%matplotlib inline
import pandas as pd
import socket
host = socket.getfqdn()
from core import load, zoom, calc, save,plots,monitor
#reload funcs after updating ./core/*.py
import importlib
importlib.reload(load)
importlib.reload(zoom)
importlib.reload(calc)
importlib.reload(save)
importlib.reload(plots)
importlib.reload(monitor)
<module 'core.monitor' from '/ccc/work/cont003/gen7420/odakatin/monitor-sedna/notebook/core/monitor.py'>
# 'month': = 'JOBID' almost month but not really,
# If you submit the job with job scheduler, above
#below are list of enviroment variable one can pass
#%env local='2"
# local : if True run dask local cluster, if not true, put number of workers
# setted in the 'local'
# if no 'local ' given, local will be setted automatically to 'True'
#%env ychunk='2'
#%env tchunk='2'
# controls chunk. 'False' sets no modification from original netcdf file's chunk.
# ychunk=10 will group the original netcdf file to 10 by 10
# tchunk=1 will chunk the time coordinate one by one
#%env control=FWC_SSH
# name of control file to be used for computation/plots/save/
#%env file_exp=
# 'file_exp': Which 'experiment' name is it?
#. this corresopnds to intake catalog name without path and .yaml
#%env year=
# for Validation, this correspoinds to path/year/month 's year
# for monitoring, this corresponids to 'date' having * means do all files in the monitoring directory
# setting it as *0[0-9] &*1[0-9]& *[2-3][0-9], the job can be separated in three lots.
#%env month=
# for monitoring this corresponds to file path path-XIOS.{month}/
#
#%env save= proceed saving? True or False , Default is setted as True
#%env plot= proceed plotting? True or False , Default is setted as True
#%env calc= proceed computation? or just load computed result? True or False , Default is setted as True
#%env save=False
%%time
# 'savefig': Do we save output in html? or not. keep it true.
savefig=True
client,cluster,control,catalog_url,month,year,daskreport,outputpath = load.set_control(host)
!mkdir -p $outputpath
!mkdir -p $daskreport
client
local True using host= irene5058.c-irene.mg1.tgcc.ccc.cea.fr starting dask cluster on local= True workers 16 10000000000 False not local in tgcc rome local cluster starting This code is running on irene5058.c-irene.mg1.tgcc.ccc.cea.fr using SEDNA_DELTA_MONITOR file experiment, read from ../lib/SEDNA_DELTA_MONITOR.yaml on year= 2012 on month= 02 outputpath= ../results/SEDNA_DELTA_MONITOR/ daskreport= ../results/dask/6312288irene5058.c-irene.mg1.tgcc.ccc.cea.fr_SEDNA_DELTA_MONITOR_02M_Sectiony/ CPU times: user 359 ms, sys: 234 ms, total: 593 ms Wall time: 11 s
Client
|
Cluster
|
df=load.controlfile(control)
#Take out 'later' tagged computations
#df=df[~df['Value'].str.contains('later')]
df
Value | Inputs | Equation | Zone | Plot | Colourmap | MinMax | Unit | Oldname | Unnamed: 10 | |
---|---|---|---|---|---|---|---|---|---|---|
Section | gridT.votemper,gridS.vosaline,gridV.vomecrty,p... | data.unify_chunks().persist() | FramS | section | None | {'vosaline': (33,36.2), 'votemper': (-2,6), 'v... | None | S-1 |
Each computation consists of
%%time
import os
calcswitch=os.environ.get('calc', 'True')
loaddata=((df.Inputs != '').any())
print('calcswitch=',calcswitch,'df.Inputs != nothing',loaddata)
data = load.datas(catalog_url,df.Inputs,month,year,daskreport) if ((calcswitch=='True' )*loaddata) else 0
data
calcswitch= True df.Inputs != nothing True ../lib/SEDNA_DELTA_MONITOR.yaml using param_xios reading ../lib/SEDNA_DELTA_MONITOR.yaml using param_xios reading <bound method DataSourceBase.describe of sources: param_xios: args: combine: by_coords concat_dim: y urlpath: /ccc/work/cont003/gen7420/odakatin/CONFIGS/SEDNA/SEDNA-I/SEDNA_Domain_cfg_Tgt_20210423_tsh10m_L1/param_f32/x_*.nc xarray_kwargs: compat: override coords: minimal data_vars: minimal parallel: true description: SEDNA NEMO parameters from MPI output nav_lon lat fails driver: intake_xarray.netcdf.NetCDFSource metadata: catalog_dir: /ccc/work/cont003/gen7420/odakatin/monitor-sedna/notebook/../lib/ > {'name': 'param_xios', 'container': 'xarray', 'plugin': ['netcdf'], 'driver': ['netcdf'], 'description': 'SEDNA NEMO parameters from MPI output nav_lon lat fails', 'direct_access': 'forbid', 'user_parameters': [{'name': 'path', 'description': 'file coordinate', 'type': 'str', 'default': '/ccc/work/cont003/gen7420/odakatin/CONFIGS/SEDNA/MESH/SEDNA_mesh_mask_Tgt_20210423_tsh10m_L1/param'}], 'metadata': {}, 'args': {'urlpath': '/ccc/work/cont003/gen7420/odakatin/CONFIGS/SEDNA/SEDNA-I/SEDNA_Domain_cfg_Tgt_20210423_tsh10m_L1/param_f32/x_*.nc', 'combine': 'by_coords', 'concat_dim': 'y'}} 0 read gridS ['vosaline'] using load_data_xios reading gridS using load_data_xios reading <bound method DataSourceBase.describe of sources: data_xios: args: combine: by_coords concat_dim: time_counter,x,y urlpath: /ccc/scratch/cont003/gen7420/talandel/SEDNA/SEDNA-DELTA-S/SPLIT/1d/2012/02/*gridS_*_0[0-5][0-9][0-9].nc xarray_kwargs: compat: override coords: minimal data_vars: minimal drop_variables: !!set deptht_bounds: null depthu_bounds: null nav_lat: null nav_lon: null time_centerd: null time_centered_bounds: null time_counter_bounds: null parallel: true preprocess: !!python/name:core.load.prep '' description: SEDNA NEMO outputs from different xios server driver: intake_xarray.netcdf.NetCDFSource metadata: catalog_dir: /ccc/work/cont003/gen7420/odakatin/monitor-sedna/notebook/../lib/ > took 54.640953540802 seconds 0 merging gridS ['vosaline'] 1 read gridT ['votemper'] using load_data_xios reading gridT using load_data_xios reading <bound method DataSourceBase.describe of sources: data_xios: args: combine: by_coords concat_dim: time_counter,x,y urlpath: /ccc/scratch/cont003/gen7420/talandel/SEDNA/SEDNA-DELTA-S/SPLIT/1d/2012/02/*gridT_*_0[0-5][0-9][0-9].nc xarray_kwargs: compat: override coords: minimal data_vars: minimal drop_variables: !!set deptht_bounds: null depthu_bounds: null nav_lat: null nav_lon: null time_centerd: null time_centered_bounds: null time_counter_bounds: null parallel: true preprocess: !!python/name:core.load.prep '' description: SEDNA NEMO outputs from different xios server driver: intake_xarray.netcdf.NetCDFSource metadata: catalog_dir: /ccc/work/cont003/gen7420/odakatin/monitor-sedna/notebook/../lib/ > took 4.702946186065674 seconds 1 merging gridT ['votemper'] took 0.24201154708862305 seconds 2 read gridV ['vomecrty'] using load_data_xios reading gridV using load_data_xios reading <bound method DataSourceBase.describe of sources: data_xios: args: combine: by_coords concat_dim: time_counter,x,y urlpath: /ccc/scratch/cont003/gen7420/talandel/SEDNA/SEDNA-DELTA-S/SPLIT/1d/2012/02/*gridV_*_0[0-5][0-9][0-9].nc xarray_kwargs: compat: override coords: minimal data_vars: minimal drop_variables: !!set deptht_bounds: null depthu_bounds: null nav_lat: null nav_lon: null time_centerd: null time_centered_bounds: null time_counter_bounds: null parallel: true preprocess: !!python/name:core.load.prep '' description: SEDNA NEMO outputs from different xios server driver: intake_xarray.netcdf.NetCDFSource metadata: catalog_dir: /ccc/work/cont003/gen7420/odakatin/monitor-sedna/notebook/../lib/ > took 4.309510707855225 seconds 2 merging gridV ['vomecrty'] took 0.23908114433288574 seconds param mask2d will be included in data param depth will be included in data param nav_lon will be included in data param nav_lat will be included in data param mask will be included in data
--------------------------------------------------------------------------- ValueError Traceback (most recent call last) <timed exec> in <module> /ccc/work/cont003/gen7420/odakatin/monitor-sedna/notebook/core/load.py in datas(catalog_url, dfi, month, year, daskreport, lazy) 607 ychunk=os.environ.get('ychunk',"False" ) 608 if not ('False' in ychunk) : --> 609 ychunk=int(ychunk.replace('\'', '').replace('\"', '')) 610 data=datas_y_rechunk(data,daskreport=daskreport,ychunk=ychunk) 611 tchunk=os.environ.get('tchunk',"False" ) ValueError: invalid literal for int() with base 10: ''
%%time
monitor.auto(df,data,savefig,daskreport,outputpath,file_exp='SEDNA'
)
--------------------------------------------------------------------------- NameError Traceback (most recent call last) <timed eval> in <module> NameError: name 'data' is not defined