def make_reports(inputDir, overwrite = False):
"""
Makes HTML reports for all valid input files found in the supplied directory.
Notes
------
A valid input file:
- ends with ``.tsv``
- does not have an accompanying file with the same path that ends in ``.html``
Parameters
----------
inputDir: str
path to directory to search for input files
"""
for input in find_input_IR_tsvs(inputDir):
params = {}
input_basename = os.path.splitext(input)[0]
# check for params files
input_json = input_basename + ".json"
if os.path.exists(input_json):
with open(input_json) as f:
params.update(json.load(f))
# check for output already exists
output = input_basename + ".html"
if os.path.exists(output):
if overwrite is True:
print(">>> Overwriting output: {0}".format(output))
report.make_report(input = input, output = output, params = params)
else:
print(">>> Generating output: {0}".format(output))
report.make_report(input = input, output = output, params = params)
def make_report_for_olympiad(dir_name):
src_dir = '../data/' + dir_name
out_dir = '../out/' + dir_name
results_dir = out_dir + '/results'
os.makedirs(results_dir, exist_ok=True)
funcs = get_all_evaluation_functions()
for i in range(len(funcs)):
out_file_path = results_dir + '/out_%s.json' % i
funcs[i](src_dir, out_file_path)
file_report_name = out_dir + '/' + dir_name + '.html'
make_report(results_dir,
file_report_name,
report_title='Olympiad from "%s"' % dir_name)
def main():
if args.vcs:
try:
clone_repository(args.vcs, args.path)
except GitError as err:
logger.critical(err)
exit(1)
ast_trees = get_trees(get_filenames(args.path))
if args.words == "verbs":
words = get_verbs_from_trees(ast_trees)
else:
words = get_nouns_from_trees(ast_trees)
print('-' * 80)
print('total %s words, %s unique' % (len(words), len(set(words))))
print('-' * 80)
top_size = args.count if args.count else None
make_report(args.format, args.report_file,
collections.Counter(words).most_common(top_size))
def main():
start_time = datetime.now()
local_db.create_db()
local_db.initial_scan()
local_db.add_audit()
import_scripts()
network_analysis = retrieve_audit_info()
if network_analysis != None:
local_db.add_SNMP_SSH_info(network_analysis[0], network_analysis[1])
else:
print('Warning: SNMP and SSH services is unavailable')
end_time = datetime.now()
duration = end_time - start_time
local_db.add_time(start_time.time().isoformat(timespec='seconds'),
end_time.time().isoformat(timespec='seconds'),
round(duration.total_seconds(), 2))
make_report()
local_db.close()
def main():
parser = argparse.ArgumentParser(
description="Create a variant calling report.")
parser.add_argument("input",
help="Input file in GA4GH metrics format",
nargs="*")
parser.add_argument(
"-o",
"--output",
help="Output file name for reports, e.g. 'report' to write "
"report.html",
required=True)
parser.add_argument("-m",
"--comparison-method",
default="default",
dest="comparison_method",
help="The comparison method that was used.")
parser.add_argument(
"-l",
"--result-list",
default=[],
dest="result_list",
action="append",
help="Result list in delimited format. Must have these columns: "
"method, comparisonmethod, and files.")
parser.add_argument(
"--roc-max-datapoints",
help=
"Maximum number of data points in a ROC (higher numbers might slow down our plotting)",
dest="roc_datapoints",
type=int,
default=1000)
parser.add_argument(
"--roc-resolution",
help=
"Minimum difference in precision / recall covered by the ROC curves.",
dest="roc_diff",
type=float,
default=0.005)
parser.add_argument(
"--min-recall",
help=
"Minimum recall for ROC curves (use to reduce size of output file by "
"clipping the bits of the ROC that are not meaningful)",
dest="min_recall",
type=float,
default=0.2)
parser.add_argument(
"--min-precision",
help=
"Minimum precision for ROC curves (use to reduce size of output file by"
" clipping the bits of the ROC that are not meaningful)",
dest="min_precision",
type=float,
default=0.0)
args = parser.parse_args()
if args.output.endswith(".gz"):
args.output = gzip.GzipFile(args.output, "w")
elif not args.output.endswith(".html"):
args.output += ".html"
if args.input:
metrics = report.metrics.read_qfy_csv(args.input,
args.comparison_method)
else:
metrics = []
for l in args.result_list:
print "reading %s" % l
csvfile = open(l, 'rb')
dialect = csv.Sniffer().sniff(csvfile.read(8192))
csvfile.seek(0)
dr = csv.DictReader(csvfile, dialect=dialect)
for row in dr:
rfiles = [x.strip() for x in row["files"].split(",")]
for i, rfile in enumerate(rfiles):
if not os.path.exists(rfile):
rfiles[i] = os.path.abspath(
os.path.join(os.path.dirname(l), rfile))
row_metrics = report.metrics.read_qfy_csv(
rfiles,
method=row["method"],
cmethod=row["comparisonmethod"],
roc_metrics=["METRIC.Precision", "METRIC.Recall"],
roc_diff=args.roc_diff,
max_data_points=args.roc_datapoints,
minmax={
"METRIC.Precision": {
"min": args.min_precision
},
"METRIC.Recall": {
"min": args.min_recall
}
})
metrics = metrics + row_metrics
loader = jinja2.FileSystemLoader(searchpath=TEMPLATEDIR)
env = jinja2.Environment(loader=loader)
template_vars = {"content": report.make_report(metrics)}
if not metrics:
raise Exception("No inputs specified.")
template = env.get_template("report.jinja2.html")
template.stream(**template_vars).dump(args.output)
except Exception as e:
print(f"problem dataset {i + 1}: {e}")
traceback.print_exc()
pass
print()
scores = pd.DataFrame(scores)
print(scores)
print(scores.sum() / len(scores))
scores.to_csv(output_fname, index=False)
if __name__ == '__main__':
real_data_path = "../data/with_class"
synth_data_path = "../data/synthetic"
real_data = util.read_data(real_data_path)
synthetic_data = util.read_data(synth_data_path)
start = datetime.datetime.now()
print(start)
os.mkdir(f'exp{start}')
test(synthetic_data, f"exp{start}/synth_test.csv")
test(real_data, f"exp{start}/real_test.csv")
print(datetime.datetime.now())
make_report(f'exp{start}')
def main():
parser = argparse.ArgumentParser(description="Create a variant calling report.")
parser.add_argument("input", help="Input file in GA4GH metrics format", nargs="*")
parser.add_argument(
"-o", "--output", help="Output file name for reports, e.g. 'report' to write " "report.html", required=True
)
parser.add_argument(
"-m",
"--comparison-method",
default="default",
dest="comparison_method",
help="The comparison method that was used.",
)
parser.add_argument(
"-l",
"--result-list",
default=[],
dest="result_list",
action="append",
help="Result list in delimited format. Must have these columns: " "method, comparisonmethod, and files.",
)
parser.add_argument(
"--roc-max-datapoints",
help="Maximum number of data points in a ROC (higher numbers might slow down our plotting)",
dest="roc_datapoints",
type=int,
default=1000,
)
parser.add_argument(
"--roc-resolution",
help="Minimum difference in precision / recall covered by the ROC curves.",
dest="roc_diff",
type=float,
default=0.005,
)
parser.add_argument(
"--min-recall",
help="Minimum recall for ROC curves (use to reduce size of output file by "
"clipping the bits of the ROC that are not meaningful)",
dest="min_recall",
type=float,
default=0.2,
)
parser.add_argument(
"--min-precision",
help="Minimum precision for ROC curves (use to reduce size of output file by"
" clipping the bits of the ROC that are not meaningful)",
dest="min_precision",
type=float,
default=0.0,
)
args = parser.parse_args()
if args.output.endswith(".gz"):
args.output = gzip.GzipFile(args.output, "w")
elif not args.output.endswith(".html"):
args.output += ".html"
if args.input:
metrics = report.metrics.read_qfy_csv(args.input, args.comparison_method)
else:
metrics = []
for l in args.result_list:
print "reading %s" % l
csvfile = open(l, "rb")
dialect = csv.Sniffer().sniff(csvfile.read(8192))
csvfile.seek(0)
dr = csv.DictReader(csvfile, dialect=dialect)
for row in dr:
rfiles = [x.strip() for x in row["files"].split(",")]
for i, rfile in enumerate(rfiles):
if not os.path.exists(rfile):
rfiles[i] = os.path.abspath(os.path.join(os.path.dirname(l), rfile))
row_metrics = report.metrics.read_qfy_csv(
rfiles,
method=row["method"],
cmethod=row["comparisonmethod"],
roc_metrics=["METRIC.Precision", "METRIC.Recall"],
roc_diff=args.roc_diff,
max_data_points=args.roc_datapoints,
minmax={"METRIC.Precision": {"min": args.min_precision}, "METRIC.Recall": {"min": args.min_recall}},
)
metrics = metrics + row_metrics
loader = jinja2.FileSystemLoader(searchpath=TEMPLATEDIR)
env = jinja2.Environment(loader=loader)
template_vars = {"content": report.make_report(metrics)}
if not metrics:
raise Exception("No inputs specified.")
template = env.get_template("report.jinja2.html")
template.stream(**template_vars).dump(args.output)
def main():
logging.basicConfig(level=logging.DEBUG,
format='%(asctime)s %(levelname)s %(message)s')
parser = argparse.ArgumentParser(
description="Benchmark a set of VCFs against "
"Platinum Genomes and Genome in a Bottle + produce a report.")
parser.add_argument("input", help="input VCF files", nargs="+")
parser.add_argument("-o",
"--output",
help="output HTML file name",
dest="output")
parser.add_argument(
"-O",
"--keep-results",
help="Specify a folder in which to keep the hap.py results.",
dest="keep_results",
default=None)
parser.add_argument("-e",
"--engine",
help="Comparison engine to use (xcmp / vcfeval)",
default="xcmp",
choices=["xcmp", "vcfeval"])
parser.add_argument("--truthset",
dest="truthset",
help="Add truthsets in this format: 'name:vcf:bed'",
default=[],
action="append")
parser.add_argument(
"--stratification",
dest="stratification",
help=
"Use a list of stratification regions (this should point to the TSV file). "
"Prefix with 'fixchr:' to add chr prefix to chromosome names.",
default=None)
parser.add_argument("--roc",
help="ROC feature to use",
dest="roc",
default="QUAL")
parser.add_argument(
"--ignore-filters",
help="Filters to ignore (e.g. --ignore-filters MQ,lowQual).",
dest="roc_filters",
default=None)
parser.add_argument("--happy-extra",
help="hap.py extra arguments",
dest="happy_extra",
default=None)
parser.add_argument("-r",
"--reference",
help="Reference FASTA file to use",
dest="reference",
default="hg19")
parser.add_argument(
"--roc-max-datapoints",
help=
"Maximum number of data points in a ROC (higher numbers might slow down our plotting)",
dest="roc_datapoints",
type=int,
default=1000)
parser.add_argument(
"--roc-resolution",
help=
"Minimum difference in precision / recall covered by the ROC curves.",
dest="roc_diff",
type=float,
default=0.005)
parser.add_argument(
"--min-recall",
help=
"Minimum recall for ROC curves (use to reduce size of output file by "
"clipping the bits of the ROC that are not meaningful)",
dest="min_recall",
type=float,
default=0.2)
parser.add_argument(
"--min-precision",
help=
"Minimum precision for ROC curves (use to reduce size of output file by"
" clipping the bits of the ROC that are not meaningful)",
dest="min_precision",
type=float,
default=0.0)
args = parser.parse_args()
if args.output.endswith(".gz"):
args.output = gzip.GzipFile(args.output, "w")
elif not args.output.endswith(".html"):
args.output += ".html"
if args.keep_results and not os.path.isdir(args.keep_results):
raise Exception("Output folder %s does not exist." % args.keep_results)
truthsets = []
for x in args.truthset:
xs = x.split(":", 2)
if len(xs) != 3:
raise Exception("Invalid custom truthset: %s" % x)
truthsets.append(Truthset(xs[0], xs[1], xs[2]))
resultlist = []
metrics = []
logging.debug(str(DEPS))
logging.debug(str(args))
try:
results_for_report = []
for query in args.input:
methodname = os.path.basename(query)
if methodname.endswith(".gz"):
methodname = methodname[:-3]
if methodname.endswith(".vcf"):
methodname = methodname[:-4]
if methodname.endswith(".bcf"):
methodname = methodname[:-4]
for ts in truthsets:
logging.info("Comparing %s against %s" % (query, ts.name))
# run with PG
extra_args = args.happy_extra
if extra_args is None:
extra_args = ""
if args.stratification:
if args.stratification.startswith("fixchr:"):
extra_args += " --stratification-fixchr"
args.stratification = args.stratification[len("fixchr:"
):]
extra_args += " --stratification %s" % args.stratification
happyfiles = happy(query,
truth=ts.vcf,
truth_conf=ts.bed,
roc_feature=args.roc,
roc_filters=args.roc_filters,
engine=args.engine,
ref=args.reference,
extra=extra_args,
output_folder=args.keep_results,
output_prefix=args.engine + "_" +
methodname)
resultlist += happyfiles
results_for_report.append({
"method":
methodname,
"cmethod":
ts.name + "-" + args.engine,
"files":
[r for r in happyfiles if r.endswith(".roc.all.csv.gz")]
})
if args.keep_results:
with open(
os.path.join(
args.keep_results,
os.path.splitext(os.path.basename(args.output))[0] +
".tsv"), "w") as f:
print >> f, "\t".join(["method", "comparisonmethod", "files"])
for row in results_for_report:
print >> f, "\t".join([
row["method"], row["cmethod"], ",".join(row["files"])
])
for row in results_for_report:
row_metrics = report.metrics.read_qfy_csv(
row["files"],
method=row["method"],
cmethod=row["cmethod"],
roc_metrics=["METRIC.Precision", "METRIC.Recall"],
roc_diff=args.roc_diff,
max_data_points=args.roc_datapoints,
minmax={
"METRIC.Precision": {
"min": args.min_precision
},
"METRIC.Recall": {
"min": args.min_recall
}
})
metrics = metrics + row_metrics
finally:
# clean results unless we want to keep the scratch space
if not args.keep_results:
for r in resultlist:
try:
os.unlink(r)
except:
pass
logging.info("Making report %s from %i results" %
(args.output, len(metrics)))
loader = jinja2.FileSystemLoader(searchpath=TEMPLATEDIR)
env = jinja2.Environment(loader=loader)
template_vars = {"content": report.make_report(metrics)}
template = env.get_template("report.jinja2.html")
template.stream(**template_vars).dump(args.output)
def main():
logging.basicConfig(level=logging.DEBUG, format="%(asctime)s %(levelname)s %(message)s")
parser = argparse.ArgumentParser(
description="Benchmark a set of VCFs against " "Platinum Genomes and Genome in a Bottle + produce a report."
)
parser.add_argument("input", help="input VCF files", nargs="+")
parser.add_argument("-o", "--output", help="output HTML file name", dest="output")
parser.add_argument(
"-O",
"--keep-results",
help="Specify a folder in which to keep the hap.py results.",
dest="keep_results",
default=None,
)
parser.add_argument(
"-e", "--engine", help="Comparison engine to use (xcmp / vcfeval)", default="xcmp", choices=["xcmp", "vcfeval"]
)
parser.add_argument(
"--truthset", dest="truthset", help="Add truthsets in this format: 'name:vcf:bed'", default=[], action="append"
)
parser.add_argument(
"--stratification",
dest="stratification",
help="Use a list of stratification regions (this should point to the TSV file). "
"Prefix with 'fixchr:' to add chr prefix to chromosome names.",
default=None,
)
parser.add_argument("--roc", help="ROC feature to use", dest="roc", default="QUAL")
parser.add_argument(
"--ignore-filters",
help="Filters to ignore (e.g. --ignore-filters MQ,lowQual).",
dest="roc_filters",
default=None,
)
parser.add_argument("--happy-extra", help="hap.py extra arguments", dest="happy_extra", default=None)
parser.add_argument("-r", "--reference", help="Reference FASTA file to use", dest="reference", default="hg19")
parser.add_argument(
"--roc-max-datapoints",
help="Maximum number of data points in a ROC (higher numbers might slow down our plotting)",
dest="roc_datapoints",
type=int,
default=1000,
)
parser.add_argument(
"--roc-resolution",
help="Minimum difference in precision / recall covered by the ROC curves.",
dest="roc_diff",
type=float,
default=0.005,
)
parser.add_argument(
"--min-recall",
help="Minimum recall for ROC curves (use to reduce size of output file by "
"clipping the bits of the ROC that are not meaningful)",
dest="min_recall",
type=float,
default=0.2,
)
parser.add_argument(
"--min-precision",
help="Minimum precision for ROC curves (use to reduce size of output file by"
" clipping the bits of the ROC that are not meaningful)",
dest="min_precision",
type=float,
default=0.0,
)
args = parser.parse_args()
if args.output.endswith(".gz"):
args.output = gzip.GzipFile(args.output, "w")
elif not args.output.endswith(".html"):
args.output += ".html"
if args.keep_results and not os.path.isdir(args.keep_results):
raise Exception("Output folder %s does not exist." % args.keep_results)
truthsets = []
for x in args.truthset:
xs = x.split(":", 2)
if len(xs) != 3:
raise Exception("Invalid custom truthset: %s" % x)
truthsets.append(Truthset(xs[0], xs[1], xs[2]))
resultlist = []
metrics = []
logging.debug(str(DEPS))
logging.debug(str(args))
try:
results_for_report = []
for query in args.input:
methodname = os.path.basename(query)
if methodname.endswith(".gz"):
methodname = methodname[:-3]
if methodname.endswith(".vcf"):
methodname = methodname[:-4]
if methodname.endswith(".bcf"):
methodname = methodname[:-4]
for ts in truthsets:
logging.info("Comparing %s against %s" % (query, ts.name))
# run with PG
extra_args = args.happy_extra
if extra_args is None:
extra_args = ""
if args.stratification:
if args.stratification.startswith("fixchr:"):
extra_args += " --stratification-fixchr"
args.stratification = args.stratification[len("fixchr:") :]
extra_args += " --stratification %s" % args.stratification
happyfiles = happy(
query,
truth=ts.vcf,
truth_conf=ts.bed,
roc_feature=args.roc,
roc_filters=args.roc_filters,
engine=args.engine,
ref=args.reference,
extra=extra_args,
output_folder=args.keep_results,
output_prefix=args.engine + "_" + methodname,
)
resultlist += happyfiles
results_for_report.append(
{
"method": methodname,
"cmethod": ts.name + "-" + args.engine,
"files": [r for r in happyfiles if r.endswith(".roc.all.csv.gz")],
}
)
if args.keep_results:
with open(
os.path.join(args.keep_results, os.path.splitext(os.path.basename(args.output))[0] + ".tsv"), "w"
) as f:
print >> f, "\t".join(["method", "comparisonmethod", "files"])
for row in results_for_report:
print >> f, "\t".join([row["method"], row["cmethod"], ",".join(row["files"])])
for row in results_for_report:
row_metrics = report.metrics.read_qfy_csv(
row["files"],
method=row["method"],
cmethod=row["cmethod"],
roc_metrics=["METRIC.Precision", "METRIC.Recall"],
roc_diff=args.roc_diff,
max_data_points=args.roc_datapoints,
minmax={"METRIC.Precision": {"min": args.min_precision}, "METRIC.Recall": {"min": args.min_recall}},
)
metrics = metrics + row_metrics
finally:
# clean results unless we want to keep the scratch space
if not args.keep_results:
for r in resultlist:
try:
os.unlink(r)
except:
pass
logging.info("Making report %s from %i results" % (args.output, len(metrics)))
loader = jinja2.FileSystemLoader(searchpath=TEMPLATEDIR)
env = jinja2.Environment(loader=loader)
template_vars = {"content": report.make_report(metrics)}
template = env.get_template("report.jinja2.html")
template.stream(**template_vars).dump(args.output)
report.GROUPS = [
# '613019056d70a378bf512cbc2cf0bb01128033da092b79fb576a5eb2bc4e7a11', # 运营日报测试群
'7e95f199da13a7483da519274be704c8d0e8557bc55323f67441e9e116fe66c6', # TiDev
'a8d8a969caebddef437c9582f93da87713b9bd4b0fea6da490ed1f0b7e50ae35', # DDC 大群
]
report.STORAGE_URL = 'http://tiadmin.jinns.top/query/temp<rand>.png'
report.REPORT_URL = 'http://tiadmin.jinns.top/query/ddc_daily_report/'
target = '12:00'
while True:
now = datetime.datetime.now()
code = now.strftime('%H:%M')
print(now, code, target)
if code == target:
print('-------------------------------')
try:
report.make_report()
report.push_report()
except Exception as e:
print('========================')
traceback.print_exc()
print('========================')
time.sleep(60)
time.sleep(20)
random_agent,
report_folder,
word_list,
dir_list,
fnm_list,
ext_list,
url_set,
) = get_settings(args)
print("[2] : Done")
print("[3] : Start Scanning according to settings")
initial_result, final_result = scan.scan_entry(url, port, max_depth,
timeout, delay,
random_agent, word_list,
url_set)
print("[3] : Done")
print("[4] : Making Reports")
# make reports
report.make_report(initial_result, report_folder, "initial")
report.make_report(final_result, report_folder, "final")
print("[4] : All DONE! Check ./reports/ ")
"""
print("[5] : Merge Testing")
dirchecker.dirchecker_entry(args, dir_list, fnm_list, ext_list)
print("[5] : Merge Done")
"""
