform-Aware Neural Architecture Search for Mobile](https://arxiv.org/pdf/1807.11626.pdf)
## Requirement
* Python 2.7+
* Tensorflow-gpu 1.10
## Train it
Train the [MNasNet model](https://github.com/Shathe/MNasNet-Keras-Tensorflow/blob/master/Mnasnet.py) on the MNIST dataset! just execute:
```
python train.py
```
For checking and inspecting the Mnasnet model described in the paper, execute:
```
python Mnasnet.py
```
## Train it with eager execution
Train the [MNasNet (eager) model](https://github.com/Shathe/MNasNet-Keras-Tensorflow/blob/master/MnasnetEager.py) on the MNIST dataset! just execute:
```
python train_eager.py
```
The eager execution implementation also outputs logs on Tensorboard. For its visualization:
```
tensorboard --logdir=train_log:./logs/train, test_log:./logs/test
```
## MnasNet for... Semantic Segmentation!
In this other repository, [FC-Mnasnet](https://github.com/Shathe/Semantic-Segmentation-Tensorflow-Eager) I added a decoder to the MnasNet architecture in order to turn it into a semantic segmentation model.
