When I used tensorboard as tensorboard --logdir=logs ,it told me that
See original GitHub issueWhen I used tensorboard as tensorboard --logdir=logs ,it told me that
D:\workspace\python\Human-Detection_by_tiny_YoLo_v3>tensorboard --logdir=logs Traceback (most recent call last): File "d:\software\run\python\lib\runpy.py", line 193, in _run_module_as_main "__main__", mod_spec) File "d:\software\run\python\lib\runpy.py", line 85, in _run_code exec(code, run_globals) File "D:\software\run\python\Scripts\tensorboard.exe\__main__.py", line 5, in <module> File "d:\software\run\python\lib\site-packages\tensorboard\main.py", line 44, in <module> from tensorboard import default File "d:\software\run\python\lib\site-packages\tensorboard\default.py", line 36, in <module> from tensorboard.plugins.beholder import beholder_plugin File "d:\software\run\python\lib\site-packages\tensorboard\plugins\beholder\__init__.py", line 15, in <module> from tensorboard.plugins.beholder.beholder import Beholder File "d:\software\run\python\lib\site-packages\tensorboard\plugins\beholder\beholder.py", line 199, in <module> class BeholderHook(tf.estimator.SessionRunHook): AttributeError: module 'tensorflow.estimator' has no attribute 'SessionRunHook'
- TensorBoard 1.9.0
- tensorflow-gpu1.9.0
- win10
- Python3.6
Issue Analytics
- State:
- Created 4 years ago
- Comments:5 (2 by maintainers)
Top GitHub Comments
Hi @ZZM37! The expression
tf.estimator.SessionRunHook
doesn’t actually appear in TensorBoard 1.9.0 at all—it wastf.train.SessionRunHook
until #1706, which first appeared in TensorBoard 1.13.0.It sounds like you may have multiple versions of TensorBoard installed in the same environment (which won’t work out very well). Could you please run
pip freeze
to list the packages in your environment?The reason could be due to the use of older version of tensorboard and tensorflow. Could you please try with the latest versions of both Tensorflow(1.13.1) and Tensorboard(1.13.1) and see if you are running into the same error.
Also please provide a small code snippet in order to reproduce the error from our end.