%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= irene4219.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 irene4219.c-irene.mg1.tgcc.ccc.cea.fr using SEDNA_ALPHA_MONITOR file experiment, read from ../lib/SEDNA_ALPHA_MONITOR.yaml on year= * on month= 21 outputpath= ../results/SEDNA_ALPHA_MONITOR/21/ daskreport= ../results/dask/2672790irene4219.c-irene.mg1.tgcc.ccc.cea.fr_SEDNA_ALPHA_MONITOR_21Sections/ CPU times: user 449 ms, sys: 253 ms, total: 702 ms Wall time: 12.4 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.drop_vars('vozocrtx').unify_chunks().pers... | FramS | section | None | {'vosaline': (33,36.2), 'votemper': (-2,6), 'v... | None | S-1 | ||
Section | gridT.votemper,gridS.vosaline,gridU.vozocrtx,p... | data.drop_vars('vomecrty').chunk({'y':-1}).uni... | BFGS | section | None | {'vosaline': (28,35), 'votemper': (-2,2), 'voz... | None | S-2 |
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= False df.Inputs != nothing True CPU times: user 433 µs, sys: 0 ns, total: 433 µs Wall time: 400 µs
0
%%time
monitor.auto(df,data,savefig,daskreport,outputpath,file_exp='SEDNA'
)
#calc= False #save= False #plot= True Zone='FramS' Value='Section' cmap='None' clabel='None' clim= {'vosaline': (33, 36.2), 'votemper': (-2, 6), 'vomecrty': (-0.05, 0.05)} outputpath='../results/SEDNA_ALPHA_MONITOR/21/' nc_outputpath='../nc_results/SEDNA_ALPHA_MONITOR/21/' filename='SEDNA_section_FramS_Section' #3 no computing , loading starts dtaa=save.load_data(plot=Plot,path=nc_outputpath,filename=filename) start saving data filename= ../nc_results/SEDNA_ALPHA_MONITOR/21/SEDNA_section_FramS_Section/t_*.nc load computed data completed
<xarray.Dataset> Dimensions: (t: 9, x: 554, z: 103) Coordinates: * t (t) object 2004-06-01 12:00:00 ... 2004-06-09 12:00:00 y int64 ... * x (x) int64 3749 3750 3751 3752 3753 ... 4298 4299 4300 4301 4302 * z (z) int64 1 2 3 4 5 6 7 8 9 10 ... 95 96 97 98 99 100 101 102 103 nav_lon (x) float32 dask.array<chunksize=(554,), meta=np.ndarray> nav_lat (x) float32 dask.array<chunksize=(554,), meta=np.ndarray> mask (z, x) bool dask.array<chunksize=(103, 554), meta=np.ndarray> mask2d (x) bool dask.array<chunksize=(554,), meta=np.ndarray> depth (z, x) float32 dask.array<chunksize=(103, 554), meta=np.ndarray> Data variables: vosaline (t, z, x) float32 dask.array<chunksize=(1, 103, 554), meta=np.ndarray> votemper (t, z, x) float32 dask.array<chunksize=(1, 103, 554), meta=np.ndarray> vomecrty (t, z, x) float32 dask.array<chunksize=(1, 103, 554), meta=np.ndarray>
array([cftime.DatetimeNoLeap(2004, 6, 1, 12, 0, 0, 0), cftime.DatetimeNoLeap(2004, 6, 2, 12, 0, 0, 0), cftime.DatetimeNoLeap(2004, 6, 3, 12, 0, 0, 0), cftime.DatetimeNoLeap(2004, 6, 4, 12, 0, 0, 0), cftime.DatetimeNoLeap(2004, 6, 5, 12, 0, 0, 0), cftime.DatetimeNoLeap(2004, 6, 6, 12, 0, 0, 0), cftime.DatetimeNoLeap(2004, 6, 7, 12, 0, 0, 0), cftime.DatetimeNoLeap(2004, 6, 8, 12, 0, 0, 0), cftime.DatetimeNoLeap(2004, 6, 9, 12, 0, 0, 0)], dtype=object)
array(2609)
array([3749, 3750, 3751, ..., 4300, 4301, 4302])
array([ 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103])
|
|
|
|
|
|
|
|
#5 Plotting filename= plots.section(data,path=outputpath,filename=filename,save=savefig,cmap=cmap,clim=clim,clabel=clabel) ../results/SEDNA_ALPHA_MONITOR/21/SEDNA_section_FramS_Section_20040601-20040609.html starts plotting ../results/SEDNA_ALPHA_MONITOR/21/SEDNA_section_FramS_Section_20040601-20040609.html created
Zone='BFGS' Value='Section' cmap='None' clabel='None' clim= {'vosaline': (28, 35), 'votemper': (-2, 2), 'vozocrtx': (-0.05, 0.05)} outputpath='../results/SEDNA_ALPHA_MONITOR/21/' nc_outputpath='../nc_results/SEDNA_ALPHA_MONITOR/21/' filename='SEDNA_section_BFGS_Section' #3 no computing , loading starts dtaa=save.load_data(plot=Plot,path=nc_outputpath,filename=filename) start saving data filename= ../nc_results/SEDNA_ALPHA_MONITOR/21/SEDNA_section_BFGS_Section/t_*.nc load computed data completed
<xarray.Dataset> Dimensions: (t: 9, y: 2384, z: 95) Coordinates: * t (t) object 2004-06-01 12:00:00 ... 2004-06-09 12:00:00 * y (y) int64 3494 3495 3496 3497 3498 ... 5909 5910 5911 5912 5913 x int64 ... * z (z) int64 1 2 3 4 5 6 7 8 9 10 ... 86 87 88 89 90 91 92 93 94 95 nav_lon (y) float32 dask.array<chunksize=(2384,), meta=np.ndarray> nav_lat (y) float32 dask.array<chunksize=(2384,), meta=np.ndarray> mask (z, y) bool dask.array<chunksize=(95, 2384), meta=np.ndarray> mask2d (y) bool dask.array<chunksize=(2384,), meta=np.ndarray> depth (z, y) float32 dask.array<chunksize=(95, 2384), meta=np.ndarray> Data variables: vosaline (t, z, y) float32 dask.array<chunksize=(1, 95, 2384), meta=np.ndarray> votemper (t, z, y) float32 dask.array<chunksize=(1, 95, 2384), meta=np.ndarray> vozocrtx (t, z, y) float32 dask.array<chunksize=(1, 95, 2384), meta=np.ndarray>
array([cftime.DatetimeNoLeap(2004, 6, 1, 12, 0, 0, 0), cftime.DatetimeNoLeap(2004, 6, 2, 12, 0, 0, 0), cftime.DatetimeNoLeap(2004, 6, 3, 12, 0, 0, 0), cftime.DatetimeNoLeap(2004, 6, 4, 12, 0, 0, 0), cftime.DatetimeNoLeap(2004, 6, 5, 12, 0, 0, 0), cftime.DatetimeNoLeap(2004, 6, 6, 12, 0, 0, 0), cftime.DatetimeNoLeap(2004, 6, 7, 12, 0, 0, 0), cftime.DatetimeNoLeap(2004, 6, 8, 12, 0, 0, 0), cftime.DatetimeNoLeap(2004, 6, 9, 12, 0, 0, 0)], dtype=object)
array([3494, 3495, 3496, ..., 5911, 5912, 5913])
array(2281)
array([ 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95])
|
|
|
|
|
|
|
|
#5 Plotting filename= plots.section(data,path=outputpath,filename=filename,save=savefig,cmap=cmap,clim=clim,clabel=clabel) ../results/SEDNA_ALPHA_MONITOR/21/SEDNA_section_BFGS_Section_20040601-20040609.html starts plotting ../results/SEDNA_ALPHA_MONITOR/21/SEDNA_section_BFGS_Section_20040601-20040609.html created
CPU times: user 15.4 s, sys: 2.09 s, total: 17.4 s Wall time: 40.2 s