简体繁体 English

使用Caffe进行深度学习 - Python

[英]Deep learning using Caffe - Python

原文 2017-06-07 04:50:14 9 1 python/ machine-learning/ neural-network/ deep-learning/ caffe

I am studying deep learning and trying to implement it using CAFFE- Python. 我正在研究深度学习并尝试使用CAFFE-Python实现它。 can anybody tell that how we can assign the weights to each node in input layer instead of using weight filler in caffe? 任何人都可以告诉我们如何为输入层中的每个节点分配权重而不是在caffe中使用权重填充？

1 个解决方案

There is a fundamental difference between weights and input data : the training data is used to learn the weights (aka "trainable parameters") during training. 权重和输入数据之间存在根本区别：训练数据用于在训练期间学习权重（也称为“可训练参数”）。 Once the net is trained, the training data is no longer needed while the weights are kept as part of the model to be used for testing/deployment. 一旦网络被训练，就不再需要训练数据，而权重被保留作为用于测试/部署的模型的一部分。
Make sure this difference is clear to you before you precede. 在您出现之前，请确保您已明白此区别。

Layers with trainable parameters has a filler to set the weights initially. 具有可训练参数的图层具有filler以初始设置权重。
On the other hand, an input data layer does not have trainable parameters, but it should supply the net with input data . 另一方面，输入数据层没有可训练的参数，但它应该为网络提供输入数据 。 Thus, input layers has no filler . 因此，输入层没有filler 。
Based on the type of input layer you use, you will need to prepare your training data. 根据您使用的输入图层类型，您需要准备训练数据。