# out = os.path.join(pathSMAP['Out_L3_NA'], 'DA', 'CONUSv2f1_LSTM'+yrLst[k])
# masterDict = master.wrapMaster(out, optData, optModel, optLoss, optTrain)
# master.runTrain(masterDict, cudaID=k % 3, screen='LSTM' + yrLst[k])
# k=0
optData = default.update(default.optDataSMAP,
rootDB=pathSMAP['DB_L3_NA'],
subset='CONUSv2f1',
tRange=tLst[k])
optModel = default.update(default.optLstmClose,
name='hydroDL.model.rnn.AnnModel')
optLoss = default.optLossRMSE
optTrain = default.update(default.optTrainSMAP, nEpoch=300)
out = os.path.join(pathSMAP['Out_L3_NA'], 'DA', 'CONUSv2f1_NN' + yrLst[k])
masterDict = master.wrapMaster(out, optData, optModel, optLoss, optTrain)
master.runTrain(masterDict, cudaID=k % 3, screen='NN' + yrLst[k])
# master.train(masterDict)
optData = default.update(default.optDataSMAP,
rootDB=pathSMAP['DB_L3_NA'],
subset='CONUSv2f1',
tRange=tLst[k],
daObs=1)
optModel = default.update(default.optLstmClose,
name='hydroDL.model.rnn.AnnCloseModel')
optLoss = default.optLossRMSE
optTrain = default.update(default.optTrainSMAP, nEpoch=300)
out = os.path.join(pathSMAP['Out_L3_NA'], 'DA',
'CONUSv2f1_DANN' + yrLst[k])
masterDict = master.wrapMaster(out, optData, optModel, optLoss, optTrain)
master.runTrain(masterDict, cudaID=k % 3, screen='DANN' + yrLst[k])
from hydroDL import pathSMAP, master, utils
from hydroDL.master import default
from hydroDL.post import plot, stat
import os
import matplotlib.pyplot as plt
import numpy as np
import torch
# train
optData = default.update(default.optDataSMAP,
rootDB=pathSMAP['DB_L3_NA'],
subset='CONUSv4f1',
tRange=[20150402, 20180401],
daObs=1)
optModel = default.optLstmClose
optLoss = default.optLossRMSE
optTrain = default.update(default.optTrainSMAP, nEpoch=500)
out = os.path.join(pathSMAP['Out_L3_NA'], 'DA', 'CONUS_DA_3yr')
masterDict = master.wrapMaster(out, optData, optModel, optLoss, optTrain)
master.runTrain(masterDict, cudaID=2, screen='DA')
# test
subset = 'CONUS'
tRange = [20150402, 20180401]
# out = os.path.join(pathSMAP['Out_L3_NA'], 'DA', 'CONUSv2f1_DA2015')
out = os.path.join(pathSMAP['Out_L3_NA'], 'DA', 'CONUS_DA_3yr')
df, yf, obs = master.test(out, tRange=tRange, subset=subset, batchSize=100)
# /home/kxf227/work/GitHUB/hydroDL-dev/app/closeLoop/fullCONUS.py
# optModel = default.optLstm
# optLoss = default.optLossRMSE
# optTrain = default.optTrainSMAP
# out = os.path.join(pathSMAP['Out_L3_NA'], 'ecoRegionCase',
# outName)
# masterDict = wrapMaster(out, optData, optModel, optLoss,
# optTrain)
# runTrain(masterDict, cudaID=cid % 3, screen=case)
# # train(masterDict)
# cid = cid+1
cid = 2
for case in caseLst:
for k in [0, 1]:
subset = 'ecoReg_{}_L{}_sampleLin'.format(case, k)
outName = subset + '_Forcing'
varLst = dbCsv.varForcing
optData = default.update(default.optDataSMAP,
rootDB=pathSMAP['DB_L3_NA'],
tRange=[20150401, 20160401],
varT=varLst)
optData = default.forceUpdate(optData, subset=subset)
optModel = default.optLstm
optLoss = default.optLossRMSE
optTrain = default.optTrainSMAP
out = os.path.join(pathSMAP['Out_L3_NA'], 'ecoRegionCase', outName)
masterDict = wrapMaster(out, optData, optModel, optLoss, optTrain)
runTrain(masterDict, cudaID=cid % 3, screen=case)
# train(masterDict)
cid = cid + 1
tLst = [[20150501, 20151001], [20150402, 20160401]]
tagLst = ['2015RK', '2015']
for k in range(len(tLst)):
optData = default.update(default.optDataSMAP,
varT=['APCP_FORA'],
rootDB=pathSMAP['DB_L3_NA'],
subset='CONUSv2f1',
tRange=tLst[k],
daObs=1)
optModel = default.optLstmClose
optLoss = default.optLossRMSE
optTrain = default.update(default.optTrainSMAP, nEpoch=500)
out = os.path.join(pathSMAP['Out_L3_NA'], 'DA',
'CONUSv2f1_DA_Prcp_' + tagLst[k])
masterDict = master.wrapMaster(out, optData, optModel, optLoss, optTrain)
master.runTrain(masterDict, cudaID=(k + 1) % 3, screen='DA' + tagLst[k])
optData = default.update(default.optDataSMAP,
varT=['APCP_FORA'],
rootDB=pathSMAP['DB_L3_NA'],
subset='CONUSv2f1',
tRange=tLst[k])
optModel = default.optLstm
optLoss = default.optLossRMSE
optTrain = default.update(default.optTrainSMAP, nEpoch=500)
out = os.path.join(pathSMAP['Out_L3_NA'], 'DA',
'CONUSv2f1_LSTM_Prcp_' + tagLst[k])
masterDict = master.wrapMaster(out, optData, optModel, optLoss, optTrain)
master.runTrain(masterDict, cudaID=(k + 1) % 3, screen='LSTM' + tagLst[k])
# training
else:
varLst = dbCsv.varSoilmGlobal
optData = master.default.update(master.default.optDataSMAP,
rootDB=pathSMAP['DB_L3_Global'],
subset=subsetLst[k],
tRange=[20150401, 20160401],
varT=varLst)
optModel = master.default.optLstm
optLoss = master.default.optLossSigma
optTrain = master.default.optTrainSMAP
out = os.path.join(pathSMAP['Out_L3_Global'], outLst[k] + '_' + case)
masterDict = master.wrapMaster(out, optData, optModel, optLoss,
optTrain)
master.runTrain(masterDict, cudaID=cid % 3, screen=outLst[k])
cid = cid + 1
# master.train(masterDict)
# some of them failed and rerun
# master.runTrain(
# r'/mnt/sdb/rnnSMAP/Model_SMAPgrid/L3_Global/Africa_v4f1_y1_Forcing/',
# cudaID=1,
# screen='Africa_v4f1_y1_Forcing')
# master.runTrain(
# r'/mnt/sdb/rnnSMAP/Model_SMAPgrid/L3_Global/Asia_v4f1_y1_Soilm/',
# cudaID=0,
# screen='Asia_v4f1_y1_Soilm')
# master.runTrain(
# r'/mnt/sdb/rnnSMAP/Model_SMAPgrid/L3_Global/NorthAmerica_v4f1_y1_Soilm/',
# cudaID=1,
for case in caseLst:
if case == 'Forcing':
varLst = dbCsv.varForcing
else:
varLst = dbCsv.varSoilM
optData = master.default.update(master.default.optDataSMAP,
rootDB=pathSMAP['DB_L3_NA'],
subset='CONUSv2f1',
tRange=[20150401, 20160401],
varT=varLst)
optModel = master.default.optLstm
optLoss = master.default.optLossRMSE
optTrain = master.default.optTrainSMAP
out = os.path.join(pathSMAP['Out_L3_NA'], 'CONUSv2f1_' + case)
masterDict = master.wrapMaster(out, optData, optModel, optLoss, optTrain)
master.runTrain(masterDict, cudaID=cid % 3, screen=case)
cid = cid + 1
# train for each region
cid = 0
for k in range(len(subsetLst)):
for case in caseLst:
if case == 'Forcing':
varLst = dbCsv.varForcing
else:
varLst = dbCsv.varSoilM
optData = master.default.update(master.default.optDataSMAP,
rootDB=pathSMAP['DB_L3_NA'],
subset=subsetLst[k],
tRange=[20150401, 20160401],
varT=varLst)
ndayda = list(range(1, nDay + 1))
optData = default.update(default.optDataCamels,
daObs=ndayda,
davar='streamflow')
optModel = default.update(default.optLstm, hiddenSize=HIDDENSIZE)
optLoss = default.optLoss
optTrain = default.update(default.optTrainCamels,
miniBatch=[BATCH_SIZE, RHO],
nEpoch=EPOCH)
save_path = exp_name + '/' + exp_disp + \
'/epochs{}_batch{}_rho{}_hiddensize{}'.format(optTrain['nEpoch'], optTrain['miniBatch'][0],
optTrain['miniBatch'][1], optModel['hiddenSize'])
out = os.path.join(pathCamels['Out'], save_path,
'All-90-95-DA' + str(nDay))
masterDict = master.wrapMaster(out, optData, optModel, optLoss, optTrain)
master.runTrain(masterDict, cudaID=cid % 3, screen='test-DA' + str(nDay))
cid = cid + 1
# test original model
caseLst = ['All-90-95']
nDayLst = [3, 7]
for nDay in nDayLst:
caseLst.append('All-90-95-DA' + str(nDay))
outLst = [os.path.join(pathCamels['Out'], save_path, x) for x in caseLst]
subset = 'All'
tRange = [19950101, 20000101]
predLst = list()
for out in outLst:
df, pred, obs = master.test(out,
tRange=tRange,
subset=subset,
from hydroDL import pathSMAP, master
import os
# define training options
out = os.path.join(pathSMAP['Out_L3_NA'], 'RegTest', 'CONUSv4f1_sigma')
optData = master.default.update(
master.default.optDataCsv,
rootDB=pathSMAP['DB_L3_NA'],
subset='CONUSv4f1',
tRange=[20150401, 20160401],
)
optModel = master.default.optLstm
optLoss = master.default.update(master.default.optLoss,
name='hydroDL.model.crit.SigmaLoss')
optTrain = master.default.optTrain
masterDict = master.wrapMaster(out, optData, optModel, optLoss, optTrain)
# train
master.runTrain(masterDict, cudaID=0, screenName='sigmaTest')
cid = 0
for nd in dLst:
optData = default.update(
default.optDataSMAP,
rootDB=pathSMAP['DB_L3_NA'],
subset='CONUSv2f1',
tRange=[20150501, 20160501],
daObs=nd)
optModel = default.optLstmClose
optLoss = default.optLossRMSE
optTrain = default.update(default.optTrainSMAP, nEpoch=300)
out = os.path.join(pathSMAP['Out_L3_NA'], 'DA',
'CONUSv2f1_d' + str(nd))
masterDict = master.wrapMaster(out, optData, optModel, optLoss,
optTrain)
master.runTrain(masterDict, cudaID=cid % 3, screen='d' + str(nd))
# master.train(masterDict)
cid = cid + 1
# vanila LSTM
optData = default.update(
default.optDataSMAP,
rootDB=pathSMAP['DB_L3_NA'],
subset='CONUSv2f1',
tRange=[20150501, 20160501])
optModel = default.optLstm
optLoss = default.optLossRMSE
optTrain = default.update(default.optTrainSMAP, nEpoch=300)
out = os.path.join(pathSMAP['Out_L3_NA'], 'DA', 'CONUSv2f1')
masterDict = master.wrapMaster(out, optData, optModel, optLoss, optTrain)
master.runTrain(masterDict, cudaID=0, screen='LSTM')