HTTP/2 200
content-type: application/octet-stream
x-guploader-uploadid: ABgVH8-4Gk1XAi_bLVdnMGyiBEWOsLtRuB6ab-odv9B9oTb5B_ErQKIKodujZ7J2_3ivaMu-
expires: Fri, 25 Jul 2025 04:35:24 GMT
date: Fri, 25 Jul 2025 03:35:24 GMT
cache-control: public, max-age=3600
last-modified: Fri, 16 Aug 2024 09:15:34 GMT
etag: "7b1e3777b348a35819896a5e5ab19570"
x-goog-generation: 1723799734021414
x-goog-metageneration: 1
x-goog-stored-content-encoding: identity
x-goog-stored-content-length: 29844
x-goog-hash: crc32c=LO7XNw==
x-goog-hash: md5=ex43d7NIo1gZiWpeWrGVcA==
x-goog-storage-class: MULTI_REGIONAL
accept-ranges: bytes
content-length: 29844
server: UploadServer
alt-svc: h3=":443"; ma=2592000,h3-29=":443"; ma=2592000
{
"cells": [
{
"cell_type": "markdown",
"metadata": {
"id": "iPpI7RaYoZuE"
},
"source": [
"##### Copyright 2018 The TensorFlow Authors."
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {
"cellView": "form",
"execution": {
"iopub.execute_input": "2024-08-16T02:30:56.341518Z",
"iopub.status.busy": "2024-08-16T02:30:56.340976Z",
"iopub.status.idle": "2024-08-16T02:30:56.344669Z",
"shell.execute_reply": "2024-08-16T02:30:56.344124Z"
},
"id": "hro2InpHobKk"
},
"outputs": [],
"source": [
"#@title Licensed under the Apache License, Version 2.0 (the \"License\");\n",
"# you may not use this file except in compliance with the License.\n",
"# You may obtain a copy of the License at\n",
"#\n",
"# https://www.apache.org/licenses/LICENSE-2.0\n",
"#\n",
"# Unless required by applicable law or agreed to in writing, software\n",
"# distributed under the License is distributed on an \"AS IS\" BASIS,\n",
"# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n",
"# See the License for the specific language governing permissions and\n",
"# limitations under the License."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "U9i2Dsh-ziXr"
},
"source": [
"# Customization basics: tensors and operations"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "Hndw-YcxoOJK"
},
"source": [
"
"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "6sILUVbHoSgH"
},
"source": [
"This is an introductory TensorFlow tutorial that shows how to:\n",
"\n",
"* Import the required package.\n",
"* Create and use tensors.\n",
"* Use GPU acceleration.\n",
"* Build a data pipeline with `tf.data.Dataset`."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "z1JcS5iBXMRO"
},
"source": [
"## Import TensorFlow\n",
"\n",
"To get started, import the `tensorflow` module. As of TensorFlow 2, eager execution is turned on by default. Eager execution enables a more interactive frontend to TensorFlow, which you will later explore in more detail."
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {
"execution": {
"iopub.execute_input": "2024-08-16T02:30:56.348575Z",
"iopub.status.busy": "2024-08-16T02:30:56.348049Z",
"iopub.status.idle": "2024-08-16T02:30:58.698400Z",
"shell.execute_reply": "2024-08-16T02:30:58.697666Z"
},
"id": "vjBPmYjLdFmk"
},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"2024-08-16 02:30:56.601166: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:485] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered\n",
"2024-08-16 02:30:56.622608: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:8454] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered\n",
"2024-08-16 02:30:56.629164: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1452] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered\n"
]
}
],
"source": [
"import tensorflow as tf"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "H9UySOPLXdaw"
},
"source": [
"## Tensors\n",
"\n",
"A Tensor is a multi-dimensional array. Similar to NumPy `ndarray` objects, `tf.Tensor` objects have a data type and a shape. Additionally, `tf.Tensor`s can reside in accelerator memory (like a GPU). TensorFlow offers a rich library of operations (for example, `tf.math.add`, `tf.linalg.matmul`, and `tf.linalg.inv`) that consume and produce `tf.Tensor`s. These operations automatically convert built-in Python types. For example:\n"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {
"cellView": "code",
"execution": {
"iopub.execute_input": "2024-08-16T02:30:58.702978Z",
"iopub.status.busy": "2024-08-16T02:30:58.702447Z",
"iopub.status.idle": "2024-08-16T02:31:00.907532Z",
"shell.execute_reply": "2024-08-16T02:31:00.906788Z"
},
"id": "ngUe237Wt48W"
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"tf.Tensor(3, shape=(), dtype=int32)\n",
"tf.Tensor([4 6], shape=(2,), dtype=int32)\n",
"tf.Tensor(25, shape=(), dtype=int32)\n",
"tf.Tensor(6, shape=(), dtype=int32)\n",
"tf.Tensor(13, shape=(), dtype=int32)\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"I0000 00:00:1723775459.220860 70727 cuda_executor.cc:1015] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
"I0000 00:00:1723775459.224736 70727 cuda_executor.cc:1015] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
"I0000 00:00:1723775459.228526 70727 cuda_executor.cc:1015] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
"I0000 00:00:1723775459.231708 70727 cuda_executor.cc:1015] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
"I0000 00:00:1723775459.243472 70727 cuda_executor.cc:1015] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
"I0000 00:00:1723775459.247007 70727 cuda_executor.cc:1015] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
"I0000 00:00:1723775459.250510 70727 cuda_executor.cc:1015] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
"I0000 00:00:1723775459.253505 70727 cuda_executor.cc:1015] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
"I0000 00:00:1723775459.256964 70727 cuda_executor.cc:1015] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
"I0000 00:00:1723775459.260375 70727 cuda_executor.cc:1015] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
"I0000 00:00:1723775459.263839 70727 cuda_executor.cc:1015] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
"I0000 00:00:1723775459.266764 70727 cuda_executor.cc:1015] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
"I0000 00:00:1723775460.484960 70727 cuda_executor.cc:1015] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
"I0000 00:00:1723775460.487122 70727 cuda_executor.cc:1015] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
"I0000 00:00:1723775460.489159 70727 cuda_executor.cc:1015] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
"I0000 00:00:1723775460.491212 70727 cuda_executor.cc:1015] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
"I0000 00:00:1723775460.493195 70727 cuda_executor.cc:1015] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
"I0000 00:00:1723775460.495178 70727 cuda_executor.cc:1015] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
"I0000 00:00:1723775460.497138 70727 cuda_executor.cc:1015] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
"I0000 00:00:1723775460.499106 70727 cuda_executor.cc:1015] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
"I0000 00:00:1723775460.500943 70727 cuda_executor.cc:1015] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
"I0000 00:00:1723775460.502910 70727 cuda_executor.cc:1015] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
"I0000 00:00:1723775460.504850 70727 cuda_executor.cc:1015] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
"I0000 00:00:1723775460.506819 70727 cuda_executor.cc:1015] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
"I0000 00:00:1723775460.545061 70727 cuda_executor.cc:1015] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
"I0000 00:00:1723775460.547158 70727 cuda_executor.cc:1015] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
"I0000 00:00:1723775460.549140 70727 cuda_executor.cc:1015] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
"I0000 00:00:1723775460.551249 70727 cuda_executor.cc:1015] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
"I0000 00:00:1723775460.553080 70727 cuda_executor.cc:1015] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
"I0000 00:00:1723775460.555080 70727 cuda_executor.cc:1015] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
"I0000 00:00:1723775460.557010 70727 cuda_executor.cc:1015] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
"I0000 00:00:1723775460.558972 70727 cuda_executor.cc:1015] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
"I0000 00:00:1723775460.560817 70727 cuda_executor.cc:1015] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
"I0000 00:00:1723775460.563305 70727 cuda_executor.cc:1015] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
"I0000 00:00:1723775460.565638 70727 cuda_executor.cc:1015] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n",
"I0000 00:00:1723775460.568007 70727 cuda_executor.cc:1015] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\n"
]
}
],
"source": [
"print(tf.math.add(1, 2))\n",
"print(tf.math.add([1, 2], [3, 4]))\n",
"print(tf.math.square(5))\n",
"print(tf.math.reduce_sum([1, 2, 3]))\n",
"\n",
"# Operator overloading is also supported\n",
"print(tf.math.square(2) + tf.math.square(3))"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "IDY4WsYRhP81"
},
"source": [
"Each `tf.Tensor` has a shape and a datatype:"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {
"execution": {
"iopub.execute_input": "2024-08-16T02:31:00.911410Z",
"iopub.status.busy": "2024-08-16T02:31:00.911125Z",
"iopub.status.idle": "2024-08-16T02:31:00.915973Z",
"shell.execute_reply": "2024-08-16T02:31:00.915351Z"
},
"id": "srYWH1MdJNG7"
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"tf.Tensor([[2 3]], shape=(1, 2), dtype=int32)\n",
"(1, 2)\n",
"
\n"
]
}
],
"source": [
"x = tf.linalg.matmul([[1]], [[2, 3]])\n",
"print(x)\n",
"print(x.shape)\n",
"print(x.dtype)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "eBPw8e8vrsom"
},
"source": [
"The most obvious differences between NumPy arrays and `tf.Tensor`s are:\n",
"\n",
"1. Tensors can be backed by accelerator memory (like GPU, TPU).\n",
"2. Tensors are immutable."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "Dwi1tdW3JBw6"
},
"source": [
"### NumPy compatibility\n",
"\n",
"Converting between a TensorFlow `tf.Tensor` and a NumPy `ndarray` is easy:\n",
"\n",
"* TensorFlow operations automatically convert NumPy ndarrays to Tensors.\n",
"* NumPy operations automatically convert Tensors to NumPy ndarrays.\n",
"\n",
"Tensors are explicitly converted to NumPy ndarrays using their `.numpy()` method. These conversions are typically cheap since the array and `tf.Tensor` share the underlying memory representation, if possible. However, sharing the underlying representation isn't always possible since the `tf.Tensor` may be hosted in GPU memory while NumPy arrays are always backed by host memory, and the conversion involves a copy from GPU to host memory."
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {
"execution": {
"iopub.execute_input": "2024-08-16T02:31:00.919071Z",
"iopub.status.busy": "2024-08-16T02:31:00.918828Z",
"iopub.status.idle": "2024-08-16T02:31:00.929627Z",
"shell.execute_reply": "2024-08-16T02:31:00.929042Z"
},
"id": "lCUWzso6mbqR"
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"TensorFlow operations convert numpy arrays to Tensors automatically\n",
"tf.Tensor(\n",
"[[42. 42. 42.]\n",
" [42. 42. 42.]\n",
" [42. 42. 42.]], shape=(3, 3), dtype=float64)\n",
"And NumPy operations convert Tensors to NumPy arrays automatically\n",
"[[43. 43. 43.]\n",
" [43. 43. 43.]\n",
" [43. 43. 43.]]\n",
"The .numpy() method explicitly converts a Tensor to a numpy array\n",
"[[42. 42. 42.]\n",
" [42. 42. 42.]\n",
" [42. 42. 42.]]\n"
]
}
],
"source": [
"import numpy as np\n",
"\n",
"ndarray = np.ones([3, 3])\n",
"\n",
"print(\"TensorFlow operations convert numpy arrays to Tensors automatically\")\n",
"tensor = tf.math.multiply(ndarray, 42)\n",
"print(tensor)\n",
"\n",
"\n",
"print(\"And NumPy operations convert Tensors to NumPy arrays automatically\")\n",
"print(np.add(tensor, 1))\n",
"\n",
"print(\"The .numpy() method explicitly converts a Tensor to a numpy array\")\n",
"print(tensor.numpy())"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "PBNP8yTRfu_X"
},
"source": [
"## GPU acceleration\n",
"\n",
"Many TensorFlow operations are accelerated using the GPU for computation. Without any annotations, TensorFlow automatically decides whether to use the GPU or CPU for an operation—copying the tensor between CPU and GPU memory, if necessary. Tensors produced by an operation are typically backed by the memory of the device on which the operation executed. For example:"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {
"cellView": "code",
"execution": {
"iopub.execute_input": "2024-08-16T02:31:00.932518Z",
"iopub.status.busy": "2024-08-16T02:31:00.932299Z",
"iopub.status.idle": "2024-08-16T02:31:00.937748Z",
"shell.execute_reply": "2024-08-16T02:31:00.937174Z"
},
"id": "3Twf_Rw-gQFM"
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Is there a GPU available: \n",
"[PhysicalDevice(name='/physical_device:GPU:0', device_type='GPU'), PhysicalDevice(name='/physical_device:GPU:1', device_type='GPU'), PhysicalDevice(name='/physical_device:GPU:2', device_type='GPU'), PhysicalDevice(name='/physical_device:GPU:3', device_type='GPU')]\n",
"Is the Tensor on GPU #0: \n",
"True\n"
]
}
],
"source": [
"x = tf.random.uniform([3, 3])\n",
"\n",
"print(\"Is there a GPU available: \"),\n",
"print(tf.config.list_physical_devices(\"GPU\"))\n",
"\n",
"print(\"Is the Tensor on GPU #0: \"),\n",
"print(x.device.endswith('GPU:0'))"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "vpgYzgVXW2Ud"
},
"source": [
"### Device names\n",
"\n",
"The `Tensor.device` property provides a fully qualified string name of the device hosting the contents of the tensor. This name encodes many details, such as an identifier of the network address of the host on which this program is executing and the device within that host. This is required for distributed execution of a TensorFlow program. The string ends with `GPU:` if the tensor is placed on the `N`-th GPU on the host."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "ZWZQCimzuqyP"
},
"source": [
"### Explicit device placement\n",
"\n",
"In TensorFlow, *placement* refers to how individual operations are assigned (placed on) a device for execution. As mentioned, when there is no explicit guidance provided, TensorFlow automatically decides which device to execute an operation and copies tensors to that device, if needed.\n",
"\n",
"However, TensorFlow operations can be explicitly placed on specific devices using the `tf.device` context manager. For example:"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {
"execution": {
"iopub.execute_input": "2024-08-16T02:31:00.940981Z",
"iopub.status.busy": "2024-08-16T02:31:00.940608Z",
"iopub.status.idle": "2024-08-16T02:31:01.031612Z",
"shell.execute_reply": "2024-08-16T02:31:01.030949Z"
},
"id": "RjkNZTuauy-Q"
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"On CPU:\n",
"10 loops: 43.71ms\n",
"On GPU:\n",
"10 loops: 38.91ms\n"
]
}
],
"source": [
"import time\n",
"\n",
"def time_matmul(x):\n",
" start = time.time()\n",
" for loop in range(10):\n",
" tf.linalg.matmul(x, x)\n",
"\n",
" result = time.time()-start\n",
"\n",
" print(\"10 loops: {:0.2f}ms\".format(1000*result))\n",
"\n",
"# Force execution on CPU\n",
"print(\"On CPU:\")\n",
"with tf.device(\"CPU:0\"):\n",
" x = tf.random.uniform([1000, 1000])\n",
" assert x.device.endswith(\"CPU:0\")\n",
" time_matmul(x)\n",
"\n",
"# Force execution on GPU #0 if available\n",
"if tf.config.list_physical_devices(\"GPU\"):\n",
" print(\"On GPU:\")\n",
" with tf.device(\"GPU:0\"): # Or GPU:1 for the 2nd GPU, GPU:2 for the 3rd etc.\n",
" x = tf.random.uniform([1000, 1000])\n",
" assert x.device.endswith(\"GPU:0\")\n",
" time_matmul(x)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "o1K4dlhhHtQj"
},
"source": [
"## Datasets\n",
"\n",
"This section uses the `tf.data.Dataset` API to build a pipeline for feeding data to your model. `tf.data.Dataset` is used to build performant, complex input pipelines from simple, re-usable pieces that will feed your model's training or evaluation loops. (Refer to the [tf.data: Build TensorFlow input pipelines](../../guide/data.ipynb) guide to learn more.)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "zI0fmOynH-Ne"
},
"source": [
"### Create a source `Dataset`\n",
"\n",
"Create a *source* dataset using one of the factory functions like `tf.data.Dataset.from_tensors`, `tf.data.Dataset.from_tensor_slices`, or using objects that read from files like `tf.data.TextLineDataset` or `tf.data.TFRecordDataset`. Refer to the _Reading input data_ section of the [tf.data: Build TensorFlow input pipelines](../../guide/data.ipynb) guide for more information."
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {
"execution": {
"iopub.execute_input": "2024-08-16T02:31:01.034880Z",
"iopub.status.busy": "2024-08-16T02:31:01.034641Z",
"iopub.status.idle": "2024-08-16T02:31:01.054294Z",
"shell.execute_reply": "2024-08-16T02:31:01.053633Z"
},
"id": "F04fVOHQIBiG"
},
"outputs": [],
"source": [
"ds_tensors = tf.data.Dataset.from_tensor_slices([1, 2, 3, 4, 5, 6])\n",
"\n",
"# Create a CSV file\n",
"import tempfile\n",
"_, filename = tempfile.mkstemp()\n",
"\n",
"with open(filename, 'w') as f:\n",
" f.write(\"\"\"Line 1\n",
"Line 2\n",
"Line 3\n",
" \"\"\")\n",
"\n",
"ds_file = tf.data.TextLineDataset(filename)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "vbxIhC-5IPdf"
},
"source": [
"### Apply transformations\n",
"\n",
"Use the transformations functions like `tf.data.Dataset.map`, `tf.data.Dataset.batch`, and `tf.data.Dataset.shuffle` to apply transformations to dataset records."
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {
"execution": {
"iopub.execute_input": "2024-08-16T02:31:01.057854Z",
"iopub.status.busy": "2024-08-16T02:31:01.057276Z",
"iopub.status.idle": "2024-08-16T02:31:01.071743Z",
"shell.execute_reply": "2024-08-16T02:31:01.071168Z"
},
"id": "uXSDZWE-ISsd"
},
"outputs": [],
"source": [
"ds_tensors = ds_tensors.map(tf.math.square).shuffle(2).batch(2)\n",
"\n",
"ds_file = ds_file.batch(2)"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "A8X1GNfoIZKJ"
},
"source": [
"### Iterate\n",
"\n",
"`tf.data.Dataset` objects support iteration to loop over records:"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {
"execution": {
"iopub.execute_input": "2024-08-16T02:31:01.075147Z",
"iopub.status.busy": "2024-08-16T02:31:01.074578Z",
"iopub.status.idle": "2024-08-16T02:31:01.126484Z",
"shell.execute_reply": "2024-08-16T02:31:01.125706Z"
},
"id": "ws-WKRk5Ic6-"
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Elements of ds_tensors:\n",
"tf.Tensor([4 1], shape=(2,), dtype=int32)\n",
"tf.Tensor([16 9], shape=(2,), dtype=int32)\n",
"tf.Tensor([25 36], shape=(2,), dtype=int32)\n",
"\n",
"Elements in ds_file:\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"tf.Tensor([b'Line 1' b'Line 2'], shape=(2,), dtype=string)\n",
"tf.Tensor([b'Line 3' b' '], shape=(2,), dtype=string)\n"
]
}
],
"source": [
"print('Elements of ds_tensors:')\n",
"for x in ds_tensors:\n",
" print(x)\n",
"\n",
"print('\\nElements in ds_file:')\n",
"for x in ds_file:\n",
" print(x)"
]
}
],
"metadata": {
"accelerator": "GPU",
"colab": {
"collapsed_sections": [],
"name": "basics.ipynb",
"toc_visible": true
},
"kernelspec": {
"display_name": "Python 3",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.9.19"
}
},
"nbformat": 4,
"nbformat_minor": 0
}
![](\"https://www.tensorflow.org/images/tf_logo_32px.png\"){kind=logo}
View on TensorFlow.org\n",
" ![](\"https://www.tensorflow.org/images/colab_logo_32px.png\"){kind=logo}
Run in Google Colab\n",
" ![](\"https://www.tensorflow.org/images/GitHub-Mark-32px.png\")
View source on GitHub\n",
" ![](\"https://www.tensorflow.org/images/download_logo_32px.png\"){kind=logo}
Download notebook\n",
"