MCPcopy Create free account

hub / github.com/Jack-Cherish/Deep-Learning / functions

Functions315 in github.com/Jack-Cherish/Deep-Learning

⨍ Functions 315◇ Types & classes 61
↓ 12 callersMethodinit_state_vec
初始化保存状态的向量
Tutorial/lesson-6/lstm.py:44
↓ 10 callersMethodforward
根据式1-式6进行前向计算
Tutorial/lesson-6/lstm.py:64
↓ 9 callersMethodpredict
输入向量，输出感知器的计算结果
Tutorial/lesson-2/perceptron.py:22
↓ 9 callersMethodupdate
(self, val, n=1)
Pytorch-Seg/lesson-4/train.py:169
↓ 8 callersMethodforward
前向计算
Tutorial/lesson-7/recursive.py:41
↓ 8 callersMethodforward
根据『式2』进行前向计算
Tutorial/lesson-5/rnn.py:24
↓ 6 callersMethodbackward
实现LSTM训练算法
Tutorial/lesson-6/lstm.py:102
↓ 6 callersMethodforward
计算卷积层的输出 输出结果保存在self.output_array
Tutorial/lesson-6/cnn.py:152
↓ 6 callersMethodforward
计算卷积层的输出 输出结果保存在self.output_array
Tutorial/lesson-5/cnn.py:152
↓ 6 callersMethodforward
计算卷积层的输出 输出结果保存在self.output_array
Tutorial/lesson-4/cnn.py:152
↓ 6 callersMethodpredict
使用神经网络实现预测 sample: 输入样本
Tutorial/lesson-3/fc.py:74
↓ 5 callersMethodbackward
计算传递给前一层的误差项，以及计算每个权重的梯度 前一层的误差项保存在self.delta_array 梯度保存在Filter对象的weights_grad
Tutorial/lesson-6/cnn.py:168
↓ 5 callersMethodinit_delta
初始化误差项
Tutorial/lesson-6/lstm.py:141
↓ 5 callersMethodpredict
根据输入的样本预测输出值 sample: 数组，样本的特征，也就是网络的输入向量
Tutorial/lesson-3/bp.py:256
↓ 5 callersMethodpredict
输入向量，输出感知器的计算结果
Tutorial/lesson-1/perceptron.py:22
↓ 4 callersMethodbackward
(self, output)
Tutorial/lesson-3/fc.py:57
↓ 4 callersMethodcalc_gate
计算门
Tutorial/lesson-6/lstm.py:92
↓ 4 callersFunctionelement_wise_op
(array, op)
Tutorial/lesson-5/cnn.py:91
↓ 4 callersMethodinit_weight_gradient_mat
初始化权重矩阵
Tutorial/lesson-6/lstm.py:231
↓ 4 callersMethodinit_weight_mat
初始化权重矩阵
Tutorial/lesson-6/lstm.py:53
↓ 4 callersMethodload
加载数据文件，获得全部样本的输入向量
Tutorial/lesson-3/mnist.py:47
↓ 3 callersMethod__init__
(self, in_channels, out_channels, bilinear=True)
Pytorch-Seg/lesson-2/model/unet_parts.py:44
↓ 3 callersMethod__init__
(self, in_channels, out_channels, bilinear=True)
Pytorch-Seg/lesson-1/unet_parts.py:44
↓ 3 callersMethodbackward
BPTS反向传播算法
Tutorial/lesson-7/recursive.py:52
↓ 3 callersMethodbackward
计算传递给前一层的误差项，以及计算每个权重的梯度 前一层的误差项保存在self.delta_array 梯度保存在Filter对象的weights_grad
Tutorial/lesson-5/cnn.py:168
↓ 3 callersMethodbackward
计算传递给前一层的误差项，以及计算每个权重的梯度 前一层的误差项保存在self.delta_array 梯度保存在Filter对象的weights_grad
Tutorial/lesson-4/cnn.py:168
↓ 3 callersFunctionconv
计算卷积，自动适配输入为2D和3D的情况
Tutorial/lesson-6/cnn.py:39
↓ 3 callersFunctionconv
计算卷积，自动适配输入为2D和3D的情况
Tutorial/lesson-5/cnn.py:39
↓ 3 callersFunctionconv
计算卷积，自动适配输入为2D和3D的情况
Tutorial/lesson-4/cnn.py:39
↓ 3 callersMethoddump
(self, **kwArgs)
Tutorial/lesson-7/recursive.py:116
↓ 3 callersMethodget_next_batch
获得batch_size大小的数据集 Parameters: batch_size:batch_size大小 train_flag:是否从训练集获取数据 Returns: batch_x:大小为batch_size的数据x batch_y:大小为batch_si
Discuz/train.py:106
↓ 3 callersFunctionget_patch
从输入数组中获取本次卷积的区域， 自动适配输入为2D和3D的情况
Tutorial/lesson-6/cnn.py:7
↓ 3 callersFunctionget_patch
从输入数组中获取本次卷积的区域， 自动适配输入为2D和3D的情况
Tutorial/lesson-5/cnn.py:7
↓ 3 callersFunctionget_patch
从输入数组中获取本次卷积的区域， 自动适配输入为2D和3D的情况
Tutorial/lesson-4/cnn.py:7
↓ 3 callersFunctioninit_test
()
Tutorial/lesson-6/cnn.py:316
↓ 3 callersFunctioninit_test
()
Tutorial/lesson-5/cnn.py:316
↓ 3 callersFunctioninit_test
()
Tutorial/lesson-4/cnn.py:316
↓ 3 callersMethodloss
(self, output, label)
Tutorial/lesson-3/fc.py:116
↓ 3 callersMethodtrain
训练神经网络 labels: 数组，训练样本标签。每个元素是一个样本的标签。 data_set: 二维数组，训练样本特征。每个元素是一个样本的特征。
Tutorial/lesson-3/bp.py:205
↓ 2 callersFunctionaccuracy
计算topk的准确率
Pytorch-Seg/lesson-4/train.py:19
↓ 2 callersMethodaugment
(self, image, flipCode)
Pytorch-Seg/lesson-2/utils/dataset.py:14
↓ 2 callersMethodbackward
实现BPTT算法
Tutorial/lesson-5/rnn.py:34
↓ 2 callersMethodcalc_delta
内部函数，计算每个节点的delta
Tutorial/lesson-3/bp.py:221
↓ 2 callersMethodcalc_gradient
(self, label)
Tutorial/lesson-3/fc.py:100
↓ 2 callersMethodcalc_output
根据式1计算节点的输出
Tutorial/lesson-3/bp.py:42
↓ 2 callersMethodcalculate_output_size
(input_size, filter_size, zero_padding, stride)
Tutorial/lesson-6/cnn.py:266
↓ 2 callersMethodcalculate_output_size
(input_size, filter_size, zero_padding, stride)
Tutorial/lesson-5/cnn.py:266
↓ 2 callersMethodcalculate_output_size
(input_size, filter_size, zero_padding, stride)
Tutorial/lesson-4/cnn.py:266
↓ 2 callersMethodcrack_captcha_cnn
定义CNN Parameters: w_alpha:权重系数 b_alpha:偏置系数 Returns: out:CNN输出
Discuz/train.py:217
↓ 2 callersMethodcreate_delta_array
(self)
Tutorial/lesson-6/cnn.py:261
↓ 2 callersMethodcreate_delta_array
(self)
Tutorial/lesson-5/cnn.py:261
↓ 2 callersMethodcreate_delta_array
(self)
Tutorial/lesson-4/cnn.py:261
↓ 2 callersFunctiondata_set
()
Tutorial/lesson-6/lstm.py:275
↓ 2 callersFunctiondata_set
()
Tutorial/lesson-7/recursive.py:124
↓ 2 callersFunctiondata_set
()
Tutorial/lesson-5/rnn.py:96
↓ 2 callersFunctionelement_wise_op
(array, op)
Tutorial/lesson-6/cnn.py:91
↓ 2 callersFunctionelement_wise_op
(array, op)
Tutorial/lesson-4/cnn.py:91
↓ 2 callersMethodexpand_sensitivity_map
(self, sensitivity_array)
Tutorial/lesson-6/cnn.py:241
↓ 2 callersMethodexpand_sensitivity_map
(self, sensitivity_array)
Tutorial/lesson-5/cnn.py:241
↓ 2 callersMethodexpand_sensitivity_map
(self, sensitivity_array)
Tutorial/lesson-4/cnn.py:241
↓ 2 callersFunctionfeed_dict
Make a TensorFlow feed_dict: maps data onto Tensor placeholders.
mnist.py:132
↓ 2 callersMethodforward
(self, weighted_input)
Tutorial/lesson-3/fc.py:55
↓ 2 callersMethodget_file_content
读取文件内容
Tutorial/lesson-3/mnist.py:15
↓ 2 callersMethodget_gradient
获得网络在一个样本下，每个连接上的梯度 label: 样本标签 sample: 样本输入
Tutorial/lesson-3/bp.py:247
↓ 2 callersMethodget_imgs
获取图片，并划分训练集和测试集 Parameters: rate:测试集和训练集的比例，即测试集个数/训练集个数 Returns: test_imgs:测试集 test_labels:测试集标签 train_imgs:训练集 test_labels:训练集
Discuz/train.py:75
↓ 2 callersFunctionget_result
(vec)
Tutorial/lesson-3/mnist.py:97
↓ 2 callersMethodget_weights
(self)
Tutorial/lesson-6/cnn.py:110
↓ 2 callersMethodget_weights
(self)
Tutorial/lesson-5/cnn.py:110
↓ 2 callersMethodget_weights
(self)
Tutorial/lesson-4/cnn.py:110
↓ 2 callersFunctioninit_pool_test
()
Tutorial/lesson-6/cnn.py:412
↓ 2 callersFunctioninit_pool_test
()
Tutorial/lesson-5/cnn.py:412
↓ 2 callersFunctioninit_pool_test
()
Tutorial/lesson-4/cnn.py:412
↓ 2 callersMethodload
加载数据文件，获得全部样本的标签向量
Tutorial/lesson-3/mnist.py:60
↓ 2 callersFunctionnn_layer
(input_tensor, input_dim, output_dim, layer_name, act=tf.nn.relu)
mnist.py:65
↓ 2 callersMethodnorm
(self, number)
Tutorial/lesson-3/fc.py:161
↓ 2 callersFunctionpadding
为数组增加Zero padding，自动适配输入为2D和3D的情况
Tutorial/lesson-6/cnn.py:60
↓ 2 callersFunctionpadding
为数组增加Zero padding，自动适配输入为2D和3D的情况
Tutorial/lesson-5/cnn.py:60
↓ 2 callersFunctionpadding
为数组增加Zero padding，自动适配输入为2D和3D的情况
Tutorial/lesson-4/cnn.py:60
↓ 2 callersMethodreset_state
(self)
Tutorial/lesson-6/lstm.py:258
↓ 2 callersMethodreset_state
(self)
Tutorial/lesson-7/recursive.py:66
↓ 2 callersMethodreset_state
(self)
Tutorial/lesson-5/rnn.py:89
↓ 2 callersMethodset_output
设置节点的输出值。如果节点属于输入层会用到这个函数。
Tutorial/lesson-3/bp.py:24
↓ 2 callersMethodtext2vec
文本转向量 Parameters: text:文本 Returns: vector:向量
Discuz/train.py:162
↓ 2 callersMethodtrain
输入训练数据：一组向量、与每个向量对应的label；以及训练轮数、学习率
Tutorial/lesson-2/perceptron.py:32
↓ 2 callersFunctiontrain_data_set
()
Tutorial/lesson-3/fc.py:171
↓ 2 callersFunctiontranspose
(args)
Tutorial/lesson-3/fc.py:147
↓ 2 callersMethodupdate
(self, learning_rate)
Tutorial/lesson-6/cnn.py:116
↓ 2 callersMethodupdate
(self, learning_rate)
Tutorial/lesson-5/cnn.py:116
↓ 2 callersMethodupdate
(self, learning_rate)
Tutorial/lesson-4/cnn.py:116
↓ 2 callersFunctionvariable_summaries
(var)
mnist.py:48
↓ 2 callersMethodvec2text
向量转文本 Parameters: vec:向量 Returns: 文本
Discuz/train.py:191
↓ 1 callersMethod_one_iteration
一次迭代，把所有的训练数据过一遍
Tutorial/lesson-2/perceptron.py:38
↓ 1 callersMethod_one_iteration
一次迭代，把所有的训练数据过一遍
Tutorial/lesson-1/perceptron.py:38
↓ 1 callersMethod_update_weights
按照感知器规则更新权重
Tutorial/lesson-2/perceptron.py:51
↓ 1 callersMethod_update_weights
按照感知器规则更新权重
Tutorial/lesson-1/perceptron.py:51
↓ 1 callersMethodadd_connection
(self, connection)
Tutorial/lesson-3/bp.py:179
↓ 1 callersMethodappend_downstream_connection
添加一个到下游节点的连接
Tutorial/lesson-3/bp.py:30
↓ 1 callersMethodappend_upstream_connection
添加一个到上游节点的连接
Tutorial/lesson-3/bp.py:36
↓ 1 callersFunctionbias_variable
(shape)
mnist.py:43
↓ 1 callersMethodbp_gradient
(self, sensitivity_array)
Tutorial/lesson-6/cnn.py:227
next →1–100 of 315, ranked by callers