import tensorflow as tf
import numpy as np
import matplotlib.pyplot as plt
import keras
from keras.models import Sequential
from keras.layers.core import Dense
from keras.optimizers import RMSprop
from mpl_toolkits.mplot3d import Axes3D
raw_data = np.genfromtxt('x09.txt', skip_header=36)
xs = np.array(raw_data[:,2], dtype=np.float32)
ys = np.array(raw_data[:,3], dtype=np.float32)
zs = np.array(raw_data[:,4], dtype=np.float32)
x_data = np.array(raw_data[:,2:4], dtype=np.float32)
y_data = np.array(raw_data[:,4], dtype=np.float32)
y_data = y_data.reshape((25,1))
rmsprop = RMSprop(lr=0.01)
model = Sequential()
model.add(Dense(1,input_shape=(2,)))
model.compile(loss='mse',optimizer=rmsprop)
model.summary()
hist = model.fit(x_data, y_data, epochs=1000)
print(hist.history.keys())
print("100Kg 40세 혈중지방함량치=",model.predict(np.array([100,40]).reshape(1,2)))
print("60Kg 25세 혈중지방함량치=",model.predict(np.array([60,25]).reshape(1,2)))
W_, b_ = model.get_weights()
x = np.linspace(20, 100, 50).reshape(50,1)
y = np.linspace(10, 70, 50).reshape(50,1)
X = np.concatenate((x,y), axis=1)
Z = np.matmul(X, W_) + b_
fig = plt.figure(figsize=(12,12))
ax = fig.add_subplot(111, projection='3d')
ax.scatter(x, y, Z)
ax.scatter(xs, ys, zs)
ax.set_xlabel('Weight')
ax.set_ylabel('Age')
ax.set_zlabel('Blood fat')
ax.view_init(15, 15)
plt.show()