Data augmentation

In [ ]:

    

%load_ext autoreload
%autoreload 2

%matplotlib inline


    

In [ ]:

    

#export
from exp.nb_09 import *


    

In [ ]:

    

path = datasets.untar_data(datasets.URLs.IMAGENETTE_160)
tfms = [make_rgb, ResizeFixed(128), to_byte_tensor, to_float_tensor]


    

In [ ]:

    

from nvidia.dali.pipeline import Pipeline


    

In [ ]:

    

import nvidia.dali.ops as ops
import nvidia.dali.types as types


    

In [ ]:

    

bs=8


    

In [ ]:

    

class SimplePipeline(Pipeline):
    def __init__(self, batch_size=8, num_threads=8, device_id=0):
        super(SimplePipeline, self).__init__(batch_size, num_threads, device_id, seed = 12)
        self.input = ops.FileReader(file_root = path/'train')
        # self.input = ops.FileReader(file_root = image_dir, file_list = image_dir + '/file_list.txt')
        self.resize = ops.Resize(device = "cpu", resize_x=128, resize_y=128, image_type = types.RGB,
            interp_type = types.INTERP_LINEAR)
        self.decode = ops.HostDecoder(output_type = types.RGB)

    def define_graph(self):
        jpegs, labels = self.input(name='r')
        images = self.decode(jpegs)
        images = self.resize(images)
        return (images, labels)


    

In [ ]:

    

pipe = SimplePipeline()
pipe.build()


    

In [ ]:

    

pipe_out = pipe.run()
print(pipe_out)


    

In [ ]:

    

images, labels = pipe_out
images.is_dense_tensor(), labels.is_dense_tensor()


    

In [ ]:

    

t = images.as_tensor()


    

In [ ]:

    

from nvidia.dali.plugin.pytorch import DALIGenericIterator,DALIClassificationIterator,feed_ndarray


    

In [ ]:

    

it = DALIGenericIterator(pipe, ['data','label'], pipe.epoch_size('r'))


    

In [ ]:

    

it = DALIClassificationIterator(pipe, pipe.epoch_size('r'))


    

In [ ]:

    

its = iter(it)


    

In [ ]:

    

t = next(it)[0]


    

In [ ]:

    

t['label'].cuda().long().type()


    

In [ ]:

    

t['data'].type()


    

In [ ]:

    

import numpy as np


    

In [ ]:

    

labels_tensor = labels.as_tensor()


    

In [ ]:

    

labels_tensor.shape()


    

In [ ]:

    

np.array(labels_tensor)


    

In [ ]:

    

import matplotlib.gridspec as gridspec


    

In [ ]:

    

def show_images(image_batch):
    columns = 4
    rows = len(image_batch) // (columns)
    fig = plt.figure(figsize = (32,(32 // columns) * rows))
    gs = gridspec.GridSpec(rows, columns)
    for j in range(rows*columns):
        plt.subplot(gs[j])
        plt.axis("off")
        plt.imshow(image_batch.at(j))


    

In [ ]:

    

show_images(images)


    

In [ ]: