Add files via upload

heartsound/1.py

@@ -0,0 +1,5 @@
+import pandas as pd
+import csv
+with open('1.csv', 'rb') as f:
+    reader = csv.reader(f)
+    your_list = list(reader)

heartsound/dataseta/artifact.py

@@ -0,0 +1,13 @@
+import os
+from scipy.io import wavfile
+import numpy as np
+
+with file('trainartifact.csv', 'w') as outfile:
+  for file in os.listdir('atrain/Atraining_artifact/Atraining_artifact'):
+
+    if file.endswith(".wav"):
+
+        rate, wf = wavfile.read('atrain/Atraining_artifact/Atraining_artifact/'+ file)
+        print(rate)
+        np.savetxt(outfile, [wf],fmt='%01d',delimiter=',')
+        #print(wf.shape)

heartsound/dataseta/extrastole.py

@@ -0,0 +1,13 @@
+import os
+from scipy.io import wavfile
+import numpy as np
+
+with file('extrastole.csv', 'w') as outfile:
+  for file in os.listdir('atrain/Atraining_extrahs/Atraining_extrahls'):
+
+    if file.endswith(".wav"):
+
+        rate, wf = wavfile.read('atrain/Atraining_extrahs/Atraining_extrahls/'+ file)
+        print(rate)
+        np.savetxt(outfile, [wf],fmt='%01d',delimiter=',')
+

heartsound/dataseta/ml/across.py

@@ -0,0 +1,77 @@
+from __future__ import print_function
+import numpy as np
+np.random.seed(1337)  # for reproducibility
+
+from keras.preprocessing import sequence
+from keras.models import Sequential
+from keras.layers import Dense, Dropout, Activation, Lambda
+from keras.layers import Embedding
+from keras.layers import Convolution1D,MaxPooling1D, Flatten
+from keras.datasets import imdb
+from keras import backend as K
+from sklearn.cross_validation import train_test_split
+import pandas as pd
+from keras.utils.np_utils import to_categorical
+
+from sklearn.preprocessing import Normalizer
+from keras.models import Sequential
+from keras.layers import Convolution1D, Dense, Dropout, Flatten, MaxPooling1D
+from keras.utils import np_utils
+import numpy as np
+import h5py
+from keras import callbacks
+from keras.layers import LSTM, GRU, SimpleRNN
+from keras.callbacks import CSVLogger
+from keras.callbacks import ModelCheckpoint, EarlyStopping, ReduceLROnPlateau, CSVLogger
+import csv
+from sklearn.cross_validation import StratifiedKFold
+from sklearn.cross_validation import cross_val_score
+from keras.wrappers.scikit_learn import KerasClassifier
+
+with open('dataa.csv', 'rb') as f:
+    reader = csv.reader(f)
+    your_list = list(reader)
+
+trainX = np.array(your_list)
+
+traindata = pd.read_csv('dataalabels.csv', header=None)
+Y = traindata.iloc[:,0]
+y_train1 = np.array(Y)
+y_train= to_categorical(y_train1)
+
+maxlen = 44100
+trainX = sequence.pad_sequences(trainX, maxlen=maxlen)
+
+# reshape input to be [samples, time steps, features]
+X_train = np.reshape(trainX, (trainX.shape[0], 1, trainX.shape[1]))
+
+
+def create_model():
+   print("----------------------------------")
+   model = Sequential()
+   model.add(GRU(64,input_dim=44100,return_sequences=True)) 
+   model.add(Dropout(0.1))
+   model.add(GRU(64, return_sequences=True))
+   model.add(Dropout(0.1))
+   model.add(GRU(64, return_sequences=True))
+   model.add(Dropout(0.1))
+   model.add(GRU(64, return_sequences=False))
+   model.add(Dropout(0.1))
+   model.add(Dense(4))
+   model.add(Activation('softmax'))
+   print("epoch")
+   model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])
+   return model
+
+seed = 7
+np.random.seed(seed)
+
+model = KerasClassifier(build_fn=create_model, epochs=50, batch_size=2)
+
+# evaluate using 10-fold cross validation
+kfold = StratifiedKFold(y=y_train1, n_folds=10, shuffle=True, random_state=seed)
+results = cross_val_score(model, X_train, y_train, cv=kfold)
+print(results)
+print("results mean")
+print(results.mean())
+

heartsound/dataseta/ml/trainandtest/cnn.py

@@ -0,0 +1,64 @@
+from __future__ import print_function
+import numpy as np
+np.random.seed(1337)  # for reproducibility
+
+from keras.preprocessing import sequence
+from keras.models import Sequential
+from keras.layers import Dense, Dropout, Activation, Lambda
+from keras.layers import Embedding
+from keras.layers import Convolution1D,MaxPooling1D, Flatten
+from keras.datasets import imdb
+from keras import backend as K
+from sklearn.cross_validation import train_test_split
+import pandas as pd
+from keras.utils.np_utils import to_categorical
+
+from sklearn.preprocessing import Normalizer
+from keras.models import Sequential
+from keras.layers import Convolution1D, GlobalMaxPooling1D,Dense, Dropout, Flatten, MaxPooling1D
+from keras.utils import np_utils
+import numpy as np
+import h5py
+from keras import callbacks
+from keras.layers import LSTM, GRU, SimpleRNN
+from keras.callbacks import CSVLogger
+from keras.callbacks import ModelCheckpoint, EarlyStopping, ReduceLROnPlateau, CSVLogger
+import csv
+from sklearn.cross_validation import StratifiedKFold
+from sklearn.cross_validation import cross_val_score
+from keras.wrappers.scikit_learn import KerasClassifier
+
+with open('a/traindata.csv', 'rb') as f:
+    reader = csv.reader(f)
+    your_list = list(reader)
+
+trainX = np.array(your_list)
+
+traindata = pd.read_csv('a/trainlabels.csv', header=None)
+Y = traindata.iloc[:,0]
+y_train1 = np.array(Y)
+y_train= to_categorical(y_train1)
+
+maxlen = 44100
+trainX = sequence.pad_sequences(trainX, maxlen=maxlen)
+
+print(trainX.shape)
+# reshape input to be [samples, time steps, features]
+X_train = np.reshape(trainX, (trainX.shape[0], trainX.shape[1],1))
+
+batch_size = 2
+
+model = Sequential()
+model.add(Convolution1D(128, 3, border_mode="same",activation="relu",input_shape=(44100, 1)))
+model.add(MaxPooling1D(pool_size=(2)))
+#model.add(Convolution1D(256, 6, border_mode="same",activation="relu"))
+#model.add(MaxPooling1D(pool_size=(2)))
+model.add(Flatten())
+model.add(Dense(64, activation="relu"))
+model.add(Dropout(0.5))
+model.add(Dense(4, activation="softmax"))
+
+model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])
+checkpointer = callbacks.ModelCheckpoint(filepath="logs/cnnlayer/checkpoint-{epoch:02d}.hdf5", verbose=1, save_best_only=True, monitor='loss')
+model.fit(X_train, y_train, batch_size=batch_size, nb_epoch=50, callbacks=[checkpointer])
+model.save("logs/cnnlayer/lstm1layer_model.hdf5") 

heartsound/dataseta/ml/trainandtest/lstm.py

@@ -0,0 +1,86 @@
+from __future__ import print_function
+import numpy as np
+np.random.seed(1337)  # for reproducibility
+
+from keras.preprocessing import sequence
+from keras.models import Sequential
+from keras.layers import Dense, Dropout, Activation, Lambda
+from keras.layers import Embedding
+from keras.layers import Convolution1D,MaxPooling1D, Flatten
+from keras.datasets import imdb
+from keras import backend as K
+from sklearn.cross_validation import train_test_split
+import pandas as pd
+from keras.utils.np_utils import to_categorical
+
+from sklearn.preprocessing import Normalizer
+from keras.models import Sequential
+from keras.layers import Convolution1D, Dense, Dropout, Flatten, MaxPooling1D
+from keras.utils import np_utils
+import numpy as np
+import h5py
+from keras import callbacks
+from keras.layers import LSTM, GRU, SimpleRNN
+from keras.callbacks import CSVLogger
+from keras.callbacks import ModelCheckpoint, EarlyStopping, ReduceLROnPlateau, CSVLogger
+import csv
+from sklearn.cross_validation import StratifiedKFold
+from sklearn.cross_validation import cross_val_score
+from keras.wrappers.scikit_learn import KerasClassifier
+
+with open('a/traindata.csv', 'rb') as f:
+    reader = csv.reader(f)
+    your_list = list(reader)
+
+trainX = np.array(your_list)
+
+traindata = pd.read_csv('a/trainlabels.csv', header=None)
+Y = traindata.iloc[:,0]
+y_train1 = np.array(Y)
+y_train= to_categorical(y_train1)
+
+maxlen = 2000
+trainX = sequence.pad_sequences(trainX, maxlen=maxlen)
+
+# reshape input to be [samples, time steps, features]
+X_train = np.reshape(trainX, (trainX.shape[0], 1, trainX.shape[1]))
+
+
+
+with open('a/testdata.csv', 'rb') as f:
+    reader1 = csv.reader(f)
+    your_list1 = list(reader1)
+
+testX = np.array(your_list1)
+
+testdata = pd.read_csv('a/testlabels.csv', header=None)
+Y1 = testdata.iloc[:,0]
+y_test1 = np.array(Y1)
+y_test= to_categorical(y_test1)
+
+maxlen = 2000
+testX = sequence.pad_sequences(testX, maxlen=maxlen)
+
+# reshape input to be [samples, time steps, features]
+X_test = np.reshape(testX, (testX.shape[0], 1, testX.shape[1]))
+
+
+
+batch_size = 5
+
+model = Sequential()
+model.add(LSTM(256,input_dim=2000,return_sequences=True)) 
+model.add(Dropout(0.2))
+model.add(LSTM(256, return_sequences=True))
+model.add(Dropout(0.2))
+model.add(LSTM(256, return_sequences=True))
+model.add(Dropout(0.2))
+model.add(LSTM(256, return_sequences=False))
+model.add(Dropout(0.2))
+model.add(Dense(4))
+model.add(Activation('softmax'))
+
+model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])
+checkpointer = callbacks.ModelCheckpoint(filepath="logs/lstm4layer/checkpoint-{epoch:02d}.hdf5", verbose=1, save_best_only=True, monitor='val_acc', mode='max')
+model.fit(X_train, y_train, batch_size=batch_size, nb_epoch=500, validation_data=(X_test, y_test),callbacks=[checkpointer])
+model.save("logs/lstm4layer/lstm1layer_model.hdf5") 

heartsound/dataseta/ml/trainandtest/lstm1.py

@@ -0,0 +1,86 @@
+from __future__ import print_function
+import numpy as np
+np.random.seed(1337)  # for reproducibility
+
+from keras.preprocessing import sequence
+from keras.models import Sequential
+from keras.layers import Dense, Dropout, Activation, Lambda
+from keras.layers import Embedding
+from keras.layers import Convolution1D,MaxPooling1D, Flatten
+from keras.datasets import imdb
+from keras import backend as K
+from sklearn.cross_validation import train_test_split
+import pandas as pd
+from keras.utils.np_utils import to_categorical
+
+from sklearn.preprocessing import Normalizer
+from keras.models import Sequential
+from keras.layers import Convolution1D, Dense, Dropout, Flatten, MaxPooling1D
+from keras.utils import np_utils
+import numpy as np
+import h5py
+from keras import callbacks
+from keras.layers import LSTM, GRU, SimpleRNN
+from keras.callbacks import CSVLogger
+from keras.callbacks import ModelCheckpoint, EarlyStopping, ReduceLROnPlateau, CSVLogger
+import csv
+from sklearn.cross_validation import StratifiedKFold
+from sklearn.cross_validation import cross_val_score
+from keras.wrappers.scikit_learn import KerasClassifier
+
+with open('a/testdata.csv', 'rb') as f:
+    reader = csv.reader(f)
+    your_list = list(reader)
+
+trainX = np.array(your_list)
+
+traindata = pd.read_csv('a/testlabels.csv', header=None)
+Y = traindata.iloc[:,0]
+y_train1 = np.array(Y)
+y_train= to_categorical(y_train1)
+
+maxlen = 44100
+trainX = sequence.pad_sequences(trainX, maxlen=maxlen)
+
+# reshape input to be [samples, time steps, features]
+X_train = np.reshape(trainX, (trainX.shape[0], 1, trainX.shape[1]))
+
+
+
+with open('a/testdata.csv', 'rb') as f:
+    reader1 = csv.reader(f)
+    your_list1 = list(reader1)
+
+testX = np.array(your_list1)
+
+testdata = pd.read_csv('a/testlabels.csv', header=None)
+Y1 = testdata.iloc[:,0]
+y_test1 = np.array(Y1)
+y_test= to_categorical(y_test1)
+
+maxlen = 44100
+testX = sequence.pad_sequences(testX, maxlen=maxlen)
+
+# reshape input to be [samples, time steps, features]
+X_test = np.reshape(testX, (testX.shape[0], 1, testX.shape[1]))
+
+
+
+batch_size = 2
+
+model = Sequential()
+model.add(LSTM(32,input_dim=44100,return_sequences=True)) 
+model.add(Dropout(0.1))
+model.add(LSTM(32, return_sequences=True))
+model.add(Dropout(0.1))
+model.add(LSTM(512, return_sequences=False))
+model.add(Dropout(0.1))
+#model.add(LSTM(512, return_sequences=False))
+#model.add(Dropout(0.1))
+model.add(Dense(4))
+model.add(Activation('softmax'))
+
+model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])
+checkpointer = callbacks.ModelCheckpoint(filepath="logs/lstm3layer/checkpoint-{epoch:02d}.hdf5", verbose=1, save_best_only=True, monitor='loss')
+model.fit(X_train, y_train, batch_size=batch_size, validation_data=(X_test,y_test),nb_epoch=500, callbacks=[checkpointer])
+model.save("logs/lstm3layer/lstm1layer_model.hdf5")