using namespace 数学工具构造器;

A, M = X.shape B, _ = self.tsne_matrix.shape assert M == self.n_components # X, (A,M)->(A,B,M) matrix1 = np.repeat(X, B, axis=0) matrix1=matrix1.reshape([A,B,M]) # tsne_matrix, # (B,M)-&...

#!/usr/bin/env python# -*- coding: utf-8 -*-# @Author : qichun tang# @Contact : tqichun@gmail.comfrom pathlib import PathMakefile = "Makefile"lines = Path(Makefile).read_text().split("\n")packages = Path("requirements.txt").read_text().split("\

#!/usr/bin/env python# -*- coding: utf-8 -*-# @Author : qichun tang# @Contact : tqichun@gmail.comfrom copy import deepcopyfrom pickle import loads, dumpsclass Example(): def __init__(self, data): self.data = data def __reduce__

plotTree.py# -*- coding: utf-8 -*-import sys import osimport numpy as npimport matplotlib.pyplot as pltimport treePlotter as tp # 绘制树myTree = {'root': {0: 'leaf node', 1: {'level 2': {...

def get_df(path)->pd.DataFrame: data=[] header=['id','score'] lines=Path(path).read_text().split('\n') for line in lines: words=line.split(' ') while '' in words: ...

1. 减少循环层数修改前 FeatureFlat_Euc = np.zeros([len(Bins_Euc) - 1, len(ProEleList) * len(LigEleList)], float) FeatureFlat_Ele = np.zeros([len(Bins_Ele) - 1, len(ProEleList) * len(LigEleList)], flo...

文章目录1.问题背景1.1 click终端交互结构1.2 关于Top level script environment1.3 multiprocessing只能在Top-level的代码环境中执行2. 问题解决2.1 通过在PYTHONPATH中设置新地址，并重新导入模块来解决1.问题背景在scorefunc开发中，我用click模块来组织其整个项目的目录结构和终端交互的结构。但是，被clic...

multiproccessingimport multiprocessing as mpcount = 0def job(id): global count count += 1 print(f'job {id}, {count}')pool = mp.Pool(4)for i in range(100): pool.apply_async(jo...

In [9]: i=0 ...: while(i<len(lst)): ...: if lst[i]==2: ...: del lst[i] ...: else: ...: i+=1 ...: print(lst) ...: ...

ImageNet数据集预处理：如果一个图片的长宽比例小于一定比值，则直接拉伸。如果大于一定比值，按照 左、中、右 三部分进行切割def process(img,max_ratio=1.35): if len(img.shape)==2: img=cv2.cvtColor(img,cv2.COLOR_GRAY2RGB) h,w,_=img.shape r...

我在保存csv的时候，有时候用了Index=False，有的时候没有，就造成读取feature的训练数据的时候，不知道要不要删除第一列。于是可以用这个函数进行判断：def isDataFrameRangeIndex(df:pd.DataFrame,col): return (df.iloc[:, col].values == np.arange(df.shape[0])).all()...

def symlink(origin:Path,target:Path): if target.exists(): os.remove(target.as_posix()) if target.is_symlink(): os.unlink(target.as_posix()) try: os.symlink(origin.a...

import tarfiledef init_dataset(datapath): if Path(datapath).is_file() and datapath.endswith('.tar.gz'): parent=Path(datapath).parent tar = tarfile.open(datapath) dirname=...

def split_df(df, nums): ''' reverse operation: pd.concat(dfs) ''' rows = df.shape[0] interval = rows // nums dfs = [] for i in range(nums): s = i * interval ...

昨天打包的时候，有个bug就是没有把一个路径很深的文件夹打进去。因为Python调用了那个文件夹中的一些bash脚本，出错。折腾了接近一个小时，记录一下。import pdbfrom setuptools import setup, find_packagesfrom pathlib import Pathimport osdef get_recursion_file_list...

CLI@click.option('--ncpu', '-j', type=int,help='How many CPUs to utilize,' ' 0 or 1 means single process,' ' -1...