diff --git a/.idea/aws.xml b/.idea/aws.xml new file mode 100644 index 0000000..b63b642 --- /dev/null +++ b/.idea/aws.xml @@ -0,0 +1,11 @@ + + + + + + \ No newline at end of file diff --git a/Number-Of-Pool/Four/logs/20230326-182517/train/events.out.tfevents.1679826320.WEIYUQING.20952.0.v2 b/Number-Of-Pool/Four/logs/20230326-182517/train/events.out.tfevents.1679826320.WEIYUQING.20952.0.v2 new file mode 100644 index 0000000..c8917a7 Binary files /dev/null and b/Number-Of-Pool/Four/logs/20230326-182517/train/events.out.tfevents.1679826320.WEIYUQING.20952.0.v2 differ diff --git a/Number-Of-Pool/Four/logs/20230326-182517/validation/events.out.tfevents.1679826325.WEIYUQING.20952.1.v2 b/Number-Of-Pool/Four/logs/20230326-182517/validation/events.out.tfevents.1679826325.WEIYUQING.20952.1.v2 new file mode 100644 index 0000000..d148674 Binary files /dev/null and b/Number-Of-Pool/Four/logs/20230326-182517/validation/events.out.tfevents.1679826325.WEIYUQING.20952.1.v2 differ diff --git a/Number-Of-Pool/Four/model/ecg_model_four_pool.h5 b/Number-Of-Pool/Four/model/ecg_model_four_pool.h5 new file mode 100644 index 0000000..37fb7fa Binary files /dev/null and b/Number-Of-Pool/Four/model/ecg_model_four_pool.h5 differ diff --git a/Number-Of-Pool/One/logs/20230326-180117/train/events.out.tfevents.1679824881.WEIYUQING.22764.0.v2 b/Number-Of-Pool/One/logs/20230326-180117/train/events.out.tfevents.1679824881.WEIYUQING.22764.0.v2 new file mode 100644 index 0000000..abd074f Binary files /dev/null and b/Number-Of-Pool/One/logs/20230326-180117/train/events.out.tfevents.1679824881.WEIYUQING.22764.0.v2 differ diff --git a/Number-Of-Pool/One/logs/20230326-180117/validation/events.out.tfevents.1679824886.WEIYUQING.22764.1.v2 b/Number-Of-Pool/One/logs/20230326-180117/validation/events.out.tfevents.1679824886.WEIYUQING.22764.1.v2 new file mode 100644 index 0000000..9a1e000 Binary files /dev/null and b/Number-Of-Pool/One/logs/20230326-180117/validation/events.out.tfevents.1679824886.WEIYUQING.22764.1.v2 differ diff --git a/Number-Of-Pool/One/model/ecg_model_one_pool.h5 b/Number-Of-Pool/One/model/ecg_model_one_pool.h5 new file mode 100644 index 0000000..447d538 Binary files /dev/null and b/Number-Of-Pool/One/model/ecg_model_one_pool.h5 differ diff --git a/Number-Of-Pool/Three/logs/20230326-180546/train/events.out.tfevents.1679825150.WEIYUQING.15168.0.v2 b/Number-Of-Pool/Three/logs/20230326-180546/train/events.out.tfevents.1679825150.WEIYUQING.15168.0.v2 new file mode 100644 index 0000000..4068830 Binary files /dev/null and b/Number-Of-Pool/Three/logs/20230326-180546/train/events.out.tfevents.1679825150.WEIYUQING.15168.0.v2 differ diff --git a/Number-Of-Pool/Three/logs/20230326-180546/validation/events.out.tfevents.1679825156.WEIYUQING.15168.1.v2 b/Number-Of-Pool/Three/logs/20230326-180546/validation/events.out.tfevents.1679825156.WEIYUQING.15168.1.v2 new file mode 100644 index 0000000..e0d3221 Binary files /dev/null and b/Number-Of-Pool/Three/logs/20230326-180546/validation/events.out.tfevents.1679825156.WEIYUQING.15168.1.v2 differ diff --git a/Number-Of-Pool/Three/model/ecg_model_three_pool.h5 b/Number-Of-Pool/Three/model/ecg_model_three_pool.h5 new file mode 100644 index 0000000..8871373 Binary files /dev/null and b/Number-Of-Pool/Three/model/ecg_model_three_pool.h5 differ diff --git a/Number-Of-Pool/Two/logs/20230326-135829/train/events.out.tfevents.1679810312.WEIYUQING.3296.0.v2 b/Number-Of-Pool/Two/logs/20230326-135829/train/events.out.tfevents.1679810312.WEIYUQING.3296.0.v2 new file mode 100644 index 0000000..a9b2e72 Binary files /dev/null and b/Number-Of-Pool/Two/logs/20230326-135829/train/events.out.tfevents.1679810312.WEIYUQING.3296.0.v2 differ diff --git a/Number-Of-Pool/Two/logs/20230326-135829/validation/events.out.tfevents.1679810317.WEIYUQING.3296.1.v2 b/Number-Of-Pool/Two/logs/20230326-135829/validation/events.out.tfevents.1679810317.WEIYUQING.3296.1.v2 new file mode 100644 index 0000000..d943769 Binary files /dev/null and b/Number-Of-Pool/Two/logs/20230326-135829/validation/events.out.tfevents.1679810317.WEIYUQING.3296.1.v2 differ diff --git a/Number-Of-Pool/Two/logs/20230326-175724/train/events.out.tfevents.1679824649.WEIYUQING.27860.0.v2 b/Number-Of-Pool/Two/logs/20230326-175724/train/events.out.tfevents.1679824649.WEIYUQING.27860.0.v2 new file mode 100644 index 0000000..4cc1b2b Binary files /dev/null and b/Number-Of-Pool/Two/logs/20230326-175724/train/events.out.tfevents.1679824649.WEIYUQING.27860.0.v2 differ diff --git a/Number-Of-Pool/Two/logs/20230326-175724/validation/events.out.tfevents.1679824654.WEIYUQING.27860.1.v2 b/Number-Of-Pool/Two/logs/20230326-175724/validation/events.out.tfevents.1679824654.WEIYUQING.27860.1.v2 new file mode 100644 index 0000000..8539e0f Binary files /dev/null and b/Number-Of-Pool/Two/logs/20230326-175724/validation/events.out.tfevents.1679824654.WEIYUQING.27860.1.v2 differ diff --git a/Number-Of-Pool/Two/model/ecg_model_two_pool.h5 b/Number-Of-Pool/Two/model/ecg_model_two_pool.h5 new file mode 100644 index 0000000..7d0f546 Binary files /dev/null and b/Number-Of-Pool/Two/model/ecg_model_two_pool.h5 differ diff --git a/Number-Of-Pool/Two/picture/accuracy.png b/Number-Of-Pool/Two/picture/accuracy.png new file mode 100644 index 0000000..61e4170 Binary files /dev/null and b/Number-Of-Pool/Two/picture/accuracy.png differ diff --git a/Number-Of-Pool/Two/picture/confusion_matrix.png b/Number-Of-Pool/Two/picture/confusion_matrix.png new file mode 100644 index 0000000..70c94e4 Binary files /dev/null and b/Number-Of-Pool/Two/picture/confusion_matrix.png differ diff --git a/Number-Of-Pool/Two/picture/loss.png b/Number-Of-Pool/Two/picture/loss.png new file mode 100644 index 0000000..7ad35f0 Binary files /dev/null and b/Number-Of-Pool/Two/picture/loss.png differ diff --git a/Number-Of-Pool/_mian_tensorflow_Four_PoolCNN.py b/Number-Of-Pool/_mian_tensorflow_Four_PoolCNN.py new file mode 100644 index 0000000..829385a --- /dev/null +++ b/Number-Of-Pool/_mian_tensorflow_Four_PoolCNN.py @@ -0,0 +1,85 @@ +import datetime +import os + +import numpy as np +import pywt +import wfdb +import tensorflow as tf +from tensorflow import keras +from utils_two_of_PoolCNN import loadData, plot_history_tf, plot_heat_map + +#from ECG_read_weiyuqing_without_wfdb import ECGDATAPATH +#PATH of this test +Project_PATH = "../Number-Of-Pool/Four/" +#PICTUREPATH= Project_PATH + "picture/" +log_dir = Project_PATH + "logs/" + datetime.datetime.now().strftime("%Y%m%d-%H%M%S") +model_Path_one = Project_PATH + "model/ecg_model_four_pool.h5" +#model_Path_one = Project_PATH + "ecg_model_one_layer.h5" + +RATIO = 0.3 +RANDOM_SEED = 42 +BATCH_SIZE = 128 +#迭代次数 +NUM_EPOCHS = 10 + + +def CNN_model_level_one(): + leavlOneModel = tf.keras.models.Sequential([ + + #take test for one CNN layers model + + tf.keras.layers.InputLayer(input_shape=(300,)), + tf.keras.layers.Reshape(target_shape=(300, 1)), + + tf.keras.layers.Conv1D(filters=15, kernel_size=3, strides=2, padding='same', activation='relu'), + tf.keras.layers.MaxPool1D(pool_size=4, strides=2, padding='same'), + + + + tf.keras.layers.Flatten(), + tf.keras.layers.Dense(5, activation='softmax'), + # add dropout layer, dropout rate = 0.2 + # tf.keras.layers.Dropout(rate=0.2), + # tf.keras.layers.Dense(5, activation='softmax') + ]) + return leavlOneModel + + +def main(): + print(tf.config.list_physical_devices('GPU')) + # get traxin test + X_train, X_test, y_train, y_test = loadData(RATIO, RANDOM_SEED) + + # get model or create model + if os.path.exists(model_Path_one): + print(' get model in h5 file') + model = tf.keras.models.load_model(filepath=model_Path_one) + else: + # create new model(if model unexists) + model = CNN_model_level_one() + model.compile(optimizer='adam', + loss='sparse_categorical_crossentropy', + metrics=['accuracy']) + model.summary() + # TB make + tensorboard_callback = tf.keras.callbacks.TensorBoard(log_dir=log_dir, + histogram_freq=1, + ) + # Training and validation + history = model.fit(X_train, y_train, epochs=NUM_EPOCHS, + batch_size=BATCH_SIZE, + validation_data=(X_test, y_test), + callbacks=[tensorboard_callback]) + # validation_split=RATIO, + model.save(filepath=model_Path_one) + plot_history_tf(history) + + y_pred = np.argmax(model.predict(X_test), axis=-1) + plot_heat_map(y_test, y_pred) + + +if __name__ == '__main__': + os.environ["CUDA_VISIBLE_DEVICES"] = "-1" + tf.config.list_physical_devices('GPU') + with tf.device('/gpu:0'): + main() diff --git a/Number-Of-Pool/_mian_tensorflow_One_PoolCNN.py b/Number-Of-Pool/_mian_tensorflow_One_PoolCNN.py new file mode 100644 index 0000000..b4712b1 --- /dev/null +++ b/Number-Of-Pool/_mian_tensorflow_One_PoolCNN.py @@ -0,0 +1,82 @@ +import datetime +import os + +import numpy as np +import pywt +import wfdb +import tensorflow as tf +from tensorflow import keras +from utils_two_of_PoolCNN import loadData, plot_history_tf, plot_heat_map + +#from ECG_read_weiyuqing_without_wfdb import ECGDATAPATH +#PATH of this test +Project_PATH = "../Number-Of-Pool/One/" +#PICTUREPATH= Project_PATH + "picture/" +log_dir = Project_PATH + "logs/" + datetime.datetime.now().strftime("%Y%m%d-%H%M%S") +model_Path_one = Project_PATH + "model/ecg_model_one_pool.h5" +#model_Path_one = Project_PATH + "ecg_model_one_layer.h5" + +RATIO = 0.3 +RANDOM_SEED = 42 +BATCH_SIZE = 128 +#迭代次数 +NUM_EPOCHS = 10 + + +def CNN_model_level_one(): + leavlOneModel = tf.keras.models.Sequential([ + + #take test for one CNN layers model + + tf.keras.layers.InputLayer(input_shape=(300,)), + tf.keras.layers.Reshape(target_shape=(300, 1)), + + tf.keras.layers.Conv1D(filters=15, kernel_size=3, strides=2, padding='same', activation='relu'), + tf.keras.layers.MaxPool1D(pool_size=1, strides=2, padding='same'), + + + + tf.keras.layers.Flatten(), + tf.keras.layers.Dense(5, activation='softmax'), + # add dropout layer, dropout rate = 0.2 + # tf.keras.layers.Dropout(rate=0.2), + # tf.keras.layers.Dense(5, activation='softmax') + ]) + return leavlOneModel + + +def main(): + # get traxin test + X_train, X_test, y_train, y_test = loadData(RATIO, RANDOM_SEED) + + # get model or create model + if os.path.exists(model_Path_one): + print(' get model in h5 file') + model = tf.keras.models.load_model(filepath=model_Path_one) + else: + # create new model(if model unexists) + model = CNN_model_level_one() + model.compile(optimizer='adam', + loss='sparse_categorical_crossentropy', + metrics=['accuracy']) + model.summary() + # TB make + tensorboard_callback = tf.keras.callbacks.TensorBoard(log_dir=log_dir, + histogram_freq=1, + ) + # Training and validation + history = model.fit(X_train, y_train, epochs=NUM_EPOCHS, + batch_size=BATCH_SIZE, + validation_data=(X_test, y_test), + callbacks=[tensorboard_callback]) + # validation_split=RATIO, + model.save(filepath=model_Path_one) + plot_history_tf(history) + + y_pred = np.argmax(model.predict(X_test), axis=-1) + plot_heat_map(y_test, y_pred) + + +if __name__ == '__main__': + os.environ["CUDA_VISIBLE_DEVICES"] = "-1" + main() diff --git a/Number-Of-Pool/_mian_tensorflow_Three_PoolCNN.py b/Number-Of-Pool/_mian_tensorflow_Three_PoolCNN.py new file mode 100644 index 0000000..9674fee --- /dev/null +++ b/Number-Of-Pool/_mian_tensorflow_Three_PoolCNN.py @@ -0,0 +1,85 @@ +import datetime +import os + +import numpy as np +import pywt +import wfdb +import tensorflow as tf +from tensorflow import keras +from utils_two_of_PoolCNN import loadData, plot_history_tf, plot_heat_map + +#from ECG_read_weiyuqing_without_wfdb import ECGDATAPATH +#PATH of this test +Project_PATH = "../Number-Of-Pool/Three/" +#PICTUREPATH= Project_PATH + "picture/" +log_dir = Project_PATH + "logs/" + datetime.datetime.now().strftime("%Y%m%d-%H%M%S") +model_Path_one = Project_PATH + "model/ecg_model_three_pool.h5" +#model_Path_one = Project_PATH + "ecg_model_one_layer.h5" + +RATIO = 0.3 +RANDOM_SEED = 42 +BATCH_SIZE = 128 +#迭代次数 +NUM_EPOCHS = 10 + + +def CNN_model_level_one(): + leavlOneModel = tf.keras.models.Sequential([ + + #take test for one CNN layers model + + tf.keras.layers.InputLayer(input_shape=(300,)), + tf.keras.layers.Reshape(target_shape=(300, 1)), + + tf.keras.layers.Conv1D(filters=15, kernel_size=3, strides=2, padding='same', activation='relu'), + tf.keras.layers.MaxPool1D(pool_size=3, strides=2, padding='same'), + + + + tf.keras.layers.Flatten(), + tf.keras.layers.Dense(5, activation='softmax'), + # add dropout layer, dropout rate = 0.2 + # tf.keras.layers.Dropout(rate=0.2), + # tf.keras.layers.Dense(5, activation='softmax') + ]) + return leavlOneModel + + +def main(): + print(tf.config.list_physical_devices('GPU')) + # get traxin test + X_train, X_test, y_train, y_test = loadData(RATIO, RANDOM_SEED) + + # get model or create model + if os.path.exists(model_Path_one): + print(' get model in h5 file') + model = tf.keras.models.load_model(filepath=model_Path_one) + else: + # create new model(if model unexists) + model = CNN_model_level_one() + model.compile(optimizer='adam', + loss='sparse_categorical_crossentropy', + metrics=['accuracy']) + model.summary() + # TB make + tensorboard_callback = tf.keras.callbacks.TensorBoard(log_dir=log_dir, + histogram_freq=1, + ) + # Training and validation + history = model.fit(X_train, y_train, epochs=NUM_EPOCHS, + batch_size=BATCH_SIZE, + validation_data=(X_test, y_test), + callbacks=[tensorboard_callback]) + # validation_split=RATIO, + model.save(filepath=model_Path_one) + plot_history_tf(history) + + y_pred = np.argmax(model.predict(X_test), axis=-1) + plot_heat_map(y_test, y_pred) + + +if __name__ == '__main__': + os.environ["CUDA_VISIBLE_DEVICES"] = "-1" + tf.config.list_physical_devices('GPU') + with tf.device('/gpu:0'): + main() diff --git a/Number-Of-Pool/_mian_tensorflow_Two_PoolCNN.py b/Number-Of-Pool/_mian_tensorflow_Two_PoolCNN.py new file mode 100644 index 0000000..74aa862 --- /dev/null +++ b/Number-Of-Pool/_mian_tensorflow_Two_PoolCNN.py @@ -0,0 +1,82 @@ +import datetime +import os + +import numpy as np +import pywt +import wfdb +import tensorflow as tf +from tensorflow import keras +from utils_two_of_PoolCNN import loadData, plot_history_tf, plot_heat_map + +#from ECG_read_weiyuqing_without_wfdb import ECGDATAPATH +#PATH of this test +Project_PATH = "../Number-Of-Pool/Two/" +#PICTUREPATH= Project_PATH + "picture/" +log_dir = Project_PATH + "logs/" + datetime.datetime.now().strftime("%Y%m%d-%H%M%S") +model_Path_one = Project_PATH + "model/ecg_model_two_pool.h5" +#model_Path_one = Project_PATH + "ecg_model_one_layer.h5" + +RATIO = 0.3 +RANDOM_SEED = 42 +BATCH_SIZE = 128 +#迭代次数 +NUM_EPOCHS = 10 + + +def CNN_model_level_one(): + leavlOneModel = tf.keras.models.Sequential([ + + #take test for one CNN layers model + + tf.keras.layers.InputLayer(input_shape=(300,)), + tf.keras.layers.Reshape(target_shape=(300, 1)), + + tf.keras.layers.Conv1D(filters=15, kernel_size=3, strides=2, padding='same', activation='relu'), + tf.keras.layers.MaxPool1D(pool_size=2, strides=2, padding='same'), + + + + tf.keras.layers.Flatten(), + tf.keras.layers.Dense(5, activation='softmax'), + # add dropout layer, dropout rate = 0.2 + # tf.keras.layers.Dropout(rate=0.2), + # tf.keras.layers.Dense(5, activation='softmax') + ]) + return leavlOneModel + + +def main(): + # get traxin test + X_train, X_test, y_train, y_test = loadData(RATIO, RANDOM_SEED) + + # get model or create model + if os.path.exists(model_Path_one): + print(' get model in h5 file') + model = tf.keras.models.load_model(filepath=model_Path_one) + else: + # create new model(if model unexists) + model = CNN_model_level_one() + model.compile(optimizer='adam', + loss='sparse_categorical_crossentropy', + metrics=['accuracy']) + model.summary() + # TB make + tensorboard_callback = tf.keras.callbacks.TensorBoard(log_dir=log_dir, + histogram_freq=1, + ) + # Training and validation + history = model.fit(X_train, y_train, epochs=NUM_EPOCHS, + batch_size=BATCH_SIZE, + validation_data=(X_test, y_test), + callbacks=[tensorboard_callback]) + # validation_split=RATIO, + model.save(filepath=model_Path_one) + plot_history_tf(history) + + y_pred = np.argmax(model.predict(X_test), axis=-1) + plot_heat_map(y_test, y_pred) + + +if __name__ == '__main__': + os.environ["CUDA_VISIBLE_DEVICES"] = "-1" + main() diff --git a/Number-Of-Pool/utils_four_of_PoolCNN.py b/Number-Of-Pool/utils_four_of_PoolCNN.py new file mode 100644 index 0000000..eaf1778 --- /dev/null +++ b/Number-Of-Pool/utils_four_of_PoolCNN.py @@ -0,0 +1,172 @@ +# this file give +import os + +import wfdb +import pywt +import seaborn +import numpy as np +import matplotlib.pyplot as plt +from sklearn.metrics import confusion_matrix +from sklearn.model_selection import train_test_split + +# from _mian_tensorflow import Project_PATH +Project_PATH = '../Number-Of-Pool/Four/' + +PICTUREPATH = Project_PATH + 'picture/' + +# from ECG_read_weiyuqing_without_wfdb import ECGDATAPATH +MITBITECGDATAPATH = '../MIT-BIT/' + + +# wavelet denoise preprocess using mallat algorithm + +def denoise(date): + # Wavelet db5 level9 + coeffs = pywt.wavedec(data=date, wavelet='db5', level=9) + cA9, cD9, cD8, cD7, cD6, cD5, cD4, cD3, cD2, cD1 = coeffs + # noice del + threshold = (np.median(np.abs(cD1)) / 0.6745) * (np.sqrt(2 * np.log(len(cD1)))) + # take cD 1,2 turn 0 to denoise + cD1.fill(0) + cD2.fill(0) + for i in range(1, len(coeffs) - 2): + coeffs[i] = pywt.threshold(coeffs[i], threshold) + # get signal that before wavelet + rdata = pywt.waverec(coeffs=coeffs, wavelet='db5') + return rdata + + +# get ECG data,date type and denoise +# this function without return +# use append take return data to pass data +def getDataSet(number, X_data, Y_data): + ecgClassSet = ['N', 'A', 'V', 'L', 'R'] + print('get' + number + ' ECG data') + record = wfdb.rdrecord(MITBITECGDATAPATH + number, channel_names=['MLII']) + data = record.p_signal.flatten() + # use def denoise + rdata = denoise(date=data) + + # get type of ECG data + annotation = wfdb.rdann(MITBITECGDATAPATH + number, 'atr') + Rlocation = annotation.sample # annotation.sample:get R + Rclass = annotation.symbol + + # delete stable date near R + start = 10 + end = 5 + i = start + j = len(annotation.symbol) - end + + # get NAVLR type data + while i < j: + try: + lable = ecgClassSet.index(Rclass[i]) + x_train = rdata[Rlocation[i] - 99:Rlocation[i] + 201] + X_data.append(x_train) + Y_data.append(lable) + i += 1 + except ValueError: + i += 1 + return + + +# RATIO=The ratio of the training set to the test set，usually is 0.3 +def loadData(ratio, random_seed): + numberSet = ['100', '101', '103', '105', '106', '107', '108', '109', '111', '112', '113', '114', '115', + '116', '117', '119', '121', '122', '123', '124', '200', '201', '202', '203', '205', '208', + '210', '212', '213', '214', '215', '217', '219', '220', '221', '222', '223', '228', '230', + '231', '232', '233', '234'] + dataSet = [] + labelSet = [] + for n in numberSet: + getDataSet(n, dataSet, labelSet) + + # dataSet = np.array(dataSet).reshape(-1, 300) + # labelSet = np.array(labelSet).reshape(-1, 1) + + # reshape the data and split the dataset + dataSet = np.array(dataSet).reshape(-1, 300) + lableSet = np.array(labelSet).reshape(-1) + X_train, X_test, y_train, y_test = train_test_split(dataSet, lableSet, test_size=ratio, random_state=random_seed) + # print(X_train, X_test, y_train, y_test) + return X_train, X_test, y_train, y_test + + # train_ds = np.hstack((dataSet, labelSet)) + # np.random.shuffle(train_ds) + # + # X = train_ds[:, :300].reshape(-1, 300, 1) + # Y = train_ds[:, 300] + # # cut test and train data from ECG data + # shuffle_index = np.random.permutation(len(X)) + # test_length = int(RATIO * len(shuffle_index)) + # test_index = shuffle_index[:test_length] + # train_index = shuffle_index[test_index:] + # X_test, Y_test = X[test_index:], Y[test_index] + # X_train, Y_train = X[train_index], Y[train_index] + # return X_train, Y_test, X_test, Y_train + + +# confusion matrix +def plot_heat_map(y_test, y_pred): + con_mat = confusion_matrix(y_test, y_pred) + # normalize + # con_mat_norm = con_mat.astype('float') / con_mat.sum(axis=1)[:, np.newaxis] + # con_mat_norm = np.around(con_mat_norm, decimals=2) + # os.remove(PICTUREPATH+'confusion_matrix.png') + # plot + plt.figure(figsize=(8, 8)) + seaborn.heatmap(con_mat, annot=True, fmt='.20g', cmap='Blues') + plt.ylim(0, 5) + plt.xlabel('Predicted labels') + plt.ylabel('True labels') + plt.title('Confusion Matrix') + plt.savefig(PICTUREPATH + 'confusion_matrix.png') + plt.show() + + +def plot_history_tf(history): + # os.remove(PICTUREPATH+'accuracy.png') + # os.remove(PICTUREPATH+'loss.png') + + plt.figure(figsize=(8, 8)) + plt.plot(history.history['accuracy']) + plt.plot(history.history['val_accuracy']) + plt.title('Model Accuracy') + plt.ylabel('Accuracy') + plt.xlabel('Epoch') + plt.legend(['Train', 'Test'], loc='upper left') + plt.savefig(PICTUREPATH + 'accuracy.png') + plt.show() + + plt.figure(figsize=(8, 8)) + plt.plot(history.history['loss']) + plt.plot(history.history['val_loss']) + plt.title('Model Loss') + plt.ylabel('Loss') + plt.xlabel('Epoch') + plt.legend(['Train', 'Test'], loc='upper left') + plt.savefig(PICTUREPATH + 'loss.png') + plt.show() + + +def plot_history_torch(history): + plt.figure(figsize=(8, 8)) + plt.plot(history['train_acc']) + plt.plot(history['test_acc']) + plt.title('Model Accuracy') + plt.ylabel('Accuracy') + plt.xlabel('Epoch') + plt.legend(['Train', 'Test'], loc='upper left') + plt.savefig(PICTUREPATH + 'accuracy.png') + plt.show() + + plt.figure(figsize=(8, 8)) + plt.plot(history['train_loss']) + plt.plot(history['test_loss']) + plt.title('Model Loss') + plt.ylabel('Loss') + plt.xlabel('Epoch') + plt.legend(['Train', 'Test'], loc='upper left') + plt.savefig(PICTUREPATH + 'loss.png') + plt.show() diff --git a/Number-Of-Pool/utils_one_of_PoolCNN.py b/Number-Of-Pool/utils_one_of_PoolCNN.py new file mode 100644 index 0000000..5ed3590 --- /dev/null +++ b/Number-Of-Pool/utils_one_of_PoolCNN.py @@ -0,0 +1,172 @@ +# this file give +import os + +import wfdb +import pywt +import seaborn +import numpy as np +import matplotlib.pyplot as plt +from sklearn.metrics import confusion_matrix +from sklearn.model_selection import train_test_split + +# from _mian_tensorflow import Project_PATH +Project_PATH = '../Number-Of-Pool/One/' + +PICTUREPATH = Project_PATH + 'picture/' + +# from ECG_read_weiyuqing_without_wfdb import ECGDATAPATH +MITBITECGDATAPATH = '../MIT-BIT/' + + +# wavelet denoise preprocess using mallat algorithm + +def denoise(date): + # Wavelet db5 level9 + coeffs = pywt.wavedec(data=date, wavelet='db5', level=9) + cA9, cD9, cD8, cD7, cD6, cD5, cD4, cD3, cD2, cD1 = coeffs + # noice del + threshold = (np.median(np.abs(cD1)) / 0.6745) * (np.sqrt(2 * np.log(len(cD1)))) + # take cD 1,2 turn 0 to denoise + cD1.fill(0) + cD2.fill(0) + for i in range(1, len(coeffs) - 2): + coeffs[i] = pywt.threshold(coeffs[i], threshold) + # get signal that before wavelet + rdata = pywt.waverec(coeffs=coeffs, wavelet='db5') + return rdata + + +# get ECG data,date type and denoise +# this function without return +# use append take return data to pass data +def getDataSet(number, X_data, Y_data): + ecgClassSet = ['N', 'A', 'V', 'L', 'R'] + print('get' + number + ' ECG data') + record = wfdb.rdrecord(MITBITECGDATAPATH + number, channel_names=['MLII']) + data = record.p_signal.flatten() + # use def denoise + rdata = denoise(date=data) + + # get type of ECG data + annotation = wfdb.rdann(MITBITECGDATAPATH + number, 'atr') + Rlocation = annotation.sample # annotation.sample:get R + Rclass = annotation.symbol + + # delete stable date near R + start = 10 + end = 5 + i = start + j = len(annotation.symbol) - end + + # get NAVLR type data + while i < j: + try: + lable = ecgClassSet.index(Rclass[i]) + x_train = rdata[Rlocation[i] - 99:Rlocation[i] + 201] + X_data.append(x_train) + Y_data.append(lable) + i += 1 + except ValueError: + i += 1 + return + + +# RATIO=The ratio of the training set to the test set，usually is 0.3 +def loadData(ratio, random_seed): + numberSet = ['100', '101', '103', '105', '106', '107', '108', '109', '111', '112', '113', '114', '115', + '116', '117', '119', '121', '122', '123', '124', '200', '201', '202', '203', '205', '208', + '210', '212', '213', '214', '215', '217', '219', '220', '221', '222', '223', '228', '230', + '231', '232', '233', '234'] + dataSet = [] + labelSet = [] + for n in numberSet: + getDataSet(n, dataSet, labelSet) + + # dataSet = np.array(dataSet).reshape(-1, 300) + # labelSet = np.array(labelSet).reshape(-1, 1) + + # reshape the data and split the dataset + dataSet = np.array(dataSet).reshape(-1, 300) + lableSet = np.array(labelSet).reshape(-1) + X_train, X_test, y_train, y_test = train_test_split(dataSet, lableSet, test_size=ratio, random_state=random_seed) + # print(X_train, X_test, y_train, y_test) + return X_train, X_test, y_train, y_test + + # train_ds = np.hstack((dataSet, labelSet)) + # np.random.shuffle(train_ds) + # + # X = train_ds[:, :300].reshape(-1, 300, 1) + # Y = train_ds[:, 300] + # # cut test and train data from ECG data + # shuffle_index = np.random.permutation(len(X)) + # test_length = int(RATIO * len(shuffle_index)) + # test_index = shuffle_index[:test_length] + # train_index = shuffle_index[test_index:] + # X_test, Y_test = X[test_index:], Y[test_index] + # X_train, Y_train = X[train_index], Y[train_index] + # return X_train, Y_test, X_test, Y_train + + +# confusion matrix +def plot_heat_map(y_test, y_pred): + con_mat = confusion_matrix(y_test, y_pred) + # normalize + # con_mat_norm = con_mat.astype('float') / con_mat.sum(axis=1)[:, np.newaxis] + # con_mat_norm = np.around(con_mat_norm, decimals=2) + # os.remove(PICTUREPATH+'confusion_matrix.png') + # plot + plt.figure(figsize=(8, 8)) + seaborn.heatmap(con_mat, annot=True, fmt='.20g', cmap='Blues') + plt.ylim(0, 5) + plt.xlabel('Predicted labels') + plt.ylabel('True labels') + plt.title('Confusion Matrix') + plt.savefig(PICTUREPATH + 'confusion_matrix.png') + plt.show() + + +def plot_history_tf(history): + # os.remove(PICTUREPATH+'accuracy.png') + # os.remove(PICTUREPATH+'loss.png') + + plt.figure(figsize=(8, 8)) + plt.plot(history.history['accuracy']) + plt.plot(history.history['val_accuracy']) + plt.title('Model Accuracy') + plt.ylabel('Accuracy') + plt.xlabel('Epoch') + plt.legend(['Train', 'Test'], loc='upper left') + plt.savefig(PICTUREPATH + 'accuracy.png') + plt.show() + + plt.figure(figsize=(8, 8)) + plt.plot(history.history['loss']) + plt.plot(history.history['val_loss']) + plt.title('Model Loss') + plt.ylabel('Loss') + plt.xlabel('Epoch') + plt.legend(['Train', 'Test'], loc='upper left') + plt.savefig(PICTUREPATH + 'loss.png') + plt.show() + + +def plot_history_torch(history): + plt.figure(figsize=(8, 8)) + plt.plot(history['train_acc']) + plt.plot(history['test_acc']) + plt.title('Model Accuracy') + plt.ylabel('Accuracy') + plt.xlabel('Epoch') + plt.legend(['Train', 'Test'], loc='upper left') + plt.savefig(PICTUREPATH + 'accuracy.png') + plt.show() + + plt.figure(figsize=(8, 8)) + plt.plot(history['train_loss']) + plt.plot(history['test_loss']) + plt.title('Model Loss') + plt.ylabel('Loss') + plt.xlabel('Epoch') + plt.legend(['Train', 'Test'], loc='upper left') + plt.savefig(PICTUREPATH + 'loss.png') + plt.show() diff --git a/Number-Of-Pool/utils_three_of_PoolCNN.py b/Number-Of-Pool/utils_three_of_PoolCNN.py new file mode 100644 index 0000000..8e007a5 --- /dev/null +++ b/Number-Of-Pool/utils_three_of_PoolCNN.py @@ -0,0 +1,172 @@ +# this file give +import os + +import wfdb +import pywt +import seaborn +import numpy as np +import matplotlib.pyplot as plt +from sklearn.metrics import confusion_matrix +from sklearn.model_selection import train_test_split + +# from _mian_tensorflow import Project_PATH +Project_PATH = '../Number-Of-Pool/Three/' + +PICTUREPATH = Project_PATH + 'picture/' + +# from ECG_read_weiyuqing_without_wfdb import ECGDATAPATH +MITBITECGDATAPATH = '../MIT-BIT/' + + +# wavelet denoise preprocess using mallat algorithm + +def denoise(date): + # Wavelet db5 level9 + coeffs = pywt.wavedec(data=date, wavelet='db5', level=9) + cA9, cD9, cD8, cD7, cD6, cD5, cD4, cD3, cD2, cD1 = coeffs + # noice del + threshold = (np.median(np.abs(cD1)) / 0.6745) * (np.sqrt(2 * np.log(len(cD1)))) + # take cD 1,2 turn 0 to denoise + cD1.fill(0) + cD2.fill(0) + for i in range(1, len(coeffs) - 2): + coeffs[i] = pywt.threshold(coeffs[i], threshold) + # get signal that before wavelet + rdata = pywt.waverec(coeffs=coeffs, wavelet='db5') + return rdata + + +# get ECG data,date type and denoise +# this function without return +# use append take return data to pass data +def getDataSet(number, X_data, Y_data): + ecgClassSet = ['N', 'A', 'V', 'L', 'R'] + print('get' + number + ' ECG data') + record = wfdb.rdrecord(MITBITECGDATAPATH + number, channel_names=['MLII']) + data = record.p_signal.flatten() + # use def denoise + rdata = denoise(date=data) + + # get type of ECG data + annotation = wfdb.rdann(MITBITECGDATAPATH + number, 'atr') + Rlocation = annotation.sample # annotation.sample:get R + Rclass = annotation.symbol + + # delete stable date near R + start = 10 + end = 5 + i = start + j = len(annotation.symbol) - end + + # get NAVLR type data + while i < j: + try: + lable = ecgClassSet.index(Rclass[i]) + x_train = rdata[Rlocation[i] - 99:Rlocation[i] + 201] + X_data.append(x_train) + Y_data.append(lable) + i += 1 + except ValueError: + i += 1 + return + + +# RATIO=The ratio of the training set to the test set，usually is 0.3 +def loadData(ratio, random_seed): + numberSet = ['100', '101', '103', '105', '106', '107', '108', '109', '111', '112', '113', '114', '115', + '116', '117', '119', '121', '122', '123', '124', '200', '201', '202', '203', '205', '208', + '210', '212', '213', '214', '215', '217', '219', '220', '221', '222', '223', '228', '230', + '231', '232', '233', '234'] + dataSet = [] + labelSet = [] + for n in numberSet: + getDataSet(n, dataSet, labelSet) + + # dataSet = np.array(dataSet).reshape(-1, 300) + # labelSet = np.array(labelSet).reshape(-1, 1) + + # reshape the data and split the dataset + dataSet = np.array(dataSet).reshape(-1, 300) + lableSet = np.array(labelSet).reshape(-1) + X_train, X_test, y_train, y_test = train_test_split(dataSet, lableSet, test_size=ratio, random_state=random_seed) + # print(X_train, X_test, y_train, y_test) + return X_train, X_test, y_train, y_test + + # train_ds = np.hstack((dataSet, labelSet)) + # np.random.shuffle(train_ds) + # + # X = train_ds[:, :300].reshape(-1, 300, 1) + # Y = train_ds[:, 300] + # # cut test and train data from ECG data + # shuffle_index = np.random.permutation(len(X)) + # test_length = int(RATIO * len(shuffle_index)) + # test_index = shuffle_index[:test_length] + # train_index = shuffle_index[test_index:] + # X_test, Y_test = X[test_index:], Y[test_index] + # X_train, Y_train = X[train_index], Y[train_index] + # return X_train, Y_test, X_test, Y_train + + +# confusion matrix +def plot_heat_map(y_test, y_pred): + con_mat = confusion_matrix(y_test, y_pred) + # normalize + # con_mat_norm = con_mat.astype('float') / con_mat.sum(axis=1)[:, np.newaxis] + # con_mat_norm = np.around(con_mat_norm, decimals=2) + # os.remove(PICTUREPATH+'confusion_matrix.png') + # plot + plt.figure(figsize=(8, 8)) + seaborn.heatmap(con_mat, annot=True, fmt='.20g', cmap='Blues') + plt.ylim(0, 5) + plt.xlabel('Predicted labels') + plt.ylabel('True labels') + plt.title('Confusion Matrix') + plt.savefig(PICTUREPATH + 'confusion_matrix.png') + plt.show() + + +def plot_history_tf(history): + # os.remove(PICTUREPATH+'accuracy.png') + # os.remove(PICTUREPATH+'loss.png') + + plt.figure(figsize=(8, 8)) + plt.plot(history.history['accuracy']) + plt.plot(history.history['val_accuracy']) + plt.title('Model Accuracy') + plt.ylabel('Accuracy') + plt.xlabel('Epoch') + plt.legend(['Train', 'Test'], loc='upper left') + plt.savefig(PICTUREPATH + 'accuracy.png') + plt.show() + + plt.figure(figsize=(8, 8)) + plt.plot(history.history['loss']) + plt.plot(history.history['val_loss']) + plt.title('Model Loss') + plt.ylabel('Loss') + plt.xlabel('Epoch') + plt.legend(['Train', 'Test'], loc='upper left') + plt.savefig(PICTUREPATH + 'loss.png') + plt.show() + + +def plot_history_torch(history): + plt.figure(figsize=(8, 8)) + plt.plot(history['train_acc']) + plt.plot(history['test_acc']) + plt.title('Model Accuracy') + plt.ylabel('Accuracy') + plt.xlabel('Epoch') + plt.legend(['Train', 'Test'], loc='upper left') + plt.savefig(PICTUREPATH + 'accuracy.png') + plt.show() + + plt.figure(figsize=(8, 8)) + plt.plot(history['train_loss']) + plt.plot(history['test_loss']) + plt.title('Model Loss') + plt.ylabel('Loss') + plt.xlabel('Epoch') + plt.legend(['Train', 'Test'], loc='upper left') + plt.savefig(PICTUREPATH + 'loss.png') + plt.show() diff --git a/Number-Of-Pool/utils_two_of_PoolCNN.py b/Number-Of-Pool/utils_two_of_PoolCNN.py new file mode 100644 index 0000000..2a0de78 --- /dev/null +++ b/Number-Of-Pool/utils_two_of_PoolCNN.py @@ -0,0 +1,172 @@ +# this file give +import os + +import wfdb +import pywt +import seaborn +import numpy as np +import matplotlib.pyplot as plt +from sklearn.metrics import confusion_matrix +from sklearn.model_selection import train_test_split + +# from _mian_tensorflow import Project_PATH +Project_PATH = '../Number-Of-Pool/Two/' + +PICTUREPATH = Project_PATH + 'picture/' + +# from ECG_read_weiyuqing_without_wfdb import ECGDATAPATH +MITBITECGDATAPATH = '../MIT-BIT/' + + +# wavelet denoise preprocess using mallat algorithm + +def denoise(date): + # Wavelet db5 level9 + coeffs = pywt.wavedec(data=date, wavelet='db5', level=9) + cA9, cD9, cD8, cD7, cD6, cD5, cD4, cD3, cD2, cD1 = coeffs + # noice del + threshold = (np.median(np.abs(cD1)) / 0.6745) * (np.sqrt(2 * np.log(len(cD1)))) + # take cD 1,2 turn 0 to denoise + cD1.fill(0) + cD2.fill(0) + for i in range(1, len(coeffs) - 2): + coeffs[i] = pywt.threshold(coeffs[i], threshold) + # get signal that before wavelet + rdata = pywt.waverec(coeffs=coeffs, wavelet='db5') + return rdata + + +# get ECG data,date type and denoise +# this function without return +# use append take return data to pass data +def getDataSet(number, X_data, Y_data): + ecgClassSet = ['N', 'A', 'V', 'L', 'R'] + print('get' + number + ' ECG data') + record = wfdb.rdrecord(MITBITECGDATAPATH + number, channel_names=['MLII']) + data = record.p_signal.flatten() + # use def denoise + rdata = denoise(date=data) + + # get type of ECG data + annotation = wfdb.rdann(MITBITECGDATAPATH + number, 'atr') + Rlocation = annotation.sample # annotation.sample:get R + Rclass = annotation.symbol + + # delete stable date near R + start = 10 + end = 5 + i = start + j = len(annotation.symbol) - end + + # get NAVLR type data + while i < j: + try: + lable = ecgClassSet.index(Rclass[i]) + x_train = rdata[Rlocation[i] - 99:Rlocation[i] + 201] + X_data.append(x_train) + Y_data.append(lable) + i += 1 + except ValueError: + i += 1 + return + + +# RATIO=The ratio of the training set to the test set，usually is 0.3 +def loadData(ratio, random_seed): + numberSet = ['100', '101', '103', '105', '106', '107', '108', '109', '111', '112', '113', '114', '115', + '116', '117', '119', '121', '122', '123', '124', '200', '201', '202', '203', '205', '208', + '210', '212', '213', '214', '215', '217', '219', '220', '221', '222', '223', '228', '230', + '231', '232', '233', '234'] + dataSet = [] + labelSet = [] + for n in numberSet: + getDataSet(n, dataSet, labelSet) + + # dataSet = np.array(dataSet).reshape(-1, 300) + # labelSet = np.array(labelSet).reshape(-1, 1) + + # reshape the data and split the dataset + dataSet = np.array(dataSet).reshape(-1, 300) + lableSet = np.array(labelSet).reshape(-1) + X_train, X_test, y_train, y_test = train_test_split(dataSet, lableSet, test_size=ratio, random_state=random_seed) + # print(X_train, X_test, y_train, y_test) + return X_train, X_test, y_train, y_test + + # train_ds = np.hstack((dataSet, labelSet)) + # np.random.shuffle(train_ds) + # + # X = train_ds[:, :300].reshape(-1, 300, 1) + # Y = train_ds[:, 300] + # # cut test and train data from ECG data + # shuffle_index = np.random.permutation(len(X)) + # test_length = int(RATIO * len(shuffle_index)) + # test_index = shuffle_index[:test_length] + # train_index = shuffle_index[test_index:] + # X_test, Y_test = X[test_index:], Y[test_index] + # X_train, Y_train = X[train_index], Y[train_index] + # return X_train, Y_test, X_test, Y_train + + +# confusion matrix +def plot_heat_map(y_test, y_pred): + con_mat = confusion_matrix(y_test, y_pred) + # normalize + # con_mat_norm = con_mat.astype('float') / con_mat.sum(axis=1)[:, np.newaxis] + # con_mat_norm = np.around(con_mat_norm, decimals=2) + # os.remove(PICTUREPATH+'confusion_matrix.png') + # plot + plt.figure(figsize=(8, 8)) + seaborn.heatmap(con_mat, annot=True, fmt='.20g', cmap='Blues') + plt.ylim(0, 5) + plt.xlabel('Predicted labels') + plt.ylabel('True labels') + plt.title('Confusion Matrix') + plt.savefig(PICTUREPATH + 'confusion_matrix.png') + plt.show() + + +def plot_history_tf(history): + # os.remove(PICTUREPATH+'accuracy.png') + # os.remove(PICTUREPATH+'loss.png') + + plt.figure(figsize=(8, 8)) + plt.plot(history.history['accuracy']) + plt.plot(history.history['val_accuracy']) + plt.title('Model Accuracy') + plt.ylabel('Accuracy') + plt.xlabel('Epoch') + plt.legend(['Train', 'Test'], loc='upper left') + plt.savefig(PICTUREPATH + 'accuracy.png') + plt.show() + + plt.figure(figsize=(8, 8)) + plt.plot(history.history['loss']) + plt.plot(history.history['val_loss']) + plt.title('Model Loss') + plt.ylabel('Loss') + plt.xlabel('Epoch') + plt.legend(['Train', 'Test'], loc='upper left') + plt.savefig(PICTUREPATH + 'loss.png') + plt.show() + + +def plot_history_torch(history): + plt.figure(figsize=(8, 8)) + plt.plot(history['train_acc']) + plt.plot(history['test_acc']) + plt.title('Model Accuracy') + plt.ylabel('Accuracy') + plt.xlabel('Epoch') + plt.legend(['Train', 'Test'], loc='upper left') + plt.savefig(PICTUREPATH + 'accuracy.png') + plt.show() + + plt.figure(figsize=(8, 8)) + plt.plot(history['train_loss']) + plt.plot(history['test_loss']) + plt.title('Model Loss') + plt.ylabel('Loss') + plt.xlabel('Epoch') + plt.legend(['Train', 'Test'], loc='upper left') + plt.savefig(PICTUREPATH + 'loss.png') + plt.show() diff --git a/some-test/tf_run_in_gpu.py b/some-test/tf_run_in_gpu.py index 662f7f3..9e5bd92 100644 --- a/some-test/tf_run_in_gpu.py +++ b/some-test/tf_run_in_gpu.py @@ -13,6 +13,7 @@ print("cpu_b:", cpu_b.device) print("cpu_c:", cpu_c.device) # 查看gpu是否可用 print(tf.config.list_physical_devices('GPU')) +print('asdfa') # 指定在gpu上运行 with tf.device('/gpu:0'): gpu_a = tf.random.normal([10000, 1000])