本篇文章的主要目的是對上一篇文章已經搭建好的網絡進一步豐富其內容,具體體現在6個方面:
1. 加載自己製作的訓練集
2. 數據增強
3. 斷點續訓
4. 參數提取
5. acc/loss可視化
6. 給圖識物
-
加載自己製作的訓練集
代碼:
import tensorflow as tf from PIL import Image import numpy as np import os #加載訓練集的路徑以及要保存的路徑 train_path = './mnist_image_label/mnist_train_jpg_60000/' train_txt = './mnist_image_label/mnist_train_jpg_60000.txt' x_train_savepath = './mnist_image_label/mnist_x_train.npy' y_train_savepath = './mnist_image_label/mnist_y_train.npy' #加載測試集的路徑以要保存的路徑 test_path = './mnist_image_label/mnist_test_jpg_10000/' test_txt = './mnist_image_label/mnist_test_jpg_10000.txt' x_test_savepath = './mnist_image_label/mnist_x_test.npy' y_test_savepath = './mnist_image_label/mnist_y_test.npy' #-------------------------------------------------------------------------- #---------------------------------自制數據集------------------------------- #-------------------------------------------------------------------------- def generateds(path,txt): f = open(txt,'r') # 以只讀形式打開文件 contents = f.readlines() # 讀取文件中所有行 f.close() # 關閉txt文件 x,y_ = [],[] # 建立空列表 for content in contents: # 逐行取出 value = content.split() # 以空格分開,圖片路徑爲value[0],標籤爲value[1],存入列表 img_path = path + value[0] # 拼出圖片路徑和文件名 img = Image.open(img_path) # 讀入圖片 img = np.array(img.convert('L')) # 圖片變爲8位寬灰度值的np.array格式 img = img / 255 # 數據歸一化(實現預處理) x.append(img) #歸一化後的數據,添加到列表x y_.append(value[1]) #標籤添加到列表y print('loading:' + content) #打印狀態提示 x = np.array(x) #變爲np.array格式 y_ = np.array(y_) #變爲np.array格式 y_ = y_.astype(np.int64) #變爲64位整型 return x,y_ #返回輸入特徵x,返回標籤y_ if os.path.exists(x_train_savepath) and os.path.exists(y_train_savepath) and os.path.exists(x_test_savepath) and os.path.exists(y_test_savepath): print('----------------Load Datasets---------------------') x_train_save = np.load(x_train_savepath) y_train = np.load(y_train_savepath) x_test_save = np.load(x_test_savepath) y_test = np.load(y_test_savepath) x_train = np.reshape(x_train_save,(len(x_train_save),28,28)) x_test = np.reshape(x_test_save,(len(x_test_save),28,28)) else: print('----------------Generate Datasets---------------------') x_train,y_train = generateds(train_path,train_txt) x_test,y_test = generateds(test_path,test_txt) print('----------------Save Datasets---------------------') x_train_save = np.reshape(x_train,(len(x_train), -1)) x_test_save = np.reshape(x_test,(len(x_test), -1)) np.save(x_train_savepath,x_train_save) np.save(y_train_savepath,y_train) np.save(x_test_savepath,x_test_save) np.save(y_test_savepath,y_test) model = tf.keras.models.Sequential([ tf.keras.layers.Flatten(), tf.keras.layers.Dense(128,activation = 'relu'), tf.keras.layers.Dense(10,activation = 'softmax') ]) model.compile(optimizer = 'adam', loss = tf.keras.losses.SparseCategoricalCrossentropy(from_logits = False), metrics = ['sparse_categorical_accuracy']) model.fit(x_train,y_train,batch_size = 32,epochs = 5,validation_data = (x_test,y_test),validation_freq=1) model.summary()
-
數據增強
函數介紹:Image_gen_train = tf.keras.preprocessing.image.ImageDataGenerator( Rescale = 所有數據將乘以該數值 Rotation_range = 隨機旋轉角度數範圍 Width_shift_range = 隨機寬度偏移量 Height_shift_range = 隨機高度偏移量 水平翻轉:horizontal_flip = 是否隨機水平翻轉 隨機縮放:zoom_range = 隨機縮放的範圍[1-n,1+n] ) Image_gen_train.fit(x_train)
代碼:
import tensorflow as tf from tensorflow.keras.preprocessing.image import ImageDataGenerator mnist = tf.keras.datasets.mnist (x_train,y_train),(x_test,y_test) = mnist.load_data() x_train,x_test = x_train / 255.0,x_test / 255.0 x_train = x_train.reshape(x_train.shape[0],28,28,1) #給數據增減一個維度,從(60000,28,28)reshape爲(60000,28,28,1) image_gen_train = ImageDataGenerator( rescale = 1. / 1., rotation_range = 45, width_shift_range = .15, height_shift_range = .15, horizontal_flip = False, zoom_range = 0.5 ) image_gen_train.fit(x_train) model = tf.keras.models.Sequential([ tf.keras.layers.Flatten(), tf.keras.layers.Dense(128,activation = 'relu'), tf.keras.layers.Dense(10,activation = 'softmax') ]) model.compile(optimizer = 'adam', loss = tf.keras.losses.SparseCategoricalCrossentropy(from_logits = False), metrics = ['sparse_categorical_accuracy']) model.fit(image_gen_train.flow(x_train,y_train,batch_size = 32),epochs = 5,validation_data = (x_test,y_test),validation_freq=1) model.summary()
-
斷點續訓
a. 讀取模型:函數:load_weights(路徑文件名) 例子: checkpoint_save_path = "./checkpoint/mnist.ckpt" if os.path.exists(checkpoint_save_path + '.index'): print("--------------load the model---------------") model.load_weights(checkpoint_save_path)
b. 保存模型:
函數: tf.keras.callbacks.ModelCheckpoint( filepath = 路徑文件名, save_weights_only = True/False, save_best_only = True/False ) history = mlodel.fit(callbacks = [cp_callback]) 例子: cp_callback = tf.keras.callbacks.ModelCheckpoint( filepath = checkpoint_save_path, save_weights_only = True, save_best_only = True) history = model.fit(x_train,y_train,batch_size=32,epochs=5,validation_data=(x_test,y_test),validation_freq=1,callbacks = [cp_callback])
代碼:
import tensorflow as tf import os mnist = tf.keras.datasets.mnist (x_train,y_train),(x_test,y_test) = mnist.load_data() x_train,x_test = x_train / 255.0,x_test / 255.0 model = tf.keras.models.Sequential([ tf.keras.layers.Flatten(), tf.keras.layers.Dense(128,activation = 'relu'), tf.keras.layers.Dense(10,activation = 'softmax') ]) model.compile(optimizer = 'adam', loss = tf.keras.losses.SparseCategoricalCrossentropy(from_logits = False), metrics = ['sparse_categorical_accuracy']) checkpoint_save_path = "./checkpoint/mnist.ckpt" if os.path.exists(checkpoint_save_path + '.index'): print("--------------load the model---------------") model.load_weights(checkpoint_save_path) cp_callback = tf.keras.callbacks.ModelCheckpoint( filepath = checkpoint_save_path, save_weights_only = True, save_best_only = True) history = model.fit(x_train,y_train,batch_size=32,epochs=5,validation_data=(x_test,y_test),validation_freq=1,callbacks = [cp_callback]) model.summary()
-
參數提取
代碼:import tensorflow as tf import os import numpy as np np.set_printoptions(threshold = np.inf) mnist = tf.keras.datasets.mnist (x_train,y_train),(x_test,y_test) = mnist.load_data() x_train,x_test = x_train / 255.0,x_test / 255.0 model = tf.keras.models.Sequential([ tf.keras.layers.Flatten(), tf.keras.layers.Dense(128,activation = 'relu'), tf.keras.layers.Dense(10,activation = 'softmax') ]) model.compile(optimizer = 'adam', loss = tf.keras.losses.SparseCategoricalCrossentropy(from_logits = False), metrics = ['sparse_categorical_accuracy']) checkpoint_save_path = "./checkpoint/mnist.ckpt" if os.path.exists(checkpoint_save_path + '.index'): print("--------------load the model---------------") model.load_weights(checkpoint_save_path) cp_callback = tf.keras.callbacks.ModelCheckpoint( filepath = checkpoint_save_path, save_weights_only = True, save_best_only = True) history = model.fit(x_train,y_train,batch_size=32,epochs=5,validation_data=(x_test,y_test),validation_freq=1,callbacks = [cp_callback]) model.summary() #打印參數 print(model.trainable_variables) #將參數保存到文件weights.txt中 file = open('./weights.txt','w') for v in model.trainable_variables: file.write(str(v.name) + '\n') file.write(str(v.shape) + '\n') file.write(str(v.numpy()) + '\n') file.close()
-
acc/loss可視化
history包含得出的結果:- 訓練集loss:loss
- 測試集loss:val_loss
- 訓練集準確率:sparse_categorical_accuracy
- 測試集準確率:val_sparse_cateforical_accuracy
代碼:
import tensorflow as tf import os import numpy as np from matplotlib import pyplot as plt np.set_printoptions(threshold = np.inf) mnist = tf.keras.datasets.mnist (x_train,y_train),(x_test,y_test) = mnist.load_data() x_train,x_test = x_train / 255.0,x_test / 255.0 model = tf.keras.models.Sequential([ tf.keras.layers.Flatten(), tf.keras.layers.Dense(128,activation = 'relu'), tf.keras.layers.Dense(10,activation = 'softmax') ]) model.compile(optimizer = 'adam', loss = tf.keras.losses.SparseCategoricalCrossentropy(from_logits = False), metrics = ['sparse_categorical_accuracy']) checkpoint_save_path = "./checkpoint/mnist.ckpt" if os.path.exists(checkpoint_save_path + '.index'): print("--------------load the model---------------") model.load_weights(checkpoint_save_path) cp_callback = tf.keras.callbacks.ModelCheckpoint( filepath = checkpoint_save_path, save_weights_only = True, save_best_only = True) history = model.fit(x_train,y_train,batch_size=32,epochs=5,validation_data=(x_test,y_test),validation_freq=1,callbacks = [cp_callback]) model.summary() print(model.trainable_variables) file = open('./weights.txt','w') for v in model.trainable_variables: file.write(str(v.name) + '\n') file.write(str(v.shape) + '\n') file.write(str(v.numpy()) + '\n') file.close() acc = history.history['sparse_categorical_accuracy'] val_acc = history.history['val_sparse_categorical_accuracy'] loss = history.history['loss'] val_loss = history.history['val_loss'] plt.subplot(1,2,1) plt.plot(acc,label="Training Accuracy") plt.plot(val_acc,label="Validation Accuracy") plt.title("Training abd Validation Accuracy") plt.legend() plt.subplot(1,2,2) plt.plot(loss,label="Training Loss") plt.plot(val_loss,label="Validation Loss") plt.title("Training abd Validation Loss") plt.legend() plt.show()
-
給圖識物(也就是前向傳播的應用)
-
調用函數:predict(輸入特徵,batch_size=整數) 返回前向傳播的計算結果
-
步驟:
a. 浮現模型 model = tf.keras.models.Sequential([ tf.keras.layers.Flatten(), tf.keras.layers.Dense(128, activation='relu'), tf.keras.layers.Dense(10, activation='softmax')]) b. 加載參數 model.load_weights(model_save_path) c. 預測結果 result = model.predict(x_predict)
代碼:
from PIL import Image import numpy as np import tensorflow as tf model_save_path = './checkpoint/mnist.ckpt' model = tf.keras.models.Sequential([ tf.keras.layers.Flatten(), tf.keras.layers.Dense(128, activation='relu'), tf.keras.layers.Dense(10, activation='softmax')]) model.load_weights(model_save_path) preNum = int(input("input the number of test pictures:")) for i in range(preNum): image_path = input("the path of test picture:") img = Image.open(image_path) img = img.resize((28, 28), Image.ANTIALIAS) img_arr = np.array(img.convert('L')) for i in range(28): for j in range(28): if img_arr[i][j] < 200: img_arr[i][j] = 255 else: img_arr[i][j] = 0 img_arr = img_arr / 255.0 x_predict = img_arr[tf.newaxis, ...] result = model.predict(x_predict) pred = tf.argmax(result, axis=1) print('\n') tf.print(pred)
-
下面的是筆者的微信公衆號,歡迎關注,會持續更新c++、python、tensorflow、機器學習、深度學習等系列文章