3D-ConvNeuralNet-material-property-prediction
所属分类:人工智能/神经网络/深度学习
开发工具:Python
文件大小:0KB
下载次数:1
上传日期:2020-08-15 01:13:40
上 传 者:
sh-1993
说明: 本报告包含用于材料性能预测的3D卷积神经网络的数据集代码,
(This repo includes the dataset code of 3D Convolutional Neural Network for material property prediction,)
文件列表:
data/ (0, 2020-08-14)
data/download_link.txt (171, 2020-08-14)
gallery/ (0, 2020-08-14)
gallery/NN_archi.png (134417, 2020-08-14)
gallery/Pred_vs_Truth_test.png (1022656, 2020-08-14)
gallery/clous_pt.png (6481926, 2020-08-14)
saved_models/ (0, 2020-08-14)
saved_models/my_best_model_CNN_new.h5 (19178352, 2020-08-14)
training&plot.py (12961, 2020-08-14)
# 3D-ConvNeuralNet-material-property-prediction
# Authors and citation
This repo includes the dataset/scripts of 3D Convolutional Neural Network for material property prediction in paper:
[Rao, Chengping, and Yang Liu. "Three-dimensional convolutional neural network (3D-CNN) for heterogeneous material homogenization." Computational Materials Science 184 (2020): 109850.](https://www.sciencedirect.com/science/article/abs/pii/S0927025620303414)
```
@article{rao2020three,
title={Three-dimensional convolutional neural network (3D-CNN) for heterogeneous material homogenization},
author={Rao, Chengping and Liu, Yang},
journal={Computational Materials Science},
volume={184},
pages={109850},
year={2020},
publisher={Elsevier}
}
```
# Description for each file
- **training&plot.py**: Data processing, training and postprocessing scripts ;
- **data**: The dataset for this application contains 2000 samples or (**X, Y**) pairs. **X** is the tensor of size 101x101x101 that carries the phase indication for a two-phase heterogeneous material microstructure (or Representative Volume Element). **Y** is a vector of size 12 that represents the effective material properties for the microstructure.
- **saved_model**: The model with best performance on validation set.
- **gallery**: Some figures.
# Overview
> Input **X** for the 3D Convolutional Neural Net, i.e. Cartisian grid carrying phase indication
![](https://github.com/Raocp/3D-ConvNeuralNet-material-property-prediction/blob/master/gallery/NN_archi.png?raw=true)
> Architecture of the 3D Convolutional Neural Net
> Achieved prediction versus ground truth on testing data
# Note
- The code was developed based on TensorFlow 1.10.0 and Keras 2.2.4. All the runtime performance are evaluated on GeForce GTX 1080 Ti.
近期下载者:
相关文件:
收藏者: