知识卡片
前向传播计算当前参数下的输出
前向传播把样本依次经过层操作得到得分或预测,训练还需损失与参数更新。能完成784维到10维的形状流转,只说明计算接口连通,不证明手写数字识别已有准确率。
这张卡片看不懂?
这张卡片已通过人工审查,不再接受失败反馈。
已有反馈 · 0 条
还没有人提交反馈。
在整书笔记中的位置
- · 深度学习模型
原书出处 · 1 处
- · 查看原书依据
Our MLP will operate by taking a grayscale image as input. Specifically, each pixel value of the image will be used as input. Next, the sample will be forward propagated through the MLP until it reaches the output. Forward propagation refers to the input going through each of the hidden layers and incurring their respective mathematical operations (multiplication by weights, added with bias, and passed
参考来源
- 位置:《Machine Learning Fundamentals: Concepts, Models, and Applications》第08章《深度学习模型》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Machine Learning Fundamentals: Concepts, Models, and Applications/_epub-src/OEBPS/xhtml/21_chapter08.xhtml)
- 结论依据:摘录中关于前向传播计算当前参数下的输出的定义、比较或条件支撑卡片;原书表述不严谨的范围已在覆盖记录逐项限定,不把案例结果外推为普遍结论。
- 原始内容:Our MLP will operate by taking a grayscale image as input. Specifically, each pixel value of the image will be used as input. Next, the sample will be forward propagated through the MLP until it reaches the output. Forward propagation refers to the input going through each of the hidden layers and incurring their respective mathematical operations (multiplication by weights, added with bias, and passed