import numpy as np
import matplotlib.pyplot as plt
import tensorflow as tf
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import DenseWARNING:tensorflow:From C:\Users\vrajs\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\keras\src\losses.py:2976: The name tf.losses.sparse_softmax_cross_entropy is deprecated. Please use tf.compat.v1.losses.sparse_softmax_cross_entropy instead.
rng = np.random.default_rng(2)
X = rng.random(400).reshape(-1, 2)
X[:, 1] = X[:, 1] * 4 + 11.5
X[:, 0] = X[:, 0] * (285-150) + 150
Y = np.zeros(len(X))
i = 0
for t, d in X:
y = -3/(260-175)*t + 21
if (t > 175 and t < 260 and d > 12 and d < 15 and d <= y):
Y[i] = 1
else:
Y[i] = 0
i += 1
plt.scatter(X[Y == 1, 0], X[Y == 1, 1], s=70, marker='x',
linewidth=3, c='red', label="Good")
plt.scatter(X[Y == 0, 0], X[Y == 0, 1], s=50, marker='o', label="Bad")
tr = np.linspace(175, 260, 50)
plt.plot(tr, (-3 / 85) * tr + 21, linewidth=1)
plt.axhline(y=12, linewidth=1)
plt.axvline(x=175, linewidth=1)
plt.xlabel("Feature 1", size=12)
plt.ylabel("Feature 2", size=12)
plt.legend(loc='upper right')
plt.show()
Y = Y.reshape(-1, 1)
norm_l = tf.keras.layers.Normalization(axis=-1)
norm_l.adapt(X)
Xn = norm_l(X)
Xt = np.tile(Xn, (1000, 1))
Yt = np.tile(Y, (1000, 1))WARNING:tensorflow:From C:\Users\vrajs\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\keras\src\backend.py:873: The name tf.get_default_graph is deprecated. Please use tf.compat.v1.get_default_graph instead.
model = Sequential(
[
tf.keras.Input(shape=(2,)),
Dense(3, activation='sigmoid', name='layer1'),
Dense(1, activation='sigmoid', name='layer2')
]
)
model.summary()Model: "sequential"
_________________________________________________________________
Layer (type) Output Shape Param #
=================================================================
layer1 (Dense) (None, 3) 9
layer2 (Dense) (None, 1) 4
=================================================================
Total params: 13 (52.00 Byte)
Trainable params: 13 (52.00 Byte)
Non-trainable params: 0 (0.00 Byte)
_________________________________________________________________W1, b1 = model.get_layer("layer1").get_weights() # Random Generation
W2, b2 = model.get_layer("layer2").get_weights() # Random Generation
print("W1:\n", W1, "\nb1:", b1)
print("W2:\n", W2, "\nb2:", b2)W1:
[[ 1.0034692 -0.5364701 -0.61829925]
[-0.45948058 -0.48300928 -0.2479465 ]]
b1: [0. 0. 0.]
W2:
[[ 0.795092 ]
[ 0.50782764]
[-0.8502825 ]]
b2: [0.]model.compile(
loss=tf.keras.losses.BinaryCrossentropy(),
optimizer=tf.keras.optimizers.Adam(learning_rate=0.01),
)
model.fit(
Xt, Yt,
epochs=10,
)Epoch 1/10
WARNING:tensorflow:From C:\Users\vrajs\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\keras\src\utils\tf_utils.py:492: The name tf.ragged.RaggedTensorValue is deprecated. Please use tf.compat.v1.ragged.RaggedTensorValue instead.
6250/6250 [==============================] - 11s 2ms/step - loss: 0.2032
Epoch 2/10
6250/6250 [==============================] - 10s 2ms/step - loss: 0.0970
Epoch 3/10
6250/6250 [==============================] - 11s 2ms/step - loss: 0.0321
Epoch 4/10
6250/6250 [==============================] - 12s 2ms/step - loss: 0.0198
Epoch 5/10
6250/6250 [==============================] - 10s 2ms/step - loss: 0.0146
Epoch 6/10
6250/6250 [==============================] - 10s 2ms/step - loss: 0.0116
Epoch 7/10
6250/6250 [==============================] - 10s 2ms/step - loss: 0.0095
Epoch 8/10
6250/6250 [==============================] - 11s 2ms/step - loss: 0.0080
Epoch 9/10
6250/6250 [==============================] - 11s 2ms/step - loss: 0.0068
Epoch 10/10
6250/6250 [==============================] - 10s 2ms/step - loss: 0.0059<keras.src.callbacks.History at 0x278f700ac20>W1, b1 = model.get_layer("layer1").get_weights()
W2, b2 = model.get_layer("layer2").get_weights()
print("W1:\n", W1, "\nb1:", b1)
print("W2:\n", W2, "\nb2:", b2)W1:
[[13.708808 -0.04309544 13.338246 ]
[ 0.2935187 10.918727 11.044739 ]]
b1: [14.286802 13.310424 1.352872]
W2:
[[ 27.669281]
[ 27.278173]
[-37.544296]]
b2: [-35.913513]rng = np.random.default_rng(0)
X = rng.random(400).reshape(-1, 2)
X[:, 1] = X[:, 1] * 4 + 11.5
X[:, 0] = X[:, 0] * (285-150) + 150
Y = np.zeros(len(X))
i = 0
for t, d in X:
y = -3/(260-175)*t + 21
if (t > 180 and t < 260 and d > 12 and d < 15 and d <= y):
Y[i] = 1
else:
Y[i] = 0
i += 1
X_testn = norm_l(X)
predictions = model.predict(X_testn)
predictions_thresholded = (predictions > 0.5).astype(int).reshape(-1,)
plt.scatter(X[predictions_thresholded == 1, 0], X[predictions_thresholded == 1, 1], s=70, marker='x',
linewidth=3, c='red', label="Good")
plt.scatter(X[predictions_thresholded == 0, 0],
X[predictions_thresholded == 0, 1], s=50, marker='o', label="Bad")
tr = np.linspace(175, 260, 50)
plt.plot(tr, (-3 / 85) * tr + 21, linewidth=1)
plt.axhline(y=12, linewidth=1)
plt.axvline(x=175, linewidth=1)
plt.xlabel("Feature 1", size=12)
plt.ylabel("Feature 2", size=12)
plt.legend(loc='upper right')
plt.show()7/7 [==============================] - 0s 1ms/step
def my_dense(a_in, W, b):
units = W.shape[1]
a_out = np.zeros(units)
for j in range(units):
w = W[:, j]
z = np.dot(w, a_in) + b[j]
a_out[j] = 1/(1+np.exp(-z))
return (a_out)
def my_dense(a_in, W, b):
a_out=1/(1+(np.exp(-(np.matmul(a_in,W)+b))))
return (a_out)def my_sequential(x, W1, b1, W2, b2):
a1 = my_dense(x, W1, b1)
a2 = my_dense(a1, W2, b2)
return (a2)W1_tmp = np.array([[-8.93, 0.29, 12.9], [-0.1, -7.32, 10.81]])
b1_tmp = np.array([-9.82, -9.28, 0.96])
W2_tmp = np.array([[-31.18], [-27.59], [-32.56]])
b2_tmp = np.array([15.41])def my_predict(X, W1, b1, W2, b2):
m = X.shape[0]
p = np.zeros((m, 1))
for i in range(m):
p[i, 0] = my_sequential(X[i], W1, b1, W2, b2)
return (p)rng = np.random.default_rng(0)
X = rng.random(400).reshape(-1, 2)
X[:, 1] = X[:, 1] * 4 + 11.5
X[:, 0] = X[:, 0] * (285-150) + 150
Y = np.zeros(len(X))
i = 0
for t, d in X:
y = -3/(260-175)*t + 21
if (t > 180 and t < 260 and d > 12 and d < 15 and d <= y):
Y[i] = 1
else:
Y[i] = 0
i += 1
X_tstn = norm_l(X)
predictions = my_predict(X_tstn, W1_tmp, b1_tmp, W2_tmp, b2_tmp)
predictions_thresholded = (predictions > 0.5).astype(int).reshape(-1,)
plt.scatter(X[predictions_thresholded == 1, 0], X[predictions_thresholded == 1, 1], s=70, marker='x',
linewidth=3, c='red', label="Good")
plt.scatter(X[predictions_thresholded == 0, 0],
X[predictions_thresholded == 0, 1], s=50, marker='o', label="Bad")
tr = np.linspace(175, 260, 50)
plt.plot(tr, (-3 / 85) * tr + 21, linewidth=1)
plt.axhline(y=12, linewidth=1)
plt.axvline(x=175, linewidth=1)
plt.xlabel("Feature 1", size=12)
plt.ylabel("Feature 2", size=12)
plt.legend(loc='upper right')
plt.show()C:\Users\vrajs\AppData\Local\Temp\ipykernel_14280\3457910700.py:5: DeprecationWarning: Conversion of an array with ndim > 0 to a scalar is deprecated, and will error in future. Ensure you extract a single element from your array before performing this operation. (Deprecated NumPy 1.25.)
p[i, 0] = my_sequential(X[i], W1, b1, W2, b2)