DL之LiRDNNCNN：利用LiR、DNN、CNN算法对MNIST手写数字图片(csv)识别数据集实现(10)分类预测

DL之LiR&DNN&CNN：利用LiR、DNN、CNN算法对MNIST手写数字图片(csv)识别数据集实现(10)分类预测

输出结果

设计思路

核心代码

输出结果

数据集：Dataset之MNIST：MNIST(手写数字图片识别+csv文件)数据集简介、下载、使用方法之详细攻略

设计思路

核心代码

classifier = skflow.TensorFlowLinearClassifier(n_classes=10, learning_rate=0.01)
classifier.fit(X_train, y_train)
linear_y_predict = classifier.predict(X_test)classifier = skflow.TensorFlowDNNClassifier(hidden_units=[200, 50, 10], n_classes = 10,learning_rate=0.01)
classifier.fit(X_train, y_train)
dnn_y_predict = classifier.predict(X_test)classifier = skflow.TensorFlowEstimator(model_fn=conv_model, n_classes=10, steps=20000,learning_rate=0.001)
classifier.fit(X_train, y_train)
classifier.predict(X_test)

class TensorFlowDNNClassifier(TensorFlowEstimator, ClassifierMixin):"""TensorFlow DNN Classifier model.Parameters:hidden_units: List of hidden units per layer.n_classes: Number of classes in the target.tf_master: TensorFlow master. Empty string is default for local.batch_size: Mini batch size.steps: Number of steps to run over data.optimizer: Optimizer name (or class), for example "SGD", "Adam","Adagrad".learning_rate: If this is constant float value, no decay function is used.Instead, a customized decay function can be passed that acceptsglobal_step as parameter and returns a Tensor.e.g. exponential decay function:def exp_decay(global_step):return tf.train.exponential_decay(learning_rate=0.1, global_step,decay_steps=2, decay_rate=0.001)class_weight: None or list of n_classes floats. Weight associated withclasses for loss computation. If not given, all classes are suppose to haveweight one.tf_random_seed: Random seed for TensorFlow initializers.Setting this value, allows consistency between reruns.continue_training: when continue_training is True, once initializedmodel will be continuely trained on every call of fit.num_cores: Number of cores to be used. (default: 4)early_stopping_rounds: Activates early stopping if this is not None.Loss needs to decrease at least every every <early_stopping_rounds>round(s) to continue training. (default: None)max_to_keep: The maximum number of recent checkpoint files to keep.As new files are created, older files are deleted.If None or 0, all checkpoint files are kept.Defaults to 5 (that is, the 5 most recent checkpoint files are kept.)keep_checkpoint_every_n_hours: Number of hours between each checkpointto be saved. The default value of 10,000 hours effectively disables the feature."""def __init__(self, hidden_units, n_classes, tf_master="", batch_size=32, steps=200, optimizer="SGD", learning_rate=0.1, class_weight=None, tf_random_seed=42, continue_training=False, num_cores=4, verbose=1, early_stopping_rounds=None, max_to_keep=5, keep_checkpoint_every_n_hours=10000):self.hidden_units = hidden_unitssuper(TensorFlowDNNClassifier, self).__init__(model_fn=self._model_fn, n_classes=n_classes, tf_master=tf_master, batch_size=batch_size, steps=steps, optimizer=optimizer, learning_rate=learning_rate, class_weight=class_weight, tf_random_seed=tf_random_seed, continue_training=continue_training, num_cores=4, verbose=verbose, early_stopping_rounds=early_stopping_rounds, max_to_keep=max_to_keep, keep_checkpoint_every_n_hours=keep_checkpoint_every_n_hours)def _model_fn(self, X, y):return models.get_dnn_model(self.hidden_units, models.logistic_regression)(X, y)@propertydef weights_(self):"""Returns weights of the DNN weight layers."""weights = []for layer in range(len(self.hidden_units)):weights.append(self.get_tensor_value('dnn/layer%d/Linear/Matrix:0' % layer))weights.append(self.get_tensor_value('logistic_regression/weights:0'))return weights@propertydef bias_(self):"""Returns bias of the DNN's bias layers."""biases = []for layer in range(len(self.hidden_units)):biases.append(self.get_tensor_value('dnn/layer%d/Linear/Bias:0' % layer))biases.append(self.get_tensor_value('logistic_regression/bias:0'))return biases

DL之LiRDNNCNN：利用LiR、DNN、CNN算法对MNIST手写数字图片(csv)识别数据集实现(10)分类预测相关推荐

DL之DNN：利用DNN【784→50→100→10】算法对MNIST手写数字图片识别数据集进行预测、模型优化
DL之DNN:利用DNN[784→50→100→10]算法对MNIST手写数字图片识别数据集进行预测.模型优化导读目的是建立三层神经网络,进一步理解DNN内部的运作机制目录输出结果设计思路 ...
DL之DNN：利用DNN算法对mnist手写数字图片识别数据集(sklearn自带,1797*64)训练、预测(95%)
DL之DNN:利用DNN算法对mnist手写数字图片识别数据集(sklearn自带,1797*64)训练.预测(95%) 目录数据集展示输出结果设计代码数据集展示先查看sklearn自带di ...
TF：利用是Softmax回归+GD算法实现MNIST手写数字图片识别(10000张图片测试得到的准确率为92%)
TF:利用是Softmax回归+GD算法实现MNIST手写数字图片识别(10000张图片测试得到的准确率为92%) 目录设计思路全部代码设计思路全部代码 #TF:利用是Softmax回归+GD ...
TF之LSTM：利用LSTM算法对mnist手写数字图片数据集(TF函数自带)训练、评估(偶尔100%准确度，交叉熵验证)
TF之LSTM:利用LSTM算法对mnist手写数字图片数据集(TF函数自带)训练.评估(偶尔100%准确度,交叉熵验证) 目录输出结果设计思路代码设计输出结果第 0 accuracy 0. ...
DL之CNN：利用卷积神经网络算法(2→2,基于Keras的API-Functional)利用MNIST(手写数字图片识别)数据集实现多分类预测
DL之CNN:利用卷积神经网络算法(2→2,基于Keras的API-Functional)利用MNIST(手写数字图片识别)数据集实现多分类预测目录输出结果设计思路核心代码输出结果下边两张 ...
DL之CNN：利用卷积神经网络算法(2→2,基于Keras的API-Sequential)利用MNIST(手写数字图片识别)数据集实现多分类预测
DL之CNN:利用卷积神经网络算法(2→2,基于Keras的API-Sequential)利用MNIST(手写数字图片识别)数据集实现多分类预测目录输出结果设计思路核心代码输出结果 1.10 ...
TF之NN：利用DNN算法(SGD+softmax+cross_entropy)对mnist手写数字图片识别训练集(TF自带函数下载)实现87.4%识别
TF之NN:利用DNN算法(SGD+softmax+cross_entropy)对mnist手写数字图片识别训练集(TF自带函数下载)实现87.4%识别目录输出结果代码设计输出结果代码设计 ...
TF之LSTM：利用多层LSTM算法对MNIST手写数字识别数据集进行多分类
TF之LSTM:利用多层LSTM算法对MNIST手写数字识别数据集进行多分类目录设计思路实现代码设计思路更新-- 实现代码 # -*- coding:utf-8 -*- import ten ...
TF之DNN：利用DNN【784→500→10】对MNIST手写数字图片识别数据集(TF自带函数下载)预测(98%)+案例理解DNN过程
TF之DNN:利用DNN[784→500→10]对MNIST手写数字图片识别数据集(TF自带函数下载)预测(98%)+案例理解DNN过程目录输出结果案例理解DNN过程思路代码设计输出结果案 ...

DL之LiRDNNCNN：利用LiR、DNN、CNN算法对MNIST手写数字图片(csv)识别数据集实现(10)分类预测

输出结果

设计思路

核心代码

DL之LiRDNNCNN：利用LiR、DNN、CNN算法对MNIST手写数字图片(csv)识别数据集实现(10)分类预测相关推荐

最新文章

热门文章