I just got this very unexpected exception:
super().fit(df, sampling_temperature=sampling_temperature, **kwargs)
File "/home/johann/software/anaconda/envs/sipfront/lib/python3.11/site-packages/orbit/forecaster/forecaster.py", line 164, in fit
_posterior_samples, training_metrics = estimator.fit(
^^^^^^^^^^^^^^
File "/home/johann/software/anaconda/envs/sipfront/lib/python3.11/site-packages/orbit/estimators/stan_estimator.py", line 143, in fit
stan_mcmc_fit = compiled_mod.sample(
^^^^^^^^^^^^^^^^^^^^
File "/home/johann/software/anaconda/envs/sipfront/lib/python3.11/site-packages/cmdstanpy/model.py", line 1098, in sample
progress_hook("Done", -1) # -1 == all chains finished
^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/johann/software/anaconda/envs/sipfront/lib/python3.11/site-packages/cmdstanpy/model.py", line 2155, in progress_hook
pbar.postfix[0]["value"] = 'Sampling completed'
^^^^^^^^^^^^
AttributeError: 'tqdm_asyncio' object has no attribute 'postfix'
This happens when using
os.environ['TQDM_DISABLE'] = '1'
before importing orbit modules in an attempt to disable those annotying progress bars. As far as I know setting this environment variable is a standard way to disable the output and seems to be the only way as orbit does not have any setting to do it in some other way. So when that environment variable is set, orbit should definitely NOT crash.
I just got this very unexpected exception:
This happens when using
before importing orbit modules in an attempt to disable those annotying progress bars. As far as I know setting this environment variable is a standard way to disable the output and seems to be the only way as orbit does not have any setting to do it in some other way. So when that environment variable is set, orbit should definitely NOT crash.