def load_conf(self, args):
"""
Load the configuration file from standard paths or in arguments.
"""
self._collapse_arg(args, "cfpath")
if args.cfpath:
self.config = conf.load_conf(Path(args.cfpath).resolve())
else:
self.config = conf.default_conf()
print("Should use Python >= 3.6")
sys.exit()
cmt.ARGS = args.parse_arguments(sys.argv[1:])
if cmt.ARGS["version"]:
print(cmt.VERSION)
sys.exit()
err = args.get_invalid_modules_in_args()
if len(err) > 0:
print("ERR - Unknow module(s) : " + ','.join(err))
sys.exit()
# conf.yml, conf.d/*.yml
cmt.CONF = conf.load_conf()
# if cron mode, introduce a small uase (offset) tô spread the load on metrology servers
if cmt.ARGS['cron']:
mypause = conf.get_startoffset()
time.sleep(mypause)
# Persist
cmt.PERSIST = persist.Persist(file=cmt.DEFAULT_PERSIST_FILE)
if cmt.ARGS["nopersist"]:
cmt.PERSIST.dict = {}
lastrun = cmt.PERSIST.get_key("cmt_last_run", 0)
# remote conf (url) or cached conf
conf.load_conf_remote(cmt.CONF)
from src import american_community_survey as amc
from src import utils
from src import download_spark
## START
# Initiate the parser
args = utils.get_argparser().parse_args()
utils.printNowToFile("starting:")
utils.printNowToFile("downloading spark")
download_spark.download(os.getcwd())
###############################################################
if args.host and args.port:
spark = conf.load_conf(args.host, args.port)
else:
spark = conf.load_conf_default()
spark.sparkContext.addPyFile('ridge_regression.py')
import ridge_regression as rr
## PREPROCESSING: CLEANING
## path to dataset
DATA_PATH = './dataset'
df = amc.load_dataset(DATA_PATH, spark)
###############################################################
## PREPROCESSING: FEATURES ENGINEERING
# name of the target column and remove all the rows where 'PINCP' is null
#!/usr/bin/env python3
"""Evaluate variant call predictions against GiaB truth
"""
import sys
import tempfile
import os
import shutil
import subprocess
#import gzip
#import glob
#import io
from conf import load_conf
CONF = load_conf()
assert os.path.exists(CONF['highconf_regions'])
for f in CONF['truth_vcf'].values():
assert os.path.exists(f)
def num_vars_from_vcf(f):
#cmd = "zgrep -vc '^#' {}".format(f) fails on empty files
# ignore h**o to ref which can come from samples extracted from joint calls
cmd = "awk '$10 !~ /^0.0:/' {} | zgrep -v '^#' | wc -l".format(f)
res = subprocess.check_output(cmd, stderr=subprocess.STDOUT, shell=True)
return int(res.decode().strip())
def main(vartype, predvcf, truthvcf):
"""main function"""