import numpy as np
import pandas as pd
from keras.models import Sequential
from keras.layers import Dense, Dropout
from keras.optimizers import RMSprop
from keras.utils import np_utils, plot_model
from sklearn.preprocessing import quantile_transform
from keras.models import load_model

X_train=pd.read_csv('m.csv', sep=',',header=None).values
X_test =pd.read_table('prom.mat',header=None).iloc[:,1:21].values
X_test[X_test<=0]=1e-5
X_test =np.log2(X_test)
X_test = quantile_transform(X_test, n_quantiles=10, random_state=0)
np.percentile(X_test,[1,50,99])
y_train=np_utils.to_categorical(np.repeat(np.arange(0,16,1),1000))

model=Sequential()

model.add(Dense(100, activation='relu', input_shape=(X_train.shape[1],)))
model.add(Dropout(0.2))
model.add(Dense(50, activation='relu'))
model.add(Dropout(0.2))
model.add(Dense(y_train.shape[1], activation='softmax'))

model.compile(loss="categorical_crossentropy",optimizer=RMSprop(),metrics=["accuracy"])

model.fit(X_train,y_train, epochs=100, batch_size=10, verbose=1)

loss, accuracy = model.evaluate(X_train, y_train, verbose=1)
print("Accuracy = {:.2f}".format(accuracy))
pred = model.predict(X_test)
y_test = np.argmax(pred, axis=1)
y_test[np.where(np.max(pred,axis=1)<0.9)]=-1
np.savetxt("pred.csv", y_test.astype(int), fmt='%i', delimiter=",")
plot_model(model, to_file='model.png', show_shapes=True)
model.save('model.h5')