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

[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