{ "cells": [ { "cell_type": "markdown", "metadata": {}, "source": [ "[![下载Notebook](https://mindspore-website.obs.cn-north-4.myhuaweicloud.com/website-images/master/resource/_static/logo_notebook.svg)](https://mindspore-website.obs.cn-north-4.myhuaweicloud.com/notebook/master/tutorials/zh_cn/advanced/dataset/mindspore_sampler.ipynb) \n", "[![下载样例代码](https://mindspore-website.obs.cn-north-4.myhuaweicloud.com/website-images/master/resource/_static/logo_download_code.svg)](https://mindspore-website.obs.cn-north-4.myhuaweicloud.com/notebook/master/tutorials/zh_cn/advanced/dataset/mindspore_sampler.py) \n", "[![查看源文件](https://mindspore-website.obs.cn-north-4.myhuaweicloud.com/website-images/master/resource/_static/logo_source.svg)](https://gitee.com/mindspore/docs/blob/master/tutorials/source_zh_cn/advanced/dataset/sampler.ipynb)\n", "\n", "# 数据采样\n", "\n", "为满足训练需求,解决诸如数据集过大或样本类别分布不均等问题,MindSpore提供了多种不同用途的采样器(Sampler),帮助用户对数据集进行不同形式的采样。用户只需在加载数据集时传入采样器对象,即可实现数据的采样。\n", "\n", "MindSpore目前提供了如`RandomSampler`、`WeightedRandomSampler`、`SubsetRandomSampler`等多种采样器。此外,用户也可以根据需要实现自定义的采样器类。\n", "\n", "> 更多采样器的使用方法参见[采样器API文档](https://www.mindspore.cn/docs/zh-CN/master/api_python/mindspore.dataset.html#采样器)。\n", "\n", "## 采样器\n", "\n", "下面主要以CIFAR-10数据集为例,介绍几种常用MindSpore采样器的使用方法。\n", "\n", "![cifar10](https://mindspore-website.obs.cn-north-4.myhuaweicloud.com/website-images/master/tutorials/source_zh_cn/advanced/dataset/images/cifar10.jpg)\n", "\n", "> 本章节中的示例代码依赖`matplotlib`,可使用命令`pip install matplotlib`安装。如本文档以Notebook运行时,完成安装后需要重启kernel才能执行后续代码。" ] }, { "cell_type": "code", "execution_count": 1, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Downloading data from https://mindspore-website.obs.cn-north-4.myhuaweicloud.com/notebook/datasets/cifar-10-binary.tar.gz (162.2 MB)\n", "\n", "file_sizes: 100%|████████████████████████████| 170M/170M [00:16<00:00, 10.4MB/s]\n", "Extracting tar.gz file...\n", "Successfully downloaded / unzipped to ./\n" ] } ], "source": [ "from download import download\n", "\n", "url = \"https://mindspore-website.obs.cn-north-4.myhuaweicloud.com/notebook/datasets/cifar-10-binary.tar.gz\"\n", "\n", "path = download(url, \"./\", kind=\"tar.gz\", replace=True)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "解压后数据集文件的目录结构如下:\n", "\n", "```text\n", ".\n", "└── cifar-10-batches-bin\n", " ├── batches.meta.txt\n", " ├── data_batch_1.bin\n", " ├── data_batch_2.bin\n", " ├── data_batch_3.bin\n", " ├── data_batch_4.bin\n", " ├── data_batch_5.bin\n", " ├── readme.html\n", " └── test_batch.bin\n", "```\n", "\n", "### RandomSampler\n", "\n", "从索引序列中随机采样指定数目的数据。\n", "\n", "下面的样例使用随机采样器,分别从数据集中有放回和无放回地随机采样5个数据,并打印展示。为了便于观察有放回与无放回的效果,这里自定义了一个数据量较小的数据集。" ] }, { "cell_type": "code", "execution_count": 2, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "With Replacement: 4 5 6 6 1 \n", "Without Replacement: 4 1 5 6 2 " ] } ], "source": [ "from mindspore.dataset import RandomSampler, NumpySlicesDataset\n", "\n", "np_data = [1, 2, 3, 4, 5, 6, 7, 8] # 数据集\n", "\n", "# 定义有放回采样器,采样5条数据\n", "sampler1 = RandomSampler(replacement=True, num_samples=5)\n", "dataset1 = NumpySlicesDataset(np_data, column_names=[\"data\"], sampler=sampler1)\n", "\n", "print(\"With Replacement: \", end='')\n", "for data in dataset1.create_tuple_iterator(output_numpy=True):\n", " print(data[0], end=' ')\n", "\n", "# 定义无放回采样器,采样5条数据\n", "sampler2 = RandomSampler(replacement=False, num_samples=5)\n", "dataset2 = NumpySlicesDataset(np_data, column_names=[\"data\"], sampler=sampler2)\n", "\n", "print(\"\\nWithout Replacement: \", end='')\n", "for data in dataset2.create_tuple_iterator(output_numpy=True):\n", " print(data[0], end=' ')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "从上面的打印结果可以看出,使用有放回采样器时,同一条数据可能会被多次获取;使用无放回采样器时,同一条数据只能被获取一次。\n", "\n", "### WeightedRandomSampler\n", "\n", "指定长度为N的采样概率列表,按照概率在前N个样本中随机采样指定数目的数据。\n", "\n", "下面的样例使用带权随机采样器从CIFAR-10数据集的前10个样本中按概率获取6个样本,并展示已读取数据的形状和标签。" ] }, { "cell_type": "code", "execution_count": 3, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Image shape: (32, 32, 3) , Label: 6\n", "Image shape: (32, 32, 3) , Label: 6\n", "Image shape: (32, 32, 3) , Label: 9\n", "Image shape: (32, 32, 3) , Label: 9\n", "Image shape: (32, 32, 3) , Label: 9\n", "Image shape: (32, 32, 3) , Label: 6\n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "import math\n", "import matplotlib.pyplot as plt\n", "from mindspore.dataset import WeightedRandomSampler, Cifar10Dataset\n", "%matplotlib inline\n", "\n", "DATA_DIR = \"./cifar-10-batches-bin/\"\n", "\n", "# 指定前10个样本的采样概率并进行采样\n", "weights = [0.8, 0.5, 0, 0, 0, 0, 0, 0, 0, 0]\n", "sampler = WeightedRandomSampler(weights, num_samples=6)\n", "dataset = Cifar10Dataset(DATA_DIR, sampler=sampler) # 加载数据\n", "\n", "def plt_result(dataset, row):\n", " \"\"\"显示采样结果\"\"\"\n", " num = 1\n", " for data in dataset.create_dict_iterator(output_numpy=True):\n", " print(\"Image shape:\", data['image'].shape, \", Label:\", data['label'])\n", " plt.subplot(row, math.ceil(dataset.get_dataset_size() / row), num)\n", " image = data['image']\n", " plt.imshow(image, interpolation=\"None\")\n", " num += 1\n", "\n", "plt_result(dataset, 2)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "从上面的打印结果可以看出,本次在前面一共10个样本中随机采样了6条数据,只有前面两个采样概率不为0的样本才有机会被采样。\n", "\n", "### SubsetRandomSampler\n", "\n", "从指定样本索引子序列中随机采样指定数目的样本数据。\n", "\n", "下面的样例使用子序列随机采样器从CIFAR-10数据集的指定子序列中抽样3个样本,并展示已读取数据的形状和标签。" ] }, { "cell_type": "code", "execution_count": 4, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Image shape: (32, 32, 3) , Label: 9\n", "Image shape: (32, 32, 3) , Label: 1\n", "Image shape: (32, 32, 3) , Label: 4\n", "Image shape: (32, 32, 3) , Label: 9\n", "Image shape: (32, 32, 3) , Label: 6\n", "Image shape: (32, 32, 3) , Label: 1\n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "from mindspore.dataset import SubsetRandomSampler\n", "\n", "# 指定样本索引序列\n", "indices = [0, 1, 2, 3, 4, 5, 6, 7, 8, 9]\n", "sampler = SubsetRandomSampler(indices, num_samples=6)\n", "# 加载数据\n", "dataset = Cifar10Dataset(DATA_DIR, sampler=sampler)\n", "\n", "plt_result(dataset, 2)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "从上面的打印结果可以看到,采样器从索引序列中随机采样了6个样本。\n", "\n", "### PKSampler\n", "\n", "在指定的数据集类别P中,每种类别各采样K条数据。\n", "\n", "下面的样例使用PK采样器从CIFAR-10数据集中每种类别抽样2个样本,最多20个样本,并展示已读取数据的形状和标签。" ] }, { "cell_type": "code", "execution_count": 5, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Image shape: (32, 32, 3) , Label: 0\n", "Image shape: (32, 32, 3) , Label: 0\n", "Image shape: (32, 32, 3) , Label: 1\n", "Image shape: (32, 32, 3) , Label: 1\n", "Image shape: (32, 32, 3) , Label: 2\n", "Image shape: (32, 32, 3) , Label: 2\n", "Image shape: (32, 32, 3) , Label: 3\n", "Image shape: (32, 32, 3) , Label: 3\n", "Image shape: (32, 32, 3) , Label: 4\n", "Image shape: (32, 32, 3) , Label: 4\n" ] }, { "data": { "image/png": "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\n", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "from mindspore.dataset import PKSampler\n", "\n", "# 每种类别抽样2个样本,最多10个样本\n", "sampler = PKSampler(num_val=2, class_column='label', num_samples=10)\n", "dataset = Cifar10Dataset(DATA_DIR, sampler=sampler)\n", "\n", "plt_result(dataset, 3)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "从上面的打印结果可以看出,采样器对数据集中的每种标签都采样了2个样本,一共10个样本。\n", "\n", "### DistributedSampler\n", "\n", "在分布式训练中,对数据集分片进行采样。\n", "\n", "下面的样例使用分布式采样器将构建的数据集分为4片,在分片抽取一个样本,共采样3个样本,并展示已读取的数据。" ] }, { "cell_type": "code", "execution_count": 6, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "{'data': Tensor(shape=[], dtype=Int64, value= 0)}\n", "{'data': Tensor(shape=[], dtype=Int64, value= 4)}\n", "{'data': Tensor(shape=[], dtype=Int64, value= 8)}\n" ] } ], "source": [ "from mindspore.dataset import DistributedSampler\n", "\n", "# 自定义数据集\n", "data_source = [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11]\n", "\n", "# 构建的数据集分为4片,共采样3个数据样本\n", "sampler = DistributedSampler(num_shards=4, shard_id=0, shuffle=False, num_samples=3)\n", "dataset = NumpySlicesDataset(data_source, column_names=[\"data\"], sampler=sampler)\n", "\n", "# 打印数据集\n", "for data in dataset.create_dict_iterator():\n", " print(data)" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "从上面的打印结果可以看出,数据集被分成了4片,每片有3个样本,本次获取的是id为0的片中的样本。\n", "\n", "## 自定义采样器\n", "\n", "用户可以自定义采样器,并把它应用到数据集上。\n", "\n", "### \\_\\_iter\\_\\_ 模式\n", "\n", "用户可以继承`Sampler`基类,通过实现`__iter__`方法来自定义采样器的采样方式。\n", "\n", "下面的样例定义了一个从下标0至下标9间隔为2采样的采样器,将其作用于自定义数据集,并展示已读取数据。" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "a c e g i " ] } ], "source": [ "import mindspore.dataset as ds\n", "\n", "# 自定义采样器\n", "class MySampler(ds.Sampler):\n", " def __iter__(self):\n", " for i in range(0, 10, 2):\n", " yield i\n", "\n", "# 自定义数据集\n", "np_data = ['a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l']\n", "\n", "# 加载数据\n", "dataset = ds.NumpySlicesDataset(np_data, column_names=[\"data\"], sampler=MySampler())\n", "for data in dataset.create_tuple_iterator(output_numpy=True):\n", " print(data[0], end=' ')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "从上面的打印可以看出,自定义的采样器读取了下标为0、2、4、6、8的样本数据,这与自定义采样器的采样目的一致。\n", "\n", "### \\_\\_getitem\\_\\_ 模式\n", "\n", "用户可以定义一个采样器类,该类包含 `__init__` 、 `__getitem__` 和 `__len__` 方法。\n", "\n", "下面的样例定义了一个下标为 `[3, 4, 3, 2, 0, 11, 5, 5, 5, 9, 1, 11, 11, 11, 11, 8]` 的采样器类,将其作用于自定义数据集,并展示已读取数据。" ] }, { "cell_type": "code", "execution_count": 7, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "d e d c a l f f f j b l l l l i " ] } ], "source": [ "import mindspore.dataset as ds\n", "\n", "# 自定义采样器\n", "class MySampler():\n", " def __init__(self):\n", " self.index_ids = [3, 4, 3, 2, 0, 11, 5, 5, 5, 9, 1, 11, 11, 11, 11, 8]\n", " def __getitem__(self, index):\n", " return self.index_ids[index]\n", " def __len__(self):\n", " return len(self.index_ids)\n", "\n", "# 自定义数据集\n", "np_data = ['a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i', 'j', 'k', 'l']\n", "\n", "# 加载数据\n", "dataset = ds.NumpySlicesDataset(np_data, column_names=[\"data\"], sampler=MySampler())\n", "for data in dataset.create_tuple_iterator(output_numpy=True):\n", " print(data[0], end=' ')" ] }, { "cell_type": "markdown", "metadata": {}, "source": [ "从上面的打印可以看出,自定义的采样器读取了下标为 `[3, 4, 3, 2, 0, 11, 5, 5, 5, 9, 1, 11, 11, 11, 11, 8]` 的样本数据,这与自定义采样器的采样目的一致。" ] } ], "metadata": { "kernelspec": { "display_name": "MindSpore", "language": "python", "name": "mindspore" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.7.5" } }, "nbformat": 4, "nbformat_minor": 4 }