PS1 classification issues about supernnova HOT 6 CLOSED

It was a pippin issue! SNN looks good now!

from supernnova.

ok I lied there's still something weird

from supernnova.

we're fitting about 50% of the PS1-spectroscopically confirmed SNIa as CC. Pretty similar for SDSS. Is this normal?

from supernnova.

Newest update: Using the --plot_file command for run.py appears to prevent SNN from generating any plots.

We were working in /scratch/midway2/rkessler/PIPPIN_OUTPUT/PANOPTICON-DATA/3_CLAS/SNNTEST_PS_DATAPS1_SNNTEST_PS_PS1+MVCC/
previously, and ran job_plot.slurm around 20:00 EST.

The location that these should have been plotted in,
/scratch/midway2/rkessler/PIPPIN_OUTPUT/PANOPTICON-DATA/3_CLAS/SNNTEST_PS_DATAPS1_SNNTEST_PS_PS1+MVCC/dump/lightcurves/SNNTEST_PS_PS1+MVCC/early_prediction/

only contains files that were created 6 hours before, at 14:00 EST!

There's no error message in /scratch/midway2/rkessler/PIPPIN_OUTPUT/PANOPTICON-DATA/3_CLAS/SNNTEST_PS_DATAPS1_SNNTEST_PS_PS1+MVCC/output.log

So I'm not sure what's happening!

from supernnova.

Having a similar but separate issue for the spec sample. We couldn't get the plot_file working, as above, so we tried to run SNN on the spec sample as analogue to see the plots. Unfortunately, we're getting a strange error when running

/scratch/midway2/rkessler/PIPPIN_OUTPUT/MV_PS1_SDSS/3_CLAS/SNNTEST_PS_DATAPS1_SNNTEST_PS_PS1+MVCC/job_plot.slurm

which is detailed in /scratch/midway2/rkessler/PIPPIN_OUTPUT/MV_PS1_SDSS/3_CLAS/SNNTEST_PS_DATAPS1_SNNTEST_PS_PS1+MVCC/output.log -

^[[32m[Computing splits] ^[[0m
List files []

^[[32m[Data processing] ^[[0m<0s
Traceback (most recent call last):
File "run.py", line 210, in
raise e
File "run.py", line 44, in
make_dataset.make_dataset(settings)
File "/project2/rkessler/PRODUCTS/miniconda/envs/snn_gpu/lib/python3.6/contextlib.py", line 52, in inner
return func(*args, **kwds)
File "/project2/rkessler/PRODUCTS/classifiers/supernnova/supernnova/data/make_dataset.py", line 814, in make_dataset
build_traintestval_splits(settings)
File "/project2/rkessler/PRODUCTS/classifiers/supernnova/supernnova/data/make_dataset.py", line 96, in build_traintestval_splits
df_photo = pd.concat(list_df)
File "/project2/rkessler/PRODUCTS/miniconda/envs/snn_gpu/lib/python3.6/site-packages/pandas/core/reshape/concat.py", line 225, in concat
copy=copy, sort=sort)
File "/project2/rkessler/PRODUCTS/miniconda/envs/snn_gpu/lib/python3.6/site-packages/pandas/core/reshape/concat.py", line 259, in init
raise ValueError('No objects to concatenate')
ValueError: No objects to concatenate

This seems to be arising from the photo_window_files, but I haven't been able to figure out what the problem is

from supernnova.

.txt format is not supported with SuperNNova, only .FITS (SNANA format) or .csv (Plasticc format)

from supernnova.