Environment setup difficulties
RobertSellers opened this issue · comments
Any docker images / conda documentation for creating a suitable environment? I'm currently trying to hack apart the yaml file, and I've only attained partial success using the following docker image tensorflow/tensorflow:1.12.0-py3 with the training portion. Tips would be appreciated!
Hi,
You can create an Anaconda env using the yml file.
conda env create -f requirements. yml
See details here
Please update your Conda to version 4.8.3 (metapackage 2020.02) to avoid unexpected packages conflicts.
Hi. Thank you. Yes, I have tried the conda envioronment on Linux/Windows machines and with v 4.8.3 and I've been able to run the training tool with success with the PY3GPU environment, but I've had to remove some libs to pip to get this working BUT I'm still unable to run the save_logprob line with your example data.
C:\miniconda\envs\PY3GPU\lib\site-packages\tensorflow\python\util\tf_inspect.py:75: DeprecationWarning: inspect.getargspec() is deprecated since Python 3.0, use inspect.signature() or inspect.getfullargspec() return _inspect.getargspec(target) Traceback (most recent call last): File "geotracknet.py", line 54, in <module> from flags_config import config File "C:\repositories\GeoTrackNet\flags_config.py", line 196, in <module> config.onehot_lat_bins = math.ceil((config.lat_max-config.lat_min)/config.onehot_lat_reso) File "C:\miniconda\envs\PY3GPU\lib\site-packages\tensorflow\python\platform\flags.py", line 84, in __getattr__ wrapped(_sys.argv) File "C:\miniconda\envs\PY3GPU\lib\site-packages\absl\flags\_flagvalues.py", line 633, in __call__ name, value, suggestions=suggestions) absl.flags._exceptions.UnrecognizedFlagError: Unknown command line flag 'test_name'
I just confirmed that the conda requirements load "fine" on an ubuntu 20 anaconda install, but I'm getting the exact same error listed above.
Ok well, it looks like you've just got a syntax error in your documentation. test_name should just be testset_name.
python geotracknet.py
--mode=save_logprob
--dataset_dir=./data
--trainingset_name=ct_2017010203_10_20/ct_2017010203_10_20_train.pkl
--test_name=ct_2017010203_10_20/ct_2017010203_10_20_valid.pkl \
Hi, thank you very much for letting me know. I will update the readme.