tf_record_reader_test.py

from PIL import Image
import numpy as np
import sys
import torch
import torch_xla.utils.tf_record_reader as tfrr

a = """
image/class/label       tensor([82])
image/class/synset      n01796340
image/channels  tensor([3])
image/object/bbox/label tensor([], dtype=torch.int64)
image/width     tensor([900])
image/format    JPEG
image/height    tensor([600])
image/class/text        ptarmigan
image/object/bbox/ymin  tensor([])
image/encoded   tensor([ -1, -40,  -1,  ..., -30,  -1, -39], dtype=torch.int8)
image/object/bbox/ymax  tensor([])
image/object/bbox/xmin  tensor([])
image/filename  n01796340_812.JPEG
image/object/bbox/xmax  tensor([])
image/colorspace        RGB
"""


def decode(ex):
  w = ex['image/width'].item()
  h = ex['image/height'].item()
  image = Image.frombytes(ex['image/colorspace'], (w, h),
                          ex['image/encoded'].numpy().tobytes(),
                          ex['image/format'].lower(), 'RGB', None)
  npa = np.asarray(image)
  return torch.from_numpy(npa)


def readem(path):
  transforms = {
      'image/filename': 'STR',
      'image/class/synset': 'STR',
      'image/format': 'STR',
      'image/class/text': 'STR',
      'image/colorspace': 'STR',
  }
  r = tfrr.TfRecordReader(path, compression='', transforms=transforms)
  count = 0
  while True:
    ex = r.read_example()
    if not ex:
      break
    print('\n')
    for lbl, data in ex.items():
      print('{}\t{}'.format(lbl, data))
    decode(ex)
    count += 1
  print('\n\nDecoded {} samples'.format(count))


readem(sys.argv[1])

Great! Thanks for putting it up!

It Works like a charm! I missed the tobytes() things and whole day went just like that! :( Thanks a lot :))