Skip to content

Version of TensorFlow and tflearn to use (NOT WORKING WITH TF 1.0.1 and tflearn 0.3.1) #6

Description

@rahulraju93

Which version of TF and tflearn should I use so that the code runs seamlessly without any errors. I am getting weird errors with the tflearn package itself. Please let me know what version you have used to run this code without any problems. I am also getting the other issue which is still open about serializing the model. What have you done not to get any of these errors. Please help as I want to use this software for an important problem.

tflearn_seq2seq.py:449:


tflearn_seq2seq.py:211: in train
run_id="TFLearnSeq2Seq"
../anaconda/lib/python2.7/site-packages/tflearn/models/dnn.py:216: in fit
callbacks=callbacks)
../anaconda/lib/python2.7/site-packages/tflearn/helpers/trainer.py:339: in fit
show_metric)
../anaconda/lib/python2.7/site-packages/tflearn/helpers/trainer.py:829: in _train
sname, train_summ_str)


tag = 'acc:0', summary_str = '\n\x0b\n\x04Loss\x15\x00\x00\x00\x00\n\x14\n\rAdam/Loss/raw\x15\x86\xbe\x17@\n\x13\n\x0cacc_0/_raw\x1533\xfb=\n\x0c\n\x05acc_0\x15\x00\x00\x00\x00'

def get_value_from_summary_string(tag, summary_str):
    """ get_value_from_summary_string.

    Retrieve a summary value from a summary string.

    Arguments:
        tag: `str`. The summary tag (name).
        summary_str: `str`. The summary string to look in.

    Returns:
        A `float`. The retrieved value.

    Raises:
        `Exception` if tag not found.

    """
    # Fix for TF 0.12
    if tag[-1] == '/':
        tag = tag[:-1]
    summ = summary_pb2.Summary()
    summ.ParseFromString(summary_str)

    for row in summ.value:
        if row.tag.endswith(tag):
            return float(row.simple_value)
  raise ValueError("Tag: " + tag + " cannot be found in summaries list.")

E ValueError: Tag: acc:0 cannot be found in summaries list.

../anaconda/lib/python2.7/site-packages/tflearn/summaries.py:192: ValueError

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions