from hydroDL import pathSMAP, master
import os
from hydroDL.data import dbCsv
optData = master.default.update(master.default.optDataSMAP,
rootDB=pathSMAP['DB_L3_Global'],
subset='Globalv4f1_NorthAmerica',
tRange=[20150401, 20160401],
varT=dbCsv.varForcingGlobal)
optModel = master.default.optLstm
optLoss = master.default.optLossSigma
optTrain = master.default.optTrainSMAP
out = os.path.join(pathSMAP['Out_L3_Global'], 'test')
masterDict = master.wrapMaster(out, optData, optModel, optLoss, optTrain)
master.train(masterDict)
else:
cmd = "CUDA_VISIBLE_DEVICES={} screen -dmS {} python {} -F {} -M {}".format(
cudaID, screen, codePath, "train", mFile)
# if screen is None:
# #add some debugs Dapeng
# parser = argparse.ArgumentParser()
# parser.add_argument('-F', dest='func', type=str, default='train')
# parser.add_argument('-M', dest='mFile', type=str, default=mFile)
# args = parser.parse_args()
# if args.func == 'train':
# mDict = master.readMasterFile(args.mFile)
# master.train(mDict)
# # out = mDict['out']
# # email.sendEmail(subject='Training Done', text=out)
print(cmd)
os.system(cmd)
if __name__ == "__main__":
parser = argparse.ArgumentParser()
parser.add_argument("-F", dest="func", type=str)
parser.add_argument("-M", dest="mFile", type=str)
args = parser.parse_args()
if args.func == "train":
mDict = master.readMasterFile(args.mFile)
master.train(mDict)
# out = mDict['out']
# email.sendEmail(subject='Training Done', text=out)
from hydroDL import pathSMAP, master
import os
cDir = os.path.dirname(os.path.abspath(__file__))
cDir = r'/home/kxf227/work/GitHUB/pyRnnSMAP/example/'
# define training options
optData = master.updateOpt(master.default.optDataCsv,
path=os.path.join(cDir, 'data'),
subset='CONUSv4f1',
tRange=[20150401, 20160401])
optModel = master.default.optLstm
optLoss = master.default.optLoss
optTrain = master.default.optTrainSMAP
out = os.path.join(cDir, 'output', 'CONUSv4f1')
masterDict = master.wrapMaster(out, optData, optModel, optLoss, optTrain)
# train
master.train(masterDict, overwrite=True)
regionId = args.regionId
# k = [7, 8, 13]
regionId = 7
for k in range(len(subsetLst)):
kc = regionId - 1
if k != kc:
outName = 'ecoRegion{:02d}{:02d}_v2f1'.format(regionId,
k + 1) + '_Forcing'
varLst = dbCsv.varForcing
optData = default.update(default.optDataSMAP,
rootDB=pathSMAP['DB_L3_NA'],
tRange=[20150401, 20160401],
varT=varLst)
optData = default.forceUpdate(optData,
subset=[subsetLst[kc], subsetLst[k]])
optModel = default.optLstm
optLoss = default.optLossRMSE
optTrain = default.optTrainSMAP
out = os.path.join(pathSMAP['Out_L3_NA'], 'ecoRegion', outName)
masterDict = wrapMaster(out, optData, optModel, optLoss, optTrain)
train(masterDict)
'''
source /home/kxf227/anaconda3/bin/activate
conda activate pytorch
CUDA_VISIBLE_DEVICES=0 python /home/kxf227/work/GitHUB/hydroDL-dev/app/region/trainEcoComb.py --rid 7
CUDA_VISIBLE_DEVICES=1 python /home/kxf227/work/GitHUB/hydroDL-dev/app/region/trainEcoComb.py --rid 8
CUDA_VISIBLE_DEVICES=2 python /home/kxf227/work/GitHUB/hydroDL-dev/app/region/trainEcoComb.py --rid 13
'''
model,
xTrain,
yTrain,
attrs,
lossFun,
nEpoch=EPOCH,
miniBatch=[BATCH_SIZE, RHO],
saveEpoch=saveEPOCH,
saveFolder=out,
)
if interfaceOpt == 0:
# Only need to pass the wrapped configuration dict 'masterDict' for training
# nx, ny will be automatically updated later
masterDict = master.wrapMaster(out, optData, optModel, optLoss, optTrain)
master.train(masterDict)
## Not used here.
## A potential way to run batch jobs simultaneously in background through multiple GPUs and Linux screens.
## To use this, must manually set the "pathCamels['DB']" in hydroDL/__init__.py as your own root path of CAMELS data.
## Use the following master.runTrain() instead of the above master.train().
# master.runTrain(masterDict, cudaID=cid % gnum, screen='test-'+str(cid))
# cid = cid + 1
if 2 in Action:
# Train CNN-LSTM PUR model to integrate FDCs
# LCrange defines from which period to get synthetic FDC
LCTstr = str(LCrange[0]) + "-" + str(LCrange[1])
out = os.path.join(rootOut, save_path, hucdic, "Reg-85-95-Sub-Full-FDC" + LCTstr)
# out = os.path.join(rootOut, save_path, hucdic, 'Reg-85-95-Sub-5attr-FDC' + LCTstr)
# out = os.path.join(rootOut, save_path, hucdic, 'Reg-85-95-Sub-Noattr-FDC' + LCTstr)