bb_sy_w的博客

问题描述某些情况下需要将卷积/池化层后的features map的某些patch映射回原图像，就需要features map怎么变过来的怎么回去。#read imageself.object_detector = torchvision.models.detection.maskrcnn_resnet50_fpn(pretrained = True, trainable_backbone_layers = 0)images_nested = torchvision.models.detection.

由随机决策树组成的森林

在验证集上简单打分import numpy as npimport pandas as pdfrom sklearn.model_selection import train_test_splitfrom sklearn.tree import DecisionTreeRegressor#load datasetdata = pd.read_csv('train.csv')data = data.iloc[:,[1,3,4,80]]data = data.fillna(0)X_tra

官方中文文档监督学习回归分类决策树这里我们采用sklearn的内置数据试验from sklearn import treeimport graphvizfrom sklearn import datasets#加载内置数据集iris = datasets.load_iris()#实例化，采用哪种方法clf = tree.DecisionTreeClassifier()#训练clf.fit(iris.data,iris.target)#画出决策树dot_data = t