tensorlayer3/tests/performance_test/vgg/README.md

35 lines
1.3 KiB
Markdown

# Performance Test (VGG16)
### Introduction
This test compares performance of the following libraries:
1. TensorLayer v. 2.0
2. TensorFlow ([tf.keras](https://www.tensorflow.org/api_docs/python/tf/keras)) v. 2.0.0-alpha
3. [Keras](https://keras.io/) v. 2.2.4
### With GPU
__Hardware:__
- CPU: Intel(R) Xeon(R) CPU E5-2630 v4 @ 2.20GHz 40 core
- GPU: TITAN Xp
__Experiment Settings:__
- Model: [VGG16](http://www.robots.ox.ac.uk/~vgg/research/very_deep/)
- Batch size: 32
- Number of iterations: 300
__Results:__
| Mode | Lib | Data Format | Max GPU Memory Usage(MB) |Max CPU Memory Usage(MB) | Avg CPU Memory Usage(MB) | Runtime (sec) |
| :-------: | :-------------: | :-----------: | :-----------------: | :-----------------: | :-----------------: | :-----------: |
| AutoGraph | TensorFlow 2.0 | channel last | 11833 | 2161 | 2136 | 74 |
| | Tensorlayer 2.0 | channel last | 11833 | 2187 | 2169 | 76 |
| Graph | Keras | channel last | 8677 | 2580 | 2576 | 101 |
| Eager | TensorFlow 2.0 | channel last | 8723 | 2052 | 2024 | 97 |
| | TensorLayer 2.0 | channel last | 8723 | 2010 | 2007 | 95 |