![](https://img-blog.csdnimg.cn/20201014180756724.png?x-oss-process=image/resize,m_fixed,h_64,w_64)
深度学习
伽音
数据挖掘、机器学习、深度学习、自然语言处理
展开
-
NLP实践-Task7
任务链接:https://wx.zsxq.com/dweb/#/index/2222484248111.卷积参考链接:https://zhuanlan.zhihu.com/p/575758101.1卷积卷积:图像中不同数据窗口的数据和卷积核(一个滤波矩阵)作内积的操作叫做卷积。其计算过程又称为滤波,本质是提取图像不同频段的特征。卷积核:也称为滤波器filter,带着一组固定权重的...原创 2019-03-14 19:35:34 · 506 阅读 · 2 评论 -
NLP实践-Task5
任务链接:https://wx.zsxq.com/dweb/#/index/222248424811深度学习视频推荐1:https://www.icourse163.org/learn/PKU-1002536002?tid=1003797005#/learn/content深度学习视频推荐2:https://mooc.study.163.com/course/2001281002#/info...原创 2019-03-11 10:54:47 · 320 阅读 · 0 评论 -
NLP实践-Task8
1.循环神经网络参考链接1:https://blog.csdn.net/u011304078/article/details/81158370参考链接2:https://blog.csdn.net/zhaojc1995/article/details/80572098参考链接3:https://zybuluo.com/hanbingtao/note/541458参考链接4:h...原创 2019-03-17 22:18:58 · 435 阅读 · 0 评论 -
pytorch-task4
import osimport csvimport jiebaimport warningsimport numpy as npimport pandas as pdfrom tqdm import tqdmfrom gensim.models import Word2Vecimport torchimport torch.nn as nnfrom torch.optim i...原创 2019-08-13 19:31:49 · 163 阅读 · 0 评论 -
pytorch-task2
numpy实现梯度下降x = 1learning_rate = 0.1epochs = 50y = lambda x : x ** 2 - 1for epoch in range(epochs): print(epoch, x) dx = 2 * x x = x - learning_rate * dxprint(y(x))pytorch实现梯度下降...原创 2019-08-09 19:54:18 · 208 阅读 · 0 评论 -
pytorch-task3
pytorch实现逻辑回归import torchimport torch.nn as nnfrom torch.optim import SGDclass LR(nn.Module): def __init__(self): super(LR, self).__init__() self.linear = nn.Linear(2, 1) ...原创 2019-08-11 19:04:14 · 165 阅读 · 1 评论