Automated Feature Engineering: Featuretools
content
思考:
-
当特征深度过深大于2时的特征可解释性?
-
生成的特征过多带来新的问题:维度诅咒,那么如何选取特征?
feature reduction and selection …
more ...思考:
当特征深度过深大于2时的特征可解释性?
生成的特征过多带来新的问题:维度诅咒,那么如何选取特征?
feature reduction and selection …
more ...理解虚表
理解全文本查找
https://www.sqlite.org/fts5.html
http://www.sqlitetutorial.net/sqlite-full-text-search
def _is_to_exec(self, sha1):
"""Check whether the query has been run in the past.
"""
sql = f'''
SELECT DISTINCT
sha1
FROM
queries
WHERE queries …
# Definition for a Directed graph node
class DirectedGraphNode:
def __init__(self, x):
self.label = x
self.neighbors = []
class Solution:
"""
@param graph: A list of Directed graph node
@return: A list of integer
"""
def topSort(self, graph):
# 1. 统计结点入度
indegree = self.get_indegree(graph …
给定数据集\((x_1, y_1), (x_2, y_2),...,(x_n, y_n)\), 拟合一个模型\(F(x)\)。
\(F(x_1) = 1.4\)而\(y_1=1.3\)
\(F(x_2) = 0.9\)而\(y_2=0.8\)
...
在 …
more ...