• 设为首页
  • 点击收藏
  • 手机版
    手机扫一扫访问
    迪恩网络手机版
  • 关注官方公众号
    微信扫一扫关注
    公众号

JJAlmagro/subcellular_localization

原作者: [db:作者] 来自: 网络 收藏 邀请

开源软件名称(OpenSource Name):

JJAlmagro/subcellular_localization

开源软件地址(OpenSource Url):

https://github.com/JJAlmagro/subcellular_localization

开源编程语言(OpenSource Language):

Jupyter Notebook 92.1%

开源软件介绍(OpenSource Introduction):

PREDICTION OF PROTEIN SUBCELLULAR LOCALIZATION

Synopsis

Convolutional Bidirectional LSTM with attention mechanism for predicting protein subcellular localization. The model was trained using the MultiLoc dataset, which counts with 5959 proteins.

Author

Jose Juan Almagro Armenteros, DTU Bioinformatics

Protein data

There are two files in the data folder: 1) "test.npz" independent set to calculate the final performance of the model 2) "train.npz" training set.

Each file includes a numpy array with the proteins sequences already encoded in profiles, a numpy array with the masks of each sequence and a numpy vector with the target of each protein.

Training

The training is performed running the script "train.py". This is a minimal example:

python train.py -i train.npz -t test.npz

The default options are the optimals one, but the training will be really slow on CPU.

To run it on GPU use these flags before the command

THEANO_FLAGS=device=gpu0,dnn.conv.algo_fwd=time_once,dnn.conv.algo_bwd_filter=time_once,dnn.conv.algo_bwd_data=time_once,warn_float64=warn python train.py -i train.npz -t test.npz

Contributors

Ole Winther, DTU Compute Henrik Nielsen, DTU Bioinformatics Søren and Casper Kaae Sønderby, University of Copenhagen




鲜花

握手

雷人

路过

鸡蛋
该文章已有0人参与评论

请发表评论

全部评论

专题导读
上一篇:
chenjoya/2dtan: An optimized re-implementation for 2D-TAN: Learning 2D Temporal ...发布时间:2022-08-15
下一篇:
udacity/RoboND-Localization-Project发布时间:2022-08-15
热门推荐
阅读排行榜

扫描微信二维码

查看手机版网站

随时了解更新最新资讯

139-2527-9053

在线客服(服务时间 9:00~18:00)

在线QQ客服
地址:深圳市南山区西丽大学城创智工业园
电邮:jeky_zhao#qq.com
移动电话:139-2527-9053

Powered by 互联科技 X3.4© 2001-2213 极客世界.|Sitemap