%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=Fluxnet
# 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 calc=True
%%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= irene4064.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 irene4064.c-irene.mg1.tgcc.ccc.cea.fr using SEDNA_ALPHA_MONITOR file experiment, read from ../lib/SEDNA_ALPHA_MONITOR.yaml on year= * on month= 23 outputpath= ../results/SEDNA_ALPHA_MONITOR/23/ daskreport= ../results/dask/2526597irene4064.c-irene.mg1.tgcc.ccc.cea.fr_SEDNA_ALPHA_MONITOR_23Ice_quantities/ CPU times: user 301 ms, sys: 247 ms, total: 549 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 | |
---|---|---|---|---|---|---|---|---|---|---|
Ice_quantities | param.e1te2t,icemod.sivelo,icemod.sivolu,icemo... | calc.Ice_quant(data) | BBFG | Ice_intquant | None | (0,20) | cm s^(-1) | I-2 |
Each computation consists of
%%time
import os
calcswitch=os.environ.get('calc', 'True')
print('calcswitch=',calcswitch)
data = load.datas(catalog_url,df.Inputs,month,year,daskreport) if calcswitch=='True' else 0
data
calcswitch= False CPU times: user 60 µs, sys: 15 µs, total: 75 µs Wall time: 62.7 µs
0
%%time
monitor.auto(df,data,savefig,daskreport,outputpath,file_exp='SEDNA'
)
#calc= False #save= False #plot= True Zone='BBFG' Value='Ice_quantities' cmap='None' clabel='cm s^(-1)' clim= (0, 20) outputpath='../results/SEDNA_ALPHA_MONITOR/23/' nc_outputpath='../nc_results/SEDNA_ALPHA_MONITOR/23/' filename='SEDNA_Ice_intquant_BBFG_Ice_quantities' #3 no computing , loading starts dtaa=save.load_data(plot=Plot,path=nc_outputpath,filename=filename) start saving data load 1Dnc file from ../nc_results/SEDNA_ALPHA_MONITOR/23/SEDNA_Ice_intquant_BBFG_Ice_quantities*.nc
--------------------------------------------------------------------------- ValueError Traceback (most recent call last) <timed eval> in <module> /ccc/work/cont003/gen7420/odakatin/monitor-sedna/notebook/core/monitor.py in auto(df, val, savefig, daskreport, outputpath, file_exp) 68 print('dtaa=save.load_data(plot=Plot,path=nc_outputpath,filename=filename)' ) 69 with performance_report(filename=daskreport+"_calc_"+step.Value+".html"): ---> 70 data=save.load_data(plot=Plot,path=nc_outputpath,filename=filename) 71 #saveswitch=False 72 display(data) /ccc/work/cont003/gen7420/odakatin/monitor-sedna/notebook/core/save.py in load_data(plot, path, filename) 27 print('start saving data') 28 if 'int' in plot: ---> 29 data=load_integral(path,filename) 30 elif 'section' in plot: 31 data=load_twoD(path,filename,nested=False) /ccc/work/cont003/gen7420/odakatin/monitor-sedna/notebook/core/save.py in load_integral(path, filename) 71 ,concat_dim=('t') 72 ,combine='by_coords' ---> 73 ,coords='minimal') 74 75 ~/monitor/lib/python3.7/site-packages/xarray/backends/api.py in open_mfdataset(paths, chunks, concat_dim, compat, preprocess, engine, lock, data_vars, coords, combine, autoclose, parallel, join, attrs_file, **kwargs) 980 coords=coords, 981 join=join, --> 982 combine_attrs="drop", 983 ) 984 else: ~/monitor/lib/python3.7/site-packages/xarray/core/combine.py in combine_by_coords(datasets, compat, data_vars, coords, fill_value, join, combine_attrs) 760 for vars, datasets_with_same_vars in grouped_by_vars: 761 combined_ids, concat_dims = _infer_concat_order_from_coords( --> 762 list(datasets_with_same_vars) 763 ) 764 ~/monitor/lib/python3.7/site-packages/xarray/core/combine.py in _infer_concat_order_from_coords(datasets) 107 if len(datasets) > 1 and not concat_dims: 108 raise ValueError( --> 109 "Could not find any dimension coordinates to use to " 110 "order the datasets for concatenation" 111 ) ValueError: Could not find any dimension coordinates to use to order the datasets for concatenation