Webdataset for PyTorch
After switching to Pytorch for deep learning projects, I kept looking for a dataset format that can give me as good performance as TFRecord. First I tried using tensorflow to save and load TFRecor...
After switching to Pytorch for deep learning projects, I kept looking for a dataset format that can give me as good performance as TFRecord. First I tried using tensorflow to save and load TFRecor...
Recently I want to get BPnet running, but setting up environment for it is a bit tricky, so I wrote down the efforts I spent on this for future reference or anyone having trouble like me. The B...
No doubt, doing bioinformatic research requires your creativity, you need to work very hard to find the special pattern in a bunch of meaningless signal tracks, linking them to genes, making your c...
前言 latex是一种便捷的编写数学公式的语言。之前我在word中编写公式使用的是Mathtype,虽然好用,但是要收费。最近我在学习使用markdown来编写文档,而markdown对latex有着非常棒的支持,这里记录一下使用的方法。 基础语法 latex数学公式的引用有两种形式,分别是行内公式和独立公式 行内公式:用$formula$表示,例如$\sum_{i=0}^{n...
前言 之前我写了一个简单的模块numpyArrayDict.py来保存genome coverage信息。通过这个方法可以将reads在genome上的per base coverage计算完后保存在同一个HDF5文件内,而且一些其他类型的per base info(例如DNA methylation data处理后得到的每个C上甲基化的比例)也可以存储。由于h5py支持直接从文件中读取出...