1.Fashion-MNIST數據集

Fashion-MNIST數據集包括一個包含60,000個示例的訓練集和一個包含10,000個示例的測試集。每個示例是一個28x28灰度圖像，與來自以下10個類的標簽相關聯:

• T恤/上衣
• 褲子
• 套衫
• 連衣裙
• 外套
• 涼鞋
• 襯衫
• 運動鞋
• 包
• 短靴
  
  Source: https://github.com/zalandoresearch/fashion-mnist/

由于給的樣例是.ipynb格式的,因為我習慣使用pycharm，因此我將程序改到.py格式，但是由于內容較多，只能一點一點來研究相關程序。

2.cnn_preprocess.py

這個程序預處理數據，給神經網絡提供訓練數據。

---

（1）探索數據

import helper
import numpy as np
# 模塊 pickle 實現了對一個 Python 對象結構的二進制序列化和反序列化。
# 這里用來讀取fashion-mnist.p程序
import picklefilename = "fashion-mnist.p"sample_id = 17
helper.display_stats(filename, sample_id)

這是helper.py中的程序：

def _load_label_names():"""從文件中加載標簽名稱"""return ['t-shirt', 'trouser', 'pullover', 'dress', 'coat', 'sandal', 'shirt', 'sneaker', 'bag', 'ankle_boot']def load_dataset(dataset_folder_path):"""加載訓練和測試數據集"""with open(dataset_folder_path, mode='rb') as file:pickle_data = pickle.load(file)# 整理訓練和測試數據train_features = pickle_data[0].reshape((len(pickle_data[0]), 1, 28, 28)).transpose(0, 2, 3, 1)train_labels = pickle_data[1]test_features = pickle_data[2].reshape((len(pickle_data[2]), 1, 28, 28)).transpose(0, 2, 3, 1)test_labels = pickle_data[3]   return train_features, train_labels, test_features, test_labelsdef display_stats(dataset_folder_path, sample_id):"""顯示數據集的狀態"""train_features, train_labels, test_features, test_labels = load_dataset(dataset_folder_path)if not (0 <= sample_id < len(train_features)):print('{} samples in training set.  {} is out of range.'.format(len(train_features), sample_id))return Noneprint('Samples: {}'.format(len(train_features)))# 用于統計相關標簽數據print('Label Counts: {}'.format(dict(zip(*np.unique(train_labels, return_counts=True)))))print('First 20 Labels: {}'.format(train_labels[:20]))sample_image = train_features[sample_id]sample_label = train_labels[sample_id]label_names = _load_label_names()print('\nExample of Image {}:'.format(sample_id))print('Image - Min Value: {} Max Value: {}'.format(sample_image.min(), sample_image.max()))print('Image - Shape: {}'.format(sample_image.shape))print('Label - Label Id: {} Name: {}'.format(sample_label, label_names[sample_label]))plt.axis('off')plt.imshow(sample_image.squeeze(), cmap="gray")plt.show()

輸出如下：

Samples: 60000
Label Counts: {0: 6000, 1: 6000, 2: 6000, 3: 6000, 4: 6000, 5: 6000, 6: 6000, 7: 6000, 8: 6000, 9: 6000}
First 20 Labels: [9 0 0 3 0 2 7 2 5 5 0 9 5 5 7 9 1 0 6 4]Example of Image 17:
Image - Min Value: 0 Max Value: 254
Image - Shape: (28, 28, 1)
Label - Label Id: 0 Name: t-shirt

---

（2）數據預處理

實現normalize函數以獲取圖像數據x，并將其作為歸一化的Numpy數組返回。值應在0到1范圍內。返回對象的形狀應與x相同。

import problem_unittests as tests
def normalize(x):"""將0到1范圍內的樣本圖像數據列表進行規范化"""# TODO: Implement Functionoutput = np.array([image/255 for image in x])return output

（3）one-hot碼

實現該one_hot_encode功能。輸入，x是標簽列表。實現該功能以將標簽列表作為One-Hot編碼的Numpy數組返回。標簽的可能值為0到9。“one-hot code”功能應在每次調用one_hot_encode之間為每個值返回相同的編碼。

def one_hot_encode(x):"""一個獨熱碼的樣本標簽列表。為每個標簽返回一個獨熱碼向量。"""# TODO: Implement Functionone_hot = np.eye(10)[x]return None

（4）預處理和保存

預處理所有數據，并將其保存，是由10%的訓練數據進行驗證。

def _preprocess_and_save(normalize, one_hot_encode, features, labels, filename):"""對數據進行預處理并保存到文件中"""features = normalize(features)labels = one_hot_encode(labels)pickle.dump((features, labels), open(filename, 'wb'))def preprocess_and_save_data(dataset_folder_path, normalize, one_hot_encode):"""預處理訓練和驗證數據"""valid_features = []valid_labels = []train_features, train_labels, test_features, test_labels = load_dataset(dataset_folder_path)validation_count = int(len(train_features) * 0.1)# 對新的培訓數據進行預處理和保存_preprocess_and_save(normalize,one_hot_encode,train_features[:-validation_count],train_labels[:-validation_count],'preprocess_train' + '.p')# 使用部分訓練batch進行驗證valid_features.extend(train_features[-validation_count:])valid_labels.extend(train_labels[-validation_count:])# 對所有驗證數據進行預處理和保存_preprocess_and_save(normalize,one_hot_encode,np.array(valid_features),np.array(valid_labels),'preprocess_validation.p')# 對所有測試數據進行預處理和保存_preprocess_and_save(normalize,one_hot_encode,np.array(test_features),np.array(test_labels),'preprocess_test.p')

---

完整cnn_preprocess.py程序

import helper
import numpy as np
import problem_unittests as tests
import picklefilename = "fashion-mnist.p"def normalize(x):"""將0到1范圍內的樣本圖像數據列表進行歸一化"""output = np.array([image / 255 for image in x])return outputdef one_hot_encode(x):"""一個獨熱碼的樣本標簽列表。為每個標簽返回一個獨熱碼向量。"""one_hot = np.eye(10)[x]return one_hot# 對所有數據進行預處理
helper.preprocess_and_save_data(filename, normalize, one_hot_encode)

3.cnn_train.py

import helper
import numpy as np
import problem_unittests as tests
import pickle
import tensorflow as tfdef neural_net_image_input(image_shape):"""返回一批圖像輸入的張量"""input_image = tf.placeholder(tf.float32, shape=(None, *image_shape), name="x")return input_imagedef neural_net_label_input(n_classes):"""返回一個batch標簽輸入的張量"""input_label = tf.placeholder(tf.int32, shape=(None, n_classes), name="y")return input_labeldef neural_net_keep_prob_input():"""返回一個保持概率的張量"""keep_prob = tf.placeholder(tf.float32, name="keep_prob")return keep_probdef conv2d_maxpool(x_tensor, conv_num_outputs, conv_ksize, conv_strides, pool_ksize, pool_strides):"""對x_張量應用卷積和最大池化"""filter_size = [conv_ksize[0], conv_ksize[1], x_tensor.get_shape().as_list()[3], conv_num_outputs]weight = tf.Variable(tf.truncated_normal(filter_size, stddev=0.01))conv = tf.nn.conv2d(x_tensor, weight, [1, conv_strides[0], conv_strides[1], 1], padding="SAME")bias = tf.Variable(tf.zeros([conv_num_outputs]))conv = tf.nn.bias_add(conv, bias)conv = tf.nn.relu(conv)conv = tf.nn.max_pool(conv, [1, pool_ksize[0], pool_ksize[1], 1], [1, pool_strides[0], pool_strides[1], 1],padding="SAME")return convdef flatten(x_tensor):"""將x_張量壓平到(批處理大小，壓平圖像大小)，以方便接下來構建全連接層"""conv_flatten = tf.contrib.layers.flatten(x_tensor)return conv_flattendef fully_conn(x_tensor, num_outputs):"""：使用權值和偏置將一個全連接層應用到x_tensor上"""fc_layer = tf.layers.dense(x_tensor, num_outputs, activation=tf.nn.relu)return fc_layerdef output(x_tensor, num_outputs):"""使用權值和偏置應用一個輸出層"""output_layer = tf.layers.dense(x_tensor, num_outputs, activation=None)return output_layerdef conv_net(x, keep_prob):"""創建一個卷積神經網絡模型: x: 持有圖像數據的占位張量。: keep_prob: 持有dropout的占位張量保持概率。: return:表示logits的張量"""conv_layer1 = conv2d_maxpool(x, conv_num_outputs=64, conv_ksize=(5, 5), conv_strides=(2, 2),pool_ksize=(2, 2), pool_strides=(2, 2))conv_layer1 = tf.nn.dropout(conv_layer1, keep_prob)conv_layer2 = conv2d_maxpool(conv_layer1, conv_num_outputs=128, conv_ksize=(3, 3), conv_strides=(2, 2),pool_ksize=(2, 2), pool_strides=(2, 2))conv_layer2 = tf.nn.dropout(conv_layer2, keep_prob)flat_layer = flatten(conv_layer2)fc_layer1 = fully_conn(flat_layer, 256)fc_layer2 = fully_conn(fc_layer1, 128)fc_layer3 = fully_conn(fc_layer2, 64)output_layer = output(fc_layer3, 10)return output_layerdef train_neural_network(session, optimizer, keep_probability, feature_batch, label_batch):"""對批處理圖像和標簽的會話進行優化: session: 當前TensorFlow會話: optimizer: TensorFlow優化函數: keep_probability: 保持概率: feature_batch: 批量Numpy圖像數據: label_batch: 批量Numpy標簽數據"""session.run(optimizer, feed_dict={x: feature_batch, y: label_batch, keep_prob: keep_probability})def print_stats(session, feature_batch, label_batch, cost, accuracy):"""打印關于損失和驗證準確性的信息: session: 當前TensorFlow會話: feature_batch: 批量Numpy圖像數據: label_batch: 批量Numpy標簽數據: cost: TensorFlow成本函數: accuracy: TensorFlow精度函數"""l = session.run(cost, feed_dict={x: feature_batch, y: label_batch, keep_prob: 1.0})validation_accuracy = session.run(accuracy, feed_dict={x: valid_features, y: valid_labels, keep_prob: 1.0})print("The loss is: {0}, and the Validation Accuracy is: {1}".format(l, validation_accuracy))valid_features, valid_labels = pickle.load(open('preprocess_validation.p', mode='rb'))# 可調整參數
epochs = 5
batch_size = 64
keep_probability = 0.5# gpu參數
gpu_options = tf.GPUOptions(per_process_gpu_memory_fraction=0.333)# 刪除以前的 weights, bias, inputs, etc..
tf.reset_default_graph()# Inputs
x = neural_net_image_input((28, 28, 1))
y = neural_net_label_input(10)
keep_prob = neural_net_keep_prob_input()# Model
logits = conv_net(x, keep_prob)# 命名為logits張量，這樣就可以在訓練后從磁盤加載
logits = tf.identity(logits, name='logits')# 損失和優化器
cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(logits=logits, labels=y))
optimizer = tf.train.AdamOptimizer().minimize(cost)# 正確率
correct_pred = tf.equal(tf.argmax(logits, 1), tf.argmax(y, 1))
accuracy = tf.reduce_mean(tf.cast(correct_pred, tf.float32), name='accuracy')# 模型保存路徑
save_model_path = './image_classification'with tf.Session(config=tf.ConfigProto(gpu_options=gpu_options)) as sess:# 初始化變量sess.run(tf.global_variables_initializer())# 訓練周期for epoch in range(epochs):for batch_features, batch_labels in helper.load_preprocess_training_batch(batch_size):train_neural_network(sess, optimizer, keep_probability, batch_features, batch_labels)print('Epoch {:>2}:  '.format(epoch + 1), end='')print_stats(sess, batch_features, batch_labels, cost, accuracy)# 保存模型saver = tf.train.Saver()save_path = saver.save(sess, save_model_path)

其輸出為：

Epoch  1:  The loss is: 0.5249987244606018, and the Validation Accuracy is: 0.8186666369438171
Epoch  2:  The loss is: 0.4325884282588959, and the Validation Accuracy is: 0.8506666421890259
Epoch  3:  The loss is: 0.38672173023223877, and the Validation Accuracy is: 0.8629999756813049
Epoch  4:  The loss is: 0.34116020798683167, and the Validation Accuracy is: 0.8736666440963745
Epoch  5:  The loss is: 0.311458021402359, and the Validation Accuracy is: 0.8741666674613953

之后我們就可以加載訓練好的模型來測試其效果了。

4.cnn_test.py

我們使用cnn_test.py函數來確定最終的正確率，并畫出判斷結果。

import tensorflow as tf
import pickle
import helper
import random
import matplotlib.pyplot as plt
# 可調整參數
epochs = 5
batch_size = 64
keep_probability = 0.5
save_model_path = './image_classification'
n_samples = 4
top_n_predictions = 3def test_model():"""根據測試數據集測試保存的模型"""test_features, test_labels = pickle.load(open('preprocess_test.p', mode='rb'))loaded_graph = tf.Graph()config = tf.ConfigProto(device_count={'GPU': 0})with tf.Session(config=config, graph=loaded_graph) as sess:# 加載模型loader = tf.train.import_meta_graph(save_model_path + '.meta')loader.restore(sess, save_model_path)# 從加載的模型中獲取張量loaded_x = loaded_graph.get_tensor_by_name('x:0')loaded_y = loaded_graph.get_tensor_by_name('y:0')loaded_keep_prob = loaded_graph.get_tensor_by_name('keep_prob:0')loaded_logits = loaded_graph.get_tensor_by_name('logits:0')loaded_acc = loaded_graph.get_tensor_by_name('accuracy:0')# 批量獲取內存限制的準確性test_batch_acc_total = 0test_batch_count = 0for test_feature_batch, test_label_batch in helper.batch_features_labels(test_features, test_labels,batch_size):test_batch_acc_total += sess.run(loaded_acc,feed_dict={loaded_x: test_feature_batch, loaded_y: test_label_batch, loaded_keep_prob: 1.0})test_batch_count += 1print('Testing Accuracy: {}\n'.format(test_batch_acc_total / test_batch_count))# 打印隨機抽樣random_test_features, random_test_labels = tuple(zip(*random.sample(list(zip(test_features, test_labels)), n_samples)))random_test_predictions = sess.run(tf.nn.top_k(tf.nn.softmax(loaded_logits), top_n_predictions),feed_dict={loaded_x: random_test_features, loaded_y: random_test_labels, loaded_keep_prob: 1.0})helper.display_image_predictions(random_test_features, random_test_labels, random_test_predictions)test_model()
plt.show()

輸出結果如下:

Testing Accuracy: 0.8765923566878981

5.helper.py

因為使用pycharm的關系，我對helper.py進行了少許更改。

import pickle
import numpy as np
import matplotlib.pyplot as plt
from sklearn.preprocessing import LabelBinarizerdef _load_label_names():"""從文件中加載標簽名稱"""return ['t-shirt', 'trouser', 'pullover', 'dress', 'coat', 'sandal', 'shirt', 'sneaker', 'bag', 'ankle_boot']def load_dataset(dataset_folder_path):"""加載訓練和測試數據集"""with open(dataset_folder_path, mode='rb') as file:pickle_data = pickle.load(file)train_features = pickle_data[0].reshape((len(pickle_data[0]), 1, 28, 28)).transpose(0, 2, 3, 1)print(train_features.shape)train_labels = pickle_data[1]test_features = pickle_data[2].reshape((len(pickle_data[2]), 1, 28, 28)).transpose(0, 2, 3, 1)test_labels = pickle_data[3]   return train_features, train_labels, test_features, test_labelsdef display_stats(dataset_folder_path, sample_id):"""顯示數據集的狀態"""train_features, train_labels, test_features, test_labels = load_dataset(dataset_folder_path)if not (0 <= sample_id < len(train_features)):print('{} samples in training set.  {} is out of range.'.format(len(train_features), sample_id))return Noneprint('Samples: {}'.format(len(train_features)))print('Label Counts: {}'.format(dict(zip(*np.unique(train_labels, return_counts=True)))))print('First 20 Labels: {}'.format(train_labels[:20]))sample_image = train_features[sample_id]sample_label = train_labels[sample_id]label_names = _load_label_names()print('\nExample of Image {}:'.format(sample_id))print('Image - Min Value: {} Max Value: {}'.format(sample_image.min(), sample_image.max()))print('Image - Shape: {}'.format(sample_image.shape))print('Label - Label Id: {} Name: {}'.format(sample_label, label_names[sample_label]))plt.axis('off')plt.imshow(sample_image.squeeze(), cmap="gray")plt.show()def _preprocess_and_save(normalize, one_hot_encode, features, labels, filename):"""對數據進行預處理并保存到文件中"""features = normalize(features)labels = one_hot_encode(labels)pickle.dump((features, labels), open(filename, 'wb'))def preprocess_and_save_data(dataset_folder_path, normalize, one_hot_encode):"""預處理訓練和驗證數據"""valid_features = []valid_labels = []train_features, train_labels, test_features, test_labels = load_dataset(dataset_folder_path)validation_count = int(len(train_features) * 0.1)# 對新的培訓數據進行預處理和保存_preprocess_and_save(normalize,one_hot_encode,train_features[:-validation_count],train_labels[:-validation_count],'preprocess_train' + '.p')# 使用部分訓練batch進行驗證valid_features.extend(train_features[-validation_count:])valid_labels.extend(train_labels[-validation_count:])# 對所有驗證數據進行預處理和保存_preprocess_and_save(normalize,one_hot_encode,np.array(valid_features),np.array(valid_labels),'preprocess_validation.p')# 對所有測試數據進行預處理和保存_preprocess_and_save(normalize,one_hot_encode,np.array(test_features),np.array(test_labels),'preprocess_test.p')def batch_features_labels(features, labels, batch_size):"""將特征和標簽分成批次"""assert features is not None, 'features is None!'assert labels is not None, 'labels is None!'for start in range(0, len(features), batch_size):end = min(start + batch_size, len(features))yield features[start:end], labels[start:end]def load_preprocess_training_batch(batch_size):"""載預處理的訓練數據，并以<batch_size>或更小的批量返回"""filename = 'preprocess_train' + '.p'features, labels = pickle.load(open(filename, mode='rb'))# 返回批量<batch_size>或更小的訓練數據return batch_features_labels(features, labels, batch_size)def display_image_predictions(features, labels, predictions):n_classes = 10label_names = _load_label_names()label_binarizer = LabelBinarizer()label_binarizer.fit(range(n_classes))label_ids = label_binarizer.inverse_transform(np.array(labels))fig, axies = plt.subplots(nrows=4, ncols=2)fig.tight_layout()fig.suptitle('Softmax Predictions', fontsize=20, y=1.1)n_predictions = 3margin = 0.05ind = np.arange(n_predictions)width = (1. - 2. * margin) / n_predictionsfor image_i, (feature, label_id, pred_indicies, pred_values) in enumerate(zip(features, label_ids, predictions.indices, predictions.values)):pred_names = [label_names[pred_i] for pred_i in pred_indicies]correct_name = label_names[label_id]axies[image_i][0].imshow(feature.squeeze(), cmap='gray')axies[image_i][0].set_title(correct_name)axies[image_i][0].set_axis_off()axies[image_i][1].barh(ind + margin, pred_values[::-1], width)axies[image_i][1].set_yticks(ind + margin)axies[image_i][1].set_yticklabels(pred_names[::-1])axies[image_i][1].set_xticks([0, 0.5, 1.0])

以下是所有本次程序使用或者建立的數據，其中problem_unittests.py是一個關于某些模塊的測試程序，在主要程序中使用不到，只做編寫程序時測試之用。
接下來便是完全卷積網絡了。

總結

以上是生活随笔為你收集整理的Udacity机器人软件工程师课程笔记（二十八) - 卷积神经网络实例 - Fashion-MNIST数据集的全部內容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網站內容還不錯，歡迎將生活随笔推薦給好友。