Clever Geek Handbook

Tf.contrib.learn load_csv_with_header not working in TensorFlow 1.1

I installed the latest version of TensorFlow (v1.1.0) and I tried to run the tf.contrib.learn Quickstart where you intend to generate a classifier for the IRIS dataset. However, when I tried:

training_set = tf.contrib.learn.datasets.base.load_csv_with_header(
    filename=IRIS_TRAINING,
    target_dtype=np.int,
    features_dtype=np.float32)

I got an error StopIteration

.

When I checked the API, I didn't find anything about load_csv_with_header()

. Did they change it in the latest version without updating the tutorial? How can I fix this?

EDIT : I am using Python3.6 if it matters.

+3


source to share


4 answers


This is due to the difference between Python 2 and Python 3. Here's my code below works for Python 3.5:

if not os.path.exists(IRIS_TRAINING):
    raw = urllib.request.urlopen(IRIS_TRAINING_URL).read().decode()
    with open(IRIS_TRAINING, 'w') as f:
        f.write(raw)

if not os.path.exists(IRIS_TEST):
    raw = urllib.request.urlopen(IRIS_TEST_URL).read().decode()
    with open(IRIS_TEST, 'w') as f:
        f.write(raw)

What probably happened is that your code created the filename after IRIS_TRAINING

. But the file is empty. Thus StopIteration is raised

. If you look at the implementation load_csv_with_header

:

with gfile.Open(filename) as csv_file:
    data_file = csv.reader(csv_file)
    header = next(data_file)


StopIteration

thrown when it next

doesn't find any additional items to read as documented https://docs.python.org/3.5/library/exceptions.html#StopIteration

Please note the change in my code from Python 2 version as shown in the Tensorflow tutorial:

  • urllib.request.urlopen

    instead urllib.urlopen

  • decode()

    performed after read()

+6


source


StopIteration

should only happen where the csv file is empty. Have you checked that this path (IRIS_TRAINING) resolves that you have open permission?



+2


source


or you can write the csv file as binary instead of adding decode() 

if not os.path.exists(IRIS_TRAINING):
    raw = urllib.request.urlopen(IRIS_TRAINING_URL).read()
    with open(IRIS_TRAINING, 'wb') as f:
        f.write(raw)
0


source


if the above answer doesn't work, you can specify your path in iris_training.csv and iris_test.csv in urlopen () method.

-1


source






More articles:


All Articles