tensorflow.python.framework.errors_impl.InternalError: Unable to get element from the feed as bytes when use wide_n_deep_tutorial.py

Question

when I try to use wide_n_deep_tutorial.py provided by tensorflow official website, the tutorial can be run successfully, but after I change the data and corresponding feature, it will show the error below:

File "wide_n_deep_feed.py", line 224, in <module>
tf.app.run(main=main, argv=[sys.argv[0]] + unparsed)
File "/usr/lib/python2.7/site-packages/tensorflow/python/platform/app.py", line 48, in run
    _sys.exit(main(_sys.argv[:1] + flags_passthrough))
  File "wide_n_deep_feed.py", line 185, in main
    FLAGS.train_data, FLAGS.test_data)
  File "wide_n_deep_feed.py", line 166, in train_and_eval
    steps=train_steps)
  File "/usr/lib/python2.7/site-packages/tensorflow/python/estimator/estimator.py", line 241, in train
    loss = self._train_model(input_fn=input_fn, hooks=hooks)
  File "/usr/lib/python2.7/site-packages/tensorflow/python/estimator/estimator.py", line 686, in _train_model
    _, loss = mon_sess.run([estimator_spec.train_op, estimator_spec.loss])
  File "/usr/lib/python2.7/site-packages/tensorflow/python/training/monitored_session.py", line 534, in __exit__
    self._close_internal(exception_type)
  File "/usr/lib/python2.7/site-packages/tensorflow/python/training/monitored_session.py", line 569, in _close_internal
    self._sess.close()
  File "/usr/lib/python2.7/site-packages/tensorflow/python/training/monitored_session.py", line 811, in close
    self._sess.close()
  File "/usr/lib/python2.7/site-packages/tensorflow/python/training/monitored_session.py", line 908, in close
    ignore_live_threads=True)
  File "/usr/lib/python2.7/site-packages/tensorflow/python/training/coordinator.py", line 389, in join
    six.reraise(*self._exc_info_to_raise)
  File "/usr/lib/python2.7/site-packages/tensorflow/python/estimator/inputs/queues/feeding_queue_runner.py", line 94, in _run
    sess.run(enqueue_op, feed_dict=feed_dict)
  File "/usr/lib/python2.7/site-packages/tensorflow/python/client/session.py", line 895, in run
    run_metadata_ptr)
  File "/usr/lib/python2.7/site-packages/tensorflow/python/client/session.py", line 1124, in _run
    feed_dict_tensor, options, run_metadata)
  File "/usr/lib/python2.7/site-packages/tensorflow/python/client/session.py", line 1321, in _do_run
    options, run_metadata)
  File "/usr/lib/python2.7/site-packages/tensorflow/python/client/session.py", line 1340, in _do_call
    raise type(e)(node_def, op, message)
tensorflow.python.framework.errors_impl.InternalError: Unable to get element from the feed as bytes.

PS: there are some files generated in my models directory already: checkpoint, events, graph.pbtxt, model.ckpt-0.data-00000-of-00001,model.ckpt-0.index,model.ckpt-0.meta

Any answers will be appreciated.

Chuong Huynh · Accepted Answer · 2017-09-21 10:25:47Z

3

EDIT:

The error occurred when you use the dataframe with NaN value, so Tensorflow can not convert the columns to tf.Dtype, you can use the line in sample code:

df_data = df_data.dropna(how="any", axis=0)

to remove all NaN samples or you can cast the numeric column and replace all NaN value with empty string like this:

# Cast to numeric column
df_data["numeric_column_title"] = pd.to_numeric(df_data["numeric_column_title"],errors='coerce')
# Replace NaN value with 0
df_data[["numeric_column_title1", "numeric_column_title2"]] = df_data[["numeric_column_title1", "numeric_column_title2"]].fillna(value=0)
# Replace other string columns with empty string
df_data = df_data.fillna(value='')

Hope this help :)

edited Sep 21, 2017 at 10:25

answered Sep 21, 2017 at 4:29

Chuong Huynh

841 silver badge6 bronze badges

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

tensorflow.python.framework.errors_impl.InternalError: Unable to get element from the feed as bytes when use wide_n_deep_tutorial.py

1 Answer 1

Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related