Add pipeline

[heavengate]

test code

import paddle.fluid as fluid

import sys
import time

data_root = '/paddle/paddle/val'

def imagenet_pipeline():
    image, label = paddle.vision.reader.file_label_reader(data_root, batch_size=128, shuffle=True, drop_last=False)

    def decode(image):
        image = paddle.vision.ops.image_decode_random_crop(image, num_threads=4, mode='rgb')
        return image
    def resize(image):
        image = paddle.vision.ops.image_resize(image, size=224)
        return image
    def flip_normalize(image):
        mirror = paddle.vision.ops.random_flip(image)
        image = paddle.vision.ops.mirror_normalize(image, mirror)
        return image
    def preprocess(image):
        image = paddle.vision.ops.image_resize(image, size=224)
        mirror = paddle.vision.ops.random_flip(image)
        image = paddle.vision.ops.mirror_normalize(image, mirror)
        return image

    image = paddle.io.map(decode, image)
    image = paddle.io.map(resize, image)
    image = paddle.io.map(flip_normalize, image)
    # image = paddle.io.map(preprocess, image)

    return {'image': image, 'label': label}

# run in dygraph mode
# for i in range(10):
    # print(imagenet_pipeline())

# run in static mode
dataloader = paddle.io.DataLoader.from_pipeline(imagenet_pipeline)
# for _ in range(100):
for i, data in enumerate(dataloader):
    # print(i, data['image'].shape, data['image'].place, data['label'].shape, data['label'].place)
    # if i > 120:
    #     # print(data['image'], data['label'])
    #     break
    # if i % 10 == 0:
    #     print(i, data['image'].shape, data['image'].place, data['label'])
    #     sys.stdout.flush()
    if i == 20:
        tic = time.time()
    if i == 120:
        toc = time.time()
        print("cost per step: {}ms".format((toc - tic) * 10.))
        sys.stdout.flush()
        break