Added linear

Author: edwardyang12

Summer20/NeuralNetwork/tflinear.py

@@ -0,0 +1,65 @@
+from tensorflow.keras import datasets, models, layers, losses
+import tensorflow as tf
+import numpy as np
+import random
+import matplotlib.pyplot as plt
+
+# randomPoints on the sin graph
+# number - number of points
+# bounds - tuple for range of values
+def randomPoints(number, bounds, func):
+    inputList = []
+    outputList = []
+    while(number>0):
+        value = random.uniform(bounds[0],bounds[1])
+        inputList.append([value])
+        outputList.append([func(value)])
+        number = number - 1
+    return inputList, outputList
+
+def linear(x):
+    return x
+
+def coolLinear(x):
+    return 3*x + 1
+
+# generate training data
+bounds = (-10,10) # represents full system dynamics
+
+inputList, outputList = randomPoints(10000, bounds, coolLinear)
+
+# neural network code
+model = models.Sequential()
+model.add(layers.Dense(1, activation='linear', input_shape=(1,)))
+model.compile(optimizer='Adam',
+                loss=losses.MeanSquaredError(),
+                metrics=['mean_squared_error'])
+
+history = model.fit(np.array(inputList),np.array(outputList), epochs=200)
+print(model.get_weights())
+
+# plots out learning curve
+# plt.plot(history.history['mean_squared_error'], label='mean_squared_error')
+# plt.xlabel('Epoch')
+# plt.ylabel('MSE')
+# plt.ylim([0.0, 0.2])
+# plt.legend(loc='lower right')
+# plt.show()
+
+# generate test data
+inputTest, outputTest = randomPoints(10, bounds, coolLinear)
+print(model.predict(np.array(inputTest)))
+print(outputTest)
+
+# graph of the test, train points
+graph = plt.figure()
+ax = graph.add_subplot(111)
+
+x = np.linspace(-10,10,500)
+y = 3*x + 1
+
+plt.plot(x,y, label= 'y = x', markersize = 2, c='c')
+ax.scatter(inputTest, outputTest, label = 'training', c='b')
+ax.scatter(inputTest,model.predict(np.array(inputTest)), label = 'testing', c='r')
+plt.legend(loc='lower right')
+plt.show()

Summer20/NeuralNetwork/tfsin.py

@@ -48,13 +48,14 @@ def selectPoints(number, angle=90):
 
 # neural network code
 model = models.Sequential()
-model.add(layers.Dense(10, activation='tanh', input_shape=(1,)))
+model.add(layers.Dense(12, activation='tanh', input_shape=(1,)))
 model.add(layers.Dense(1, activation=None))
 model.compile(optimizer='Adam',
                 loss=losses.MeanSquaredError(),
                 metrics=['mean_squared_error'])
 
 history = model.fit(np.array(angleList),np.array(sinList), epochs=200)
+print(model.get_weights())
 
 # plots out learning curve
 # plt.plot(history.history['mean_squared_error'], label='mean_squared_error')