-
Notifications
You must be signed in to change notification settings - Fork 42
/
tensorflow_example.py
37 lines (27 loc) · 943 Bytes
/
tensorflow_example.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
import tensorflow as tf
from hyperactive import Hyperactive
mnist = tf.keras.datasets.mnist
(x_train, y_train), (x_test, y_test) = mnist.load_data()
x_train, x_test = x_train / 255.0, x_test / 255.0
def cnn(params):
nn = tf.keras.models.Sequential(
[
tf.keras.layers.Flatten(input_shape=(28, 28)),
tf.keras.layers.Dense(128, activation="relu"),
tf.keras.layers.Dropout(0.2),
tf.keras.layers.Dense(10),
]
)
loss_fn = tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True)
nn.compile(optimizer="adam", loss=loss_fn, metrics=["accuracy"])
nn.fit(x_train, y_train, epochs=5)
_, score = nn.evaluate(x=x_test, y=y_test)
return score
search_space = {
"filters_0": [16, 32, 64],
"filters_1": [16, 32, 64],
"dense_0": list(range(100, 2000, 100)),
}
hyper = Hyperactive()
hyper.add_search(cnn, search_space, n_iter=5)
hyper.run()