Waste-Sorter-master

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:2271KB
下载次数:6
上传日期:2020-05-28 13:30:29
上 传 者5496981
说明:  训练一个卷积神经网络,用fastai库(建在PyTorch上)将图像分类为纸板,玻璃,金属,纸张,塑料或垃圾。数据集下载地址如下。
(Train a convolutional neural network to classify images into cardboard, glass, metal, paper, plastic or garbage using the fastai Library (built on pytorch). The data set download address is as follows.)

文件列表:
Waste sorter.ipynb (3143398, 2019-08-16)

# Waste-Sorter This was my first attempt at building an image classifier using the fastai library, trained to sort different types of trash. The test accuracy was really high, 92%. I also wrote a [code walkthrough](https://towardsdatascience.com/how-to-build-an-image-classifier-for-waste-sorting-6d11d3c9c478) on Medium. All the code for this project is in **`Waste sorter.ipynb`**. ## How to view this project The notebook is too big to render on Github, which is why we have to use Jupyter Notebook Viewer to render it. Start with option (a) and if the link doesn't work, use option (b). ### Option (a) - Click [here](https://nbviewer.jupyter.org/github/collindching/Waste-Sorter/blob/master/Waste%20sorter.ipynb) ### Option (b) 1. Copy the link to the notebook on Github (https://github.com/collindching/Waste-Sorter/blob/master/Waste%20sorter.ipynb) 2. Follow this link [here](https://nbviewer.jupyter.org/) and paste the notebook URL to view the notebook ## How to run this notebook 1. Download **`Waste sorter.ipynb`** 2. Download the corresponding dataset [here](https://github.com/garythung/trashnet/blob/master/data/dataset-resized.zip) and move it to the same directory as the notebook 3. Run the notebook in Anaconda

近期下载者

相关文件


收藏者