TypeError in random.sample(np.array()), word2vec_basic.py with Python3.5

[DSLituiev]

Environment info

Operating System:
MacOS10.10.5 (14F1713)

1. Which pip package you installed.
   pip3 install --upgrade https://storage.googleapis.com/tensorflow/mac/tensorflow-0.7.1-cp35-none-any.whl

   git clone https://github.com/tensorflow/tensorflow
   git checkout r0.7

2. The output from python -c "import tensorflow; print(tensorflow.version)".
   0.7.1

3. numpy.version.full_version

   Out[4]: '1.11.0'

   Installed as described above with pip3, running script from commit 263d00d

Steps to reproduce

  python3 tensorflow/examples/tutorials/word2vec/word2vec_basic.py

What have you tried?

1. This simple example fails:
   random.sample(np.arange(1000), 60)

it looks like it relates to this numpy issue

Logs or other output that would be helpful

(If logs are large, please upload as attachment).

Found and verified text8.zip
Data size 17005207
Most common words (+UNK) [['UNK', 418391], (b'the', 1061396), (b'of', 593677), (b'and', 416629), (b'one', 411764)]
Sample data [5240, 3084, 12, 6, 195, 2, 3135, 46, 59, 156]
3084 -> 5240
b'originated' -> b'anarchism'
3084 -> 12
b'originated' -> b'as'
12 -> 6
b'as' -> b'a'
12 -> 3084
b'as' -> b'originated'
6 -> 12
b'a' -> b'as'
6 -> 195
b'a' -> b'term'
195 -> 2
b'term' -> b'of'
195 -> 6
b'term' -> b'a'
Traceback (most recent call last):
  File "tensorflow/examples/tutorials/word2vec/word2vec_basic.py", line 131, in <module>
    valid_examples = np.array(random.sample(np.arange(valid_window), valid_size))
  File "/Library/Frameworks/Python.framework/Versions/3.5/lib/python3.5/random.py", line 311, in sample
    raise TypeError("Population must be a sequence or set.  For dicts, use list(d).")
TypeError: Population must be a sequence or set.  For dicts, use list(d).

[DSLituiev]

Fix: line 131:
before

 valid_examples = np.array(random.sample(np.arange(valid_window), valid_size))

after

 valid_examples = np.array(random.sample(range(valid_window), valid_size))

[vrv]

Cool, thanks! Do you want to send us a PR?

[DSLituiev]

Yes I will

[vrv]

Thanks!