MindSpore Data Format Conversion

Ascend GPU CPU Data Preparation

View Source On Gitee

Overview

You can convert non-standard datasets and common datasets into the MindSpore data format (that is, MindRecord) to easily load the datasets to MindSpore for training. In addition, the performance of MindSpore is optimized in some scenarios. You can use MindRecord to achieve better performance.

Converting Non-Standard Dataset to MindRecord

The following describes how to convert CV and NLP data into MindRecord and read the MindRecord file through MindDataset.

Converting CV Dataset

This example describes how to convert a CV dataset into MindRecord and use MindDataset to load the dataset.

Create a MindRecord file containing 100 records, whose samples include the file_name (string), label (integer), and data (binary) fields. Use MindDataset to read the MindRecord file.

  1. Import related modules.

    from io import BytesIO
import os
import mindspore.dataset as ds
from mindspore.mindrecord import FileWriter
import mindspore.dataset.vision.c_transforms as vision
from PIL import Image

  2. Generate 100 images and convert them to MindRecord.

    MINDRECORD_FILE = "test.mindrecord"

if os.path.exists(MINDRECORD_FILE):
    os.remove(MINDRECORD_FILE)
    os.remove(MINDRECORD_FILE + ".db")

writer = FileWriter(file_name=MINDRECORD_FILE, shard_num=1)

cv_schema = {"file_name": {"type": "string"}, "label": {"type": "int32"}, "data": {"type": "bytes"}}
writer.add_schema(cv_schema, "it is a cv dataset")

writer.add_index(["file_name", "label"])

data = []
for i in range(100):
    i += 1

    sample = {}
    white_io = BytesIO()
    Image.new('RGB', (i*10, i*10), (255, 255, 255)).save(white_io, 'JPEG')
    image_bytes = white_io.getvalue()
    sample['file_name'] = str(i) + ".jpg"
    sample['label'] = i
    sample['data'] = white_io.getvalue()

    data.append(sample)
    if i % 10 == 0:
        writer.write_raw_data(data)
        data = []

if data:
    writer.write_raw_data(data)

writer.commit()

    MSRStatus.SUCCESS

    Parameter description:

    • MINDRECORD_FILE: path of the output MindRecord file.

  3. Read MindRecord using MindDataset.

    data_set = ds.MindDataset(dataset_file=MINDRECORD_FILE)
decode_op = vision.Decode()
data_set = data_set.map(operations=decode_op, input_columns=["data"], num_parallel_workers=2)
count = 0
for item in data_set.create_dict_iterator(output_numpy=True):
    count += 1
print("Got {} samples".format(count))

    Got 100 samples

Converting NLP Dataset

This example describes how to convert an NLP dataset into MindRecord and use MindDataset to load the dataset. The process of converting the text into the lexicographic order is omitted in this example.

Create a MindRecord file containing 100 records, whose samples include eight fields of the integer type. Use MindDataset to read the MindRecord file.

  1. Import related modules.

    import os
import numpy as np
import mindspore.dataset as ds
from mindspore.mindrecord import FileWriter

  2. Generate 100 text samples and convert them to MindRecord.

    MINDRECORD_FILE = "test.mindrecord"

if os.path.exists(MINDRECORD_FILE):
    os.remove(MINDRECORD_FILE)
    os.remove(MINDRECORD_FILE + ".db")

writer = FileWriter(file_name=MINDRECORD_FILE, shard_num=1)

nlp_schema = {"source_sos_ids": {"type": "int64", "shape": [-1]},
            "source_sos_mask": {"type": "int64", "shape": [-1]},
            "source_eos_ids": {"type": "int64", "shape": [-1]},
            "source_eos_mask": {"type": "int64", "shape": [-1]},
            "target_sos_ids": {"type": "int64", "shape": [-1]},
            "target_sos_mask": {"type": "int64", "shape": [-1]},
            "target_eos_ids": {"type": "int64", "shape": [-1]},
            "target_eos_mask": {"type": "int64", "shape": [-1]}}
writer.add_schema(nlp_schema, "it is a preprocessed nlp dataset")

data = []
for i in range(100):
    i += 1

    sample = {"source_sos_ids": np.array([i, i + 1, i + 2, i + 3, i + 4], dtype=np.int64),
            "source_sos_mask": np.array([i * 1, i * 2, i * 3, i * 4, i * 5, i * 6, i * 7], dtype=np.int64),
            "source_eos_ids": np.array([i + 5, i + 6, i + 7, i + 8, i + 9, i + 10], dtype=np.int64),
            "source_eos_mask": np.array([19, 20, 21, 22, 23, 24, 25, 26, 27], dtype=np.int64),
            "target_sos_ids": np.array([28, 29, 30, 31, 32], dtype=np.int64),
            "target_sos_mask": np.array([33, 34, 35, 36, 37, 38], dtype=np.int64),
            "target_eos_ids": np.array([39, 40, 41, 42, 43, 44, 45, 46, 47], dtype=np.int64),
            "target_eos_mask": np.array([48, 49, 50, 51], dtype=np.int64)}

    data.append(sample)
    if i % 10 == 0:
        writer.write_raw_data(data)
        data = []

if data:
    writer.write_raw_data(data)

writer.commit()

    MSRStatus.SUCCESS

    Parameter description:

    • MINDRECORD_FILE: path of the output MindRecord file.

  3. Read MindRecord using MindDataset.

    data_set = ds.MindDataset(dataset_file=MINDRECORD_FILE)
count = 0
for item in data_set.create_dict_iterator():
    count += 1
print("Got {} samples".format(count))

    Got 100 samples

Converting Common Dataset to MindRecord

MindSpore provides tool classes for converting common datasets to MindRecord. The following table lists part of common datasets and their corresponding tool classes.

Dataset

Tool Class

CIFAR-10

Cifar10ToMR

ImageNet

ImageNetToMR

TFRecord

TFRecordToMR

CSV File

CsvToMR

For details about dataset conversion, see MindSpore API.

Converting the CIFAR-10 Dataset

You can use the Cifar10ToMR class to convert the original CIFAR-10 data to MindRecord and use MindDataset to load the data.

  1. Download and decompress the CIFAR-10 dataset. Execute the following command in jupyter notebook:

    wget -N https://mindspore-website.obs.cn-north-4.myhuaweicloud.com/notebook/datasets/cifar-10-python.tar.gz --no-check-certificate
mkdir -p datasets
tar -xzf cifar-10-python.tar.gz -C datasets
tree ./datasets/cifar-10-batches-py

    ./datasets/cifar-10-batches-py
├── batches.meta
├── data_batch_1
├── data_batch_2
├── data_batch_3
├── data_batch_4
├── data_batch_5
├── readme.html
└── test_batch

0 directories, 8 files

  2. Import related modules.

    import os
import mindspore.dataset as ds
import mindspore.dataset.vision.c_transforms as vision
from mindspore.mindrecord import Cifar10ToMR

  3. Create the Cifar10ToMR object and call the transform API to convert the CIFAR-10 dataset to MindRecord.

    ds_target_path = "./datasets/mindspore_dataset_conversion/"
# clean old run files
os.system("rm -f {}*".format(ds_target_path))
os.system("mkdir -p {}".format(ds_target_path))

CIFAR10_DIR = "./datasets/cifar-10-batches-py"
MINDRECORD_FILE = "./datasets/mindspore_dataset_conversion/cifar10.mindrecord"
cifar10_transformer = Cifar10ToMR(CIFAR10_DIR, MINDRECORD_FILE)
cifar10_transformer.transform(['label'])

    MSRStatus.SUCCESS

    Parameter description:

    • CIFAR10_DIR: path of the CIFAR-10 dataset folder.

    • MINDRECORD_FILE: path of the output MindRecord file.

  4. Read MindRecord using MindDataset.

    data_set = ds.MindDataset(dataset_file=MINDRECORD_FILE)
decode_op = vision.Decode()
data_set = data_set.map(operations=decode_op, input_columns=["data"], num_parallel_workers=2)
count = 0
for item in data_set.create_dict_iterator(output_numpy=True):
    count += 1
print("Got {} samples".format(count))

    Got 50000 samples

Converting the ImageNet Dataset

You can use the ImageNetToMR class to convert the original ImageNet data (images and annotations) to MindRecord and use MindDataset to load the data.

  1. Download the ImageNet dataset, save all images in the images/ folder, and use a mapping file labels_map.txt to record the mapping between images and labels. The mapping file contains two columns, which are the directory and label ID of each type of images. The two columns are separated by spaces. The following is an example of the mapping file:

    n01440760 0
n01443537 1
n01484850 2
n01491361 3
n01494475 4
n01496331 5

    The file directory structure is as follows:

    ├─ labels_map.txt
└─ images
    └─ ......

  2. Import the dataset conversion tool class ImageNetToMR.

    import mindspore.dataset as ds
import mindspore.dataset.vision.c_transforms as vision
from mindspore.mindrecord import ImageNetToMR

  3. Create the ImageNetToMR object and call the transform API to convert the dataset to MindRecord.

    IMAGENET_MAP_FILE = "./labels_map.txt"
IMAGENET_IMAGE_DIR = "./images"
MINDRECORD_FILE = "./imagenet.mindrecord"
imagenet_transformer = ImageNetToMR(IMAGENET_MAP_FILE, IMAGENET_IMAGE_DIR, MINDRECORD_FILE, partition_number=1)
imagenet_transformer.transform()

    Parameter description:

    • IMAGENET_MAP_FILE: path of the label mapping file of the ImageNet dataset.

    • IMAGENET_IMAGE_DIR: path of the folder where all ImageNet images are stored.

    • MINDRECORD_FILE: path of the output MindRecord file.

  4. Read MindRecord using MindDataset.

    data_set = ds.MindDataset(dataset_file=MINDRECORD_FILE)
decode_op = vision.Decode()
data_set = data_set.map(operations=decode_op, input_columns=["image"], num_parallel_workers=2)
count = 0
for item in data_set.create_dict_iterator(output_numpy=True):
    count += 1
print("Got {} samples".format(count))

Converting CSV Dataset

Create a CSV file containing 5 records, convert the CSV file to MindRecord using the CsvToMR tool class, and then read the MindRecord file using MindDataset.

  1. Import related modules.

    import csv
import os
import mindspore.dataset as ds
from mindspore.mindrecord import CsvToMR

  2. Generate a CSV file and convert it to MindRecord.

    CSV_FILE = "test.csv"
MINDRECORD_FILE = "test.mindrecord"

def generate_csv():
    headers = ["id", "name", "math", "english"]
    rows = [(1, "Lily", 78.5, 90),
            (2, "Lucy", 99, 85.2),
            (3, "Mike", 65, 71),
            (4, "Tom", 95, 99),
            (5, "Jeff", 85, 78.5)]
    with open(CSV_FILE, 'w', encoding='utf-8') as f:
        writer = csv.writer(f)
        writer.writerow(headers)
        writer.writerows(rows)

generate_csv()

if os.path.exists(MINDRECORD_FILE):
    os.remove(MINDRECORD_FILE)
    os.remove(MINDRECORD_FILE + ".db")

csv_transformer = CsvToMR(CSV_FILE, MINDRECORD_FILE, partition_number=1)

csv_transformer.transform()

assert os.path.exists(MINDRECORD_FILE)
assert os.path.exists(MINDRECORD_FILE + ".db")

    Parameter description:

    • CSV_FILE: path of the CSV file.

    • MINDRECORD_FILE: path of the output MindRecord file.

  3. Read MindRecord using MindDataset.

    data_set = ds.MindDataset(dataset_file=MINDRECORD_FILE)
count = 0
for item in data_set.create_dict_iterator(output_numpy=True):
    count += 1
print("Got {} samples".format(count))

    Got 5 samples

Converting TFRecord Dataset

Currently, only TensorFlow 1.13.0-rc1 and later versions are supported.

In this part of the example, TensorFlow needs to be installed in advance. If it is not installed, execute the following command to install it. For example, when this document is running as a Notebook, after the installation is complete, you need to restart the kernel to execute the subsequent code.

os.system('pip install tensorflow') if os.system('python -c "import tensorflow"') else print("TensorFlow installed")

0

Use TensorFlow to create a TFRecord file and convert the file to MindRecord using the TFRecordToMR tool class. Read the file using MindDataset and decode the image_bytes field using the Decode operator.

  1. Import related modules.

    import collections
from io import BytesIO
import os
import mindspore.dataset as ds
from mindspore.mindrecord import TFRecordToMR
import mindspore.dataset.vision.c_transforms as vision
from PIL import Image
import tensorflow as tf

  2. Generate a TFRecord file.

    TFRECORD_FILE = "test.tfrecord"
MINDRECORD_FILE = "test.mindrecord"

def generate_tfrecord():
    def create_int_feature(values):
        if isinstance(values, list):
            feature = tf.train.Feature(int64_list=tf.train.Int64List(value=list(values)))
        else:
            feature = tf.train.Feature(int64_list=tf.train.Int64List(value=[values]))
        return feature

    def create_float_feature(values):
        if isinstance(values, list):
            feature = tf.train.Feature(float_list=tf.train.FloatList(value=list(values)))
        else:
            feature = tf.train.Feature(float_list=tf.train.FloatList(value=[values]))
        return feature

    def create_bytes_feature(values):
        if isinstance(values, bytes):
            white_io = BytesIO()
            Image.new('RGB', (10, 10), (255, 255, 255)).save(white_io, 'JPEG')
            image_bytes = white_io.getvalue()
            feature = tf.train.Feature(bytes_list=tf.train.BytesList(value=[image_bytes]))
        else:
            feature = tf.train.Feature(bytes_list=tf.train.BytesList(value=[bytes(values, encoding='utf-8')]))
        return feature

    writer = tf.io.TFRecordWriter(TFRECORD_FILE)

    example_count = 0
    for i in range(10):
        file_name = "000" + str(i) + ".jpg"
        image_bytes = bytes(str("aaaabbbbcccc" + str(i)), encoding="utf-8")
        int64_scalar = i
        float_scalar = float(i)
        int64_list = [i, i+1, i+2, i+3, i+4, i+1234567890]
        float_list = [float(i), float(i+1), float(i+2.8), float(i+3.2),
                    float(i+4.4), float(i+123456.9), float(i+98765432.1)]

        features = collections.OrderedDict()
        features["file_name"] = create_bytes_feature(file_name)
        features["image_bytes"] = create_bytes_feature(image_bytes)
        features["int64_scalar"] = create_int_feature(int64_scalar)
        features["float_scalar"] = create_float_feature(float_scalar)
        features["int64_list"] = create_int_feature(int64_list)
        features["float_list"] = create_float_feature(float_list)

        tf_example = tf.train.Example(features=tf.train.Features(feature=features))
        writer.write(tf_example.SerializeToString())
        example_count += 1
    writer.close()
    print("Write {} rows in tfrecord.".format(example_count))

generate_tfrecord()

    Write 10 rows in tfrecord.

    Parameter description:

    • TFRECORD_FILE: path of the TFRecord file.

    • MINDRECORD_FILE: path of the output MindRecord file.

  3. Convert the TFRecord to MindRecord.

    feature_dict = {"file_name": tf.io.FixedLenFeature([], tf.string),
                "image_bytes": tf.io.FixedLenFeature([], tf.string),
                "int64_scalar": tf.io.FixedLenFeature([], tf.int64),
                "float_scalar": tf.io.FixedLenFeature([], tf.float32),
                "int64_list": tf.io.FixedLenFeature([6], tf.int64),
                "float_list": tf.io.FixedLenFeature([7], tf.float32),
                }

if os.path.exists(MINDRECORD_FILE):
    os.remove(MINDRECORD_FILE)
    os.remove(MINDRECORD_FILE + ".db")

tfrecord_transformer = TFRecordToMR(TFRECORD_FILE, MINDRECORD_FILE, feature_dict, ["image_bytes"])
tfrecord_transformer.transform()

assert os.path.exists(MINDRECORD_FILE)
assert os.path.exists(MINDRECORD_FILE + ".db")

  4. Read MindRecord using MindDataset.

    data_set = ds.MindDataset(dataset_file=MINDRECORD_FILE)
decode_op = vision.Decode()
data_set = data_set.map(operations=decode_op, input_columns=["image_bytes"], num_parallel_workers=2)
count = 0
for item in data_set.create_dict_iterator(output_numpy=True):
    count += 1
print("Got {} samples".format(count))

    Got 10 samples