Using Jupyter on GPU¶

• If the import of tensorflow fails, then make sure that you have chosen the appropriate jupyter kernel.

Import modules and check versions¶

In [10]:

import sys
import numpy as np
import tensorflow as tf

import sys import numpy as np import tensorflow as tf

In [11]:

print (sys.version)

print (sys.version)

3.10.10 | packaged by conda-forge | (main, Mar 24 2023, 20:08:06) [GCC 11.3.0]

In [12]:

print (tf.__version__)

print (tf.__version__)

2.15.0

Test if Tensorflow can access a GPU¶

In [13]:

tf.config.list_physical_devices('GPU')

tf.config.list_physical_devices('GPU')

Out[13]:

[PhysicalDevice(name='/physical_device:GPU:0', device_type='GPU')]

In [14]:

device_name = tf.test.gpu_device_name()
print (device_name)

device_name = tf.test.gpu_device_name() print (device_name)

/device:GPU:0

2024-02-08 09:33:51.770011: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1929] Created device /device:GPU:0 with 9776 MB memory:  -> device: 0, name: NVIDIA GeForce RTX 2080 Ti, pci bus id: 0000:3b:00.0, compute capability: 7.5

Run a simple example from Tensorflow tutorials¶

The beginner tutorial can be found at: https://www.tensorflow.org/tutorials/quickstart/beginner

Load a dataset¶

In [15]:

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

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

Build a machine learning model¶

In [16]:

model = 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)
model.compile(optimizer='adam', loss=loss_fn, metrics=['accuracy'])

model = 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) model.compile(optimizer='adam', loss=loss_fn, metrics=['accuracy'])

Train your model¶

In [17]:

model.fit(x_train, y_train, epochs=5)

model.fit(x_train, y_train, epochs=5)

Epoch 1/5
1875/1875 [==============================] - 11s 5ms/step - loss: 0.2947 - accuracy: 0.9147
Epoch 2/5
1875/1875 [==============================] - 10s 5ms/step - loss: 0.1429 - accuracy: 0.9578
Epoch 3/5
1875/1875 [==============================] - 10s 5ms/step - loss: 0.1066 - accuracy: 0.9678
Epoch 4/5
1875/1875 [==============================] - 10s 5ms/step - loss: 0.0890 - accuracy: 0.9725
Epoch 5/5
1875/1875 [==============================] - 10s 5ms/step - loss: 0.0748 - accuracy: 0.9766

Out[17]:

<keras.src.callbacks.History at 0x2b2e31914580>

Evaluate your model¶

In [18]:

model.evaluate(x_test, y_test, verbose=2)

model.evaluate(x_test, y_test, verbose=2)

313/313 - 1s - loss: 0.0775 - accuracy: 0.9751 - 1s/epoch - 4ms/step

Out[18]:

[0.07746326178312302, 0.9750999808311462]