Deep Learning(深度学习)学习笔记整理系列之(二)

2017-05-29 微天使 Deep Learning(深度学习)学

Deep Learning(深度学习)学习笔记整理系列之(二)

四、关于特征        特征是机器学习系统的原材料,对最终模型的影响是毋庸置疑的。如果数据被很好的表达成了特征,通常线性模型就能达到满意的精度。那对于特征,我们需要考虑什么呢?4.1、特征表示的粒度        学习算法在一个什么粒度上的特征表示,才有能发挥作用?就一个图片来说,像素级的特征根本没有价值。例如下面的摩托车,从像素级别,根本得不到任何信息,其无法进行摩托车和非摩托车的区分。而如果特征是一个具有结构性(或者说有含义)的时候,比如是否具有车把手(handle),是否具有车轮(wheel),就很容易把摩托车和非摩托车区分,学习算法才能发挥作用。  4.2、初级(浅层)特征表示        既然像素级的特征表示方法没有作用,那怎样的表示才有用呢?        1995 年前后

版权声明:
本网站所有内容来源注明为“梅斯医学”或“MedSci原创”的文字、图片和音视频资料,版权均属于梅斯医学所有。非经授权,任何媒体、网站或个人不得转载,授权转载时须注明来源为“梅斯医学”。其它来源的文章系转载文章,或“梅斯号”自媒体发布的文章,仅系出于传递更多信息之目的,本站仅负责审核内容合规,其内容不代表本站立场,本站不负责内容的准确性和版权。如果存在侵权、或不希望被转载的媒体或个人可与我们联系,我们将立即进行删除处理。
在此留言
评论区 (3)
#插入话题
  1. [GetPortalCommentsPageByObjectIdResponse(id=1624575, encodeId=571b16245e57d, content=<a href='/topic/show?id=3f2d2e7087' target=_blank style='color:#2F92EE;'>#ARNI#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=39, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=2770, encryptionId=3f2d2e7087, topicName=ARNI)], attachment=null, authenticateStatus=null, createdAvatar=null, createdBy=f41c20676233, createdName=gongliu, createdTime=Wed May 31 14:04:00 CST 2017, time=2017-05-31, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=204225, encodeId=5b6320422592, content=学习了谢谢分享。, beContent=null, objectType=article, channel=null, level=null, likeNumber=50, replyNumber=0, topicName=null, topicId=null, topicList=[], attachment=null, authenticateStatus=null, createdAvatar=https://wx.qlogo.cn/mmopen/oLAjfB7s1ib06iaV3D9Tg5Kzuf9u71gZhPYMiajRtENwicoAABeQtfXOlic8ibhYSy6DvJZWUDVtjRvTfqBffr1XJ6JLtFB5kHicthl/0, createdBy=08bb2061153, createdName=189****7206, createdTime=Tue May 30 06:37:02 CST 2017, time=2017-05-30, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=204159, encodeId=a3d420415962, content=学习了很有用, beContent=null, objectType=article, channel=null, level=null, likeNumber=54, replyNumber=0, topicName=null, topicId=null, topicList=[], attachment=null, authenticateStatus=null, createdAvatar=https://wx.qlogo.cn/mmopen/oLAjfB7s1ib3IzoRn621JdKCFUI4XvNELcibocrAVU6aUzEVDSaf3rYH08PUYUiagGqxAEhj8g9WqqNvD1CupIpCrmnR2OrqHZh/0, createdBy=7f181365728, createdName=飛歌, createdTime=Tue May 30 00:07:05 CST 2017, time=2017-05-30, status=1, ipAttribution=)]
    2017-05-31 gongliu
  2. [GetPortalCommentsPageByObjectIdResponse(id=1624575, encodeId=571b16245e57d, content=<a href='/topic/show?id=3f2d2e7087' target=_blank style='color:#2F92EE;'>#ARNI#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=39, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=2770, encryptionId=3f2d2e7087, topicName=ARNI)], attachment=null, authenticateStatus=null, createdAvatar=null, createdBy=f41c20676233, createdName=gongliu, createdTime=Wed May 31 14:04:00 CST 2017, time=2017-05-31, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=204225, encodeId=5b6320422592, content=学习了谢谢分享。, beContent=null, objectType=article, channel=null, level=null, likeNumber=50, replyNumber=0, topicName=null, topicId=null, topicList=[], attachment=null, authenticateStatus=null, createdAvatar=https://wx.qlogo.cn/mmopen/oLAjfB7s1ib06iaV3D9Tg5Kzuf9u71gZhPYMiajRtENwicoAABeQtfXOlic8ibhYSy6DvJZWUDVtjRvTfqBffr1XJ6JLtFB5kHicthl/0, createdBy=08bb2061153, createdName=189****7206, createdTime=Tue May 30 06:37:02 CST 2017, time=2017-05-30, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=204159, encodeId=a3d420415962, content=学习了很有用, beContent=null, objectType=article, channel=null, level=null, likeNumber=54, replyNumber=0, topicName=null, topicId=null, topicList=[], attachment=null, authenticateStatus=null, createdAvatar=https://wx.qlogo.cn/mmopen/oLAjfB7s1ib3IzoRn621JdKCFUI4XvNELcibocrAVU6aUzEVDSaf3rYH08PUYUiagGqxAEhj8g9WqqNvD1CupIpCrmnR2OrqHZh/0, createdBy=7f181365728, createdName=飛歌, createdTime=Tue May 30 00:07:05 CST 2017, time=2017-05-30, status=1, ipAttribution=)]
    2017-05-30 189****7206

    学习了谢谢分享。

    0

  3. [GetPortalCommentsPageByObjectIdResponse(id=1624575, encodeId=571b16245e57d, content=<a href='/topic/show?id=3f2d2e7087' target=_blank style='color:#2F92EE;'>#ARNI#</a>, beContent=null, objectType=article, channel=null, level=null, likeNumber=39, replyNumber=0, topicName=null, topicId=null, topicList=[TopicDto(id=2770, encryptionId=3f2d2e7087, topicName=ARNI)], attachment=null, authenticateStatus=null, createdAvatar=null, createdBy=f41c20676233, createdName=gongliu, createdTime=Wed May 31 14:04:00 CST 2017, time=2017-05-31, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=204225, encodeId=5b6320422592, content=学习了谢谢分享。, beContent=null, objectType=article, channel=null, level=null, likeNumber=50, replyNumber=0, topicName=null, topicId=null, topicList=[], attachment=null, authenticateStatus=null, createdAvatar=https://wx.qlogo.cn/mmopen/oLAjfB7s1ib06iaV3D9Tg5Kzuf9u71gZhPYMiajRtENwicoAABeQtfXOlic8ibhYSy6DvJZWUDVtjRvTfqBffr1XJ6JLtFB5kHicthl/0, createdBy=08bb2061153, createdName=189****7206, createdTime=Tue May 30 06:37:02 CST 2017, time=2017-05-30, status=1, ipAttribution=), GetPortalCommentsPageByObjectIdResponse(id=204159, encodeId=a3d420415962, content=学习了很有用, beContent=null, objectType=article, channel=null, level=null, likeNumber=54, replyNumber=0, topicName=null, topicId=null, topicList=[], attachment=null, authenticateStatus=null, createdAvatar=https://wx.qlogo.cn/mmopen/oLAjfB7s1ib3IzoRn621JdKCFUI4XvNELcibocrAVU6aUzEVDSaf3rYH08PUYUiagGqxAEhj8g9WqqNvD1CupIpCrmnR2OrqHZh/0, createdBy=7f181365728, createdName=飛歌, createdTime=Tue May 30 00:07:05 CST 2017, time=2017-05-30, status=1, ipAttribution=)]
    2017-05-30 飛歌

    学习了很有用

    0

相关资讯

2017医药营销会呈现哪些态势?

大量行业政策的推行,2017年医药行业处于商业基础框架崩溃的边缘。

医药电商前景广阔,但是有3个建议你必须知道

医药电商的发展单纯的依靠网上平台的非目标性销售是难以做成规模的

内分泌失调是女性患上三大瘤的根源

女性患上三大瘤罪魁祸首竟是内分泌失调,了解内在密切关系,为治疗找到更加准确的病源,寻因而治、斩草除根!

Deep Learning(深度学习)学习笔记整理系列之(一)

Deep Learning(深度学习)学习笔记整理系列之(一)

干掉医药营销传统思维,打造适应性产品群

传统营销的时代已经过去,医药营销的新时代已经到来!

监管趋严,医药商业未来的日子怎么过?

目前医药商业企业面临的首先是以怎样国家认可的模式生存下去