Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/calmiLovesAI/TensorFlow2.0_Image_Classification
A TensorFlow_2.0 implementation of AlexNet and VGGNet.
https://github.com/calmiLovesAI/TensorFlow2.0_Image_Classification
Last synced: 2 months ago
JSON representation
A TensorFlow_2.0 implementation of AlexNet and VGGNet.
- Host: GitHub
- URL: https://github.com/calmiLovesAI/TensorFlow2.0_Image_Classification
- Owner: calmiLovesAI
- License: mit
- Created: 2019-07-09T08:26:47.000Z (over 5 years ago)
- Default Branch: master
- Last Pushed: 2023-01-06T02:43:07.000Z (about 2 years ago)
- Last Synced: 2024-10-27T15:27:40.014Z (3 months ago)
- Language: Python
- Homepage:
- Size: 203 KB
- Stars: 39
- Watchers: 2
- Forks: 18
- Open Issues: 3
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
- Awesome-Tensorflow2 - calmisential/TensorFlow2.0_Image_Classification
README
# TensorFlow2.0_Image_Classification(include AlexNet and VGGNet)
This project uses TensorFlow2.0 for image classification tasks.
## How to use
### Requirements
+ **Python 3.x** (My Python version is 3.6.8)
+ **TensorFlow version: 2.0.0-beta1**
+ The file directory of the dataset should look like this:
```
${dataset_root}
|——train
| |——class_name_0
| |——class_name_1
| |——class_name_2
| |——class_name_3
|——valid
| |——class_name_0
| |——class_name_1
| |——class_name_2
| |——class_name_3
|——test
|——class_name_0
|——class_name_1
|——class_name_2
|——class_name_3
```
### Train
Run the script
```
python train.py
```
to train the network on your image dataset, the final model will be stored. You can also change the corresponding training parameters in the `config.py`.
### Evaluate
To evaluate the model's performance on the test dataset, you can run `evaluate.py`.
The structure of the network is defined in `model_definition.py`, you can change the network structure to whatever you like.
## References
1. AlexNet : http://www.cs.toronto.edu/~fritz/absps/imagenet.pdf
2. VGG : https://arxiv.org/abs/1409.1556