def compute_capacity(stream: VCFStream, population: Population, repeats: int,
ratio: float, predictor_factory: PredictorFactory):
individuals = stream.individuals
split_index = int(len(individuals) * ratio)
labels = population.get_labels(individuals)
# Split once, since the call to `split` takes a lot of time to do for every
# polymorphism and repeat. We call for the number of repeats and apply the
# corresponding one, copying the same approach for all polymorphisms.
if repeats > 1:
train_target_indices_list = [
split(range(len(individuals)), split_index) for _ in range(repeats)
]
train_target_indices = [(set(i), set(j))
for i, j in train_target_indices_list]
else:
train_target_indices = [(set(range(len(individuals))),
set(range(len(individuals))))]
for polymorphism in stream:
accuracies: List[float] = []
for repeat_idx in range(repeats):
genotypes = (process_variant(i) for i in polymorphism.data_fields)
to_split = zip(individuals, genotypes, labels)
train = []
target = []
for i, val in enumerate(to_split):
if i in train_target_indices[repeat_idx][0]:
train.append(val)
if i in train_target_indices[repeat_idx][1]:
target.append(val)
_train_individuals, train_genotypes, train_labels = zip(*train)
predictor = predictor_factory.build(train_genotypes, train_labels)
count = 0.0
for _target_individual, genotype, label in target:
predictions = predictor.predict(genotype)
if label in predictions:
count += predictions[label]
accuracies.append(count / len(target))
m = mean(accuracies)
if repeats == 1:
print(f'{polymorphism.get_identifier()}\t{m:.5f}')
else:
s = stdev(accuracies)
print(f'{polymorphism.get_identifier()}\t{m:.5f}\t{s:.5f}')
def main():
df2 = pd.DataFrame()
if not (len(sys.argv) == 1 + 1):
print('\033[91m' + '✘ Error: ' + '\033[0m' +
'CSV file is missing, please add his path as argument')
sys.exit()
df = utils.dataframe(sys.argv[1])
columnsNamesArr = df.columns.values
listOfColumnNames = list(columnsNamesArr)
dico_numerals = {}
for label in listOfColumnNames:
if df[label].dtypes == str or df[label].dtypes == object:
continue
dico_numerals[label] = {
'count': 0,
'mean': 0,
'std': 0,
'min': float('inf'),
'25%': 0,
'50%': 0,
'75%': 0,
'max': float('-inf'),
'total': 0
}
for label in listOfColumnNames:
if df[label].dtypes == str or df[label].dtypes == object:
continue
for index, row in df.iterrows():
if row[label] > dico_numerals[label]['max']:
dico_numerals[label]['max'] = row[label]
if row[label] < dico_numerals[label]['min']:
dico_numerals[label]['min'] = row[label]
if np.isnan(row[label]) != True:
dico_numerals[label]['count'] += 1
dico_numerals[label]['total'] += row[label]
if (dico_numerals[label]['count'] > 0):
dico_numerals[label]['mean'] = dico_numerals[label][
'total'] / dico_numerals[label]['count']
dico_numerals[label]['25%'] = utils.calc_quantile(
df[label].dropna(), 0.25)
dico_numerals[label]['50%'] = utils.calc_quantile(
df[label].dropna(), 0.5)
dico_numerals[label]['75%'] = utils.calc_quantile(
df[label].dropna(), 0.75)
dico_numerals[label]['std'] = utils.stdev(df[label].dropna())
describe(dico_numerals)
def table_soak_vs_ndr(table, input_data):
"""Generate the table(s) with algorithm: table_soak_vs_ndr
specified in the specification file.
:param table: Table to generate.
:param input_data: Data to process.
:type table: pandas.Series
:type input_data: InputData
"""
logging.info(" Generating the table {0} ...".format(table.get(
"title", "")))
# Transform the data
logging.info(" Creating the data set for the {0} '{1}'.".format(
table.get("type", ""), table.get("title", "")))
data = input_data.filter_data(table, continue_on_error=True)
# Prepare the header of the table
try:
header = [
"Test case",
"{0} Throughput [Mpps]".format(table["reference"]["title"]),
"{0} Stdev [Mpps]".format(table["reference"]["title"]),
"{0} Throughput [Mpps]".format(table["compare"]["title"]),
"{0} Stdev [Mpps]".format(table["compare"]["title"]), "Delta [%]",
"Stdev of delta [%]"
]
header_str = ",".join(header) + "\n"
except (AttributeError, KeyError) as err:
logging.error(
"The model is invalid, missing parameter: {0}".format(err))
return
# Create a list of available SOAK test results:
tbl_dict = dict()
for job, builds in table["compare"]["data"].items():
for build in builds:
for tst_name, tst_data in data[job][str(build)].iteritems():
if tst_data["type"] == "SOAK":
tst_name_mod = tst_name.replace("-soak", "")
if tbl_dict.get(tst_name_mod, None) is None:
groups = re.search(REGEX_NIC, tst_data["parent"])
nic = groups.group(0) if groups else ""
name = "{0}-{1}".format(
nic, "-".join(tst_data["name"].split("-")[:-1]))
tbl_dict[tst_name_mod] = {
"name": name,
"ref-data": list(),
"cmp-data": list()
}
try:
tbl_dict[tst_name_mod]["cmp-data"].append(
tst_data["throughput"]["LOWER"])
except (KeyError, TypeError):
pass
tests_lst = tbl_dict.keys()
# Add corresponding NDR test results:
for job, builds in table["reference"]["data"].items():
for build in builds:
for tst_name, tst_data in data[job][str(build)].iteritems():
tst_name_mod = tst_name.replace("-ndrpdr", "").\
replace("-mrr", "")
if tst_name_mod in tests_lst:
try:
if tst_data["type"] in ("NDRPDR", "MRR", "BMRR"):
if table["include-tests"] == "MRR":
result = tst_data["result"]["receive-rate"].avg
elif table["include-tests"] == "PDR":
result = tst_data["throughput"]["PDR"]["LOWER"]
elif table["include-tests"] == "NDR":
result = tst_data["throughput"]["NDR"]["LOWER"]
else:
result = None
if result is not None:
tbl_dict[tst_name_mod]["ref-data"].append(
result)
except (KeyError, TypeError):
continue
tbl_lst = list()
for tst_name in tbl_dict.keys():
item = [
tbl_dict[tst_name]["name"],
]
data_r = tbl_dict[tst_name]["ref-data"]
if data_r:
data_r_mean = mean(data_r)
item.append(round(data_r_mean / 1000000, 2))
data_r_stdev = stdev(data_r)
item.append(round(data_r_stdev / 1000000, 2))
else:
data_r_mean = None
data_r_stdev = None
item.extend([None, None])
data_c = tbl_dict[tst_name]["cmp-data"]
if data_c:
data_c_mean = mean(data_c)
item.append(round(data_c_mean / 1000000, 2))
data_c_stdev = stdev(data_c)
item.append(round(data_c_stdev / 1000000, 2))
else:
data_c_mean = None
data_c_stdev = None
item.extend([None, None])
if data_r_mean and data_c_mean:
delta, d_stdev = relative_change_stdev(data_r_mean, data_c_mean,
data_r_stdev, data_c_stdev)
item.append(round(delta, 2))
item.append(round(d_stdev, 2))
tbl_lst.append(item)
# Sort the table according to the relative change
tbl_lst.sort(key=lambda rel: rel[-1], reverse=True)
# Generate csv tables:
csv_file = "{0}.csv".format(table["output-file"])
with open(csv_file, "w") as file_handler:
file_handler.write(header_str)
for test in tbl_lst:
file_handler.write(",".join([str(item) for item in test]) + "\n")
convert_csv_to_pretty_txt(csv_file, "{0}.txt".format(table["output-file"]))
def table_nics_comparison(table, input_data):
"""Generate the table(s) with algorithm: table_nics_comparison
specified in the specification file.
:param table: Table to generate.
:param input_data: Data to process.
:type table: pandas.Series
:type input_data: InputData
"""
logging.info(" Generating the table {0} ...".format(table.get(
"title", "")))
# Transform the data
logging.info(" Creating the data set for the {0} '{1}'.".format(
table.get("type", ""), table.get("title", "")))
data = input_data.filter_data(table, continue_on_error=True)
# Prepare the header of the tables
try:
header = [
"Test case",
]
if table["include-tests"] == "MRR":
hdr_param = "Receive Rate"
else:
hdr_param = "Throughput"
header.extend([
"{0} {1} [Mpps]".format(table["reference"]["title"], hdr_param),
"{0} Stdev [Mpps]".format(table["reference"]["title"]),
"{0} {1} [Mpps]".format(table["compare"]["title"], hdr_param),
"{0} Stdev [Mpps]".format(table["compare"]["title"]), "Delta [%]"
])
header_str = ",".join(header) + "\n"
except (AttributeError, KeyError) as err:
logging.error(
"The model is invalid, missing parameter: {0}".format(err))
return
# Prepare data to the table:
tbl_dict = dict()
for job, builds in table["data"].items():
for build in builds:
for tst_name, tst_data in data[job][str(build)].iteritems():
tst_name_mod = tst_name.replace("-ndrpdrdisc", "").\
replace("-ndrpdr", "").replace("-pdrdisc", "").\
replace("-ndrdisc", "").replace("-pdr", "").\
replace("-ndr", "").\
replace("1t1c", "1c").replace("2t1c", "1c").\
replace("2t2c", "2c").replace("4t2c", "2c").\
replace("4t4c", "4c").replace("8t4c", "4c")
tst_name_mod = re.sub(REGEX_NIC, "", tst_name_mod)
if tbl_dict.get(tst_name_mod, None) is None:
name = "-".join(tst_data["name"].split("-")[:-1])
tbl_dict[tst_name_mod] = {
"name": name,
"ref-data": list(),
"cmp-data": list()
}
try:
if table["include-tests"] == "MRR":
result = tst_data["result"]["receive-rate"].avg
elif table["include-tests"] == "PDR":
result = tst_data["throughput"]["PDR"]["LOWER"]
elif table["include-tests"] == "NDR":
result = tst_data["throughput"]["NDR"]["LOWER"]
else:
result = None
if result:
if table["reference"]["nic"] in tst_data["tags"]:
tbl_dict[tst_name_mod]["ref-data"].append(result)
elif table["compare"]["nic"] in tst_data["tags"]:
tbl_dict[tst_name_mod]["cmp-data"].append(result)
except (TypeError, KeyError) as err:
logging.debug("No data for {0}".format(tst_name))
logging.debug(repr(err))
# No data in output.xml for this test
tbl_lst = list()
for tst_name in tbl_dict.keys():
item = [
tbl_dict[tst_name]["name"],
]
data_t = tbl_dict[tst_name]["ref-data"]
if data_t:
item.append(round(mean(data_t) / 1000000, 2))
item.append(round(stdev(data_t) / 1000000, 2))
else:
item.extend([None, None])
data_t = tbl_dict[tst_name]["cmp-data"]
if data_t:
item.append(round(mean(data_t) / 1000000, 2))
item.append(round(stdev(data_t) / 1000000, 2))
else:
item.extend([None, None])
if item[-4] is not None and item[-2] is not None and item[-4] != 0:
item.append(int(relative_change(float(item[-4]), float(item[-2]))))
if len(item) == len(header):
tbl_lst.append(item)
# Sort the table according to the relative change
tbl_lst.sort(key=lambda rel: rel[-1], reverse=True)
# Generate csv tables:
csv_file = "{0}.csv".format(table["output-file"])
with open(csv_file, "w") as file_handler:
file_handler.write(header_str)
for test in tbl_lst:
file_handler.write(",".join([str(item) for item in test]) + "\n")
convert_csv_to_pretty_txt(csv_file, "{0}.txt".format(table["output-file"]))
def table_performance_comparison(table, input_data):
"""Generate the table(s) with algorithm: table_performance_comparison
specified in the specification file.
:param table: Table to generate.
:param input_data: Data to process.
:type table: pandas.Series
:type input_data: InputData
"""
logging.info(" Generating the table {0} ...".format(table.get(
"title", "")))
# Transform the data
logging.info(" Creating the data set for the {0} '{1}'.".format(
table.get("type", ""), table.get("title", "")))
data = input_data.filter_data(table, continue_on_error=True)
# Prepare the header of the tables
try:
header = [
"Test case",
]
if table["include-tests"] == "MRR":
hdr_param = "Receive Rate"
else:
hdr_param = "Throughput"
history = table.get("history", None)
if history:
for item in history:
header.extend([
"{0} {1} [Mpps]".format(item["title"], hdr_param),
"{0} Stdev [Mpps]".format(item["title"])
])
header.extend([
"{0} {1} [Mpps]".format(table["reference"]["title"], hdr_param),
"{0} Stdev [Mpps]".format(table["reference"]["title"]),
"{0} {1} [Mpps]".format(table["compare"]["title"], hdr_param),
"{0} Stdev [Mpps]".format(table["compare"]["title"]), "Delta [%]"
])
header_str = ",".join(header) + "\n"
except (AttributeError, KeyError) as err:
logging.error(
"The model is invalid, missing parameter: {0}".format(err))
return
# Prepare data to the table:
tbl_dict = dict()
for job, builds in table["reference"]["data"].items():
for build in builds:
for tst_name, tst_data in data[job][str(build)].iteritems():
tst_name_mod = tst_name.replace("-ndrpdrdisc", "").\
replace("-ndrpdr", "").replace("-pdrdisc", "").\
replace("-ndrdisc", "").replace("-pdr", "").\
replace("-ndr", "").\
replace("1t1c", "1c").replace("2t1c", "1c").\
replace("2t2c", "2c").replace("4t2c", "2c").\
replace("4t4c", "4c").replace("8t4c", "4c")
if "across topologies" in table["title"].lower():
tst_name_mod = tst_name_mod.replace("2n1l-", "")
if tbl_dict.get(tst_name_mod, None) is None:
groups = re.search(REGEX_NIC, tst_data["parent"])
nic = groups.group(0) if groups else ""
name = "{0}-{1}".format(
nic, "-".join(tst_data["name"].split("-")[:-1]))
if "across testbeds" in table["title"].lower() or \
"across topologies" in table["title"].lower():
name = name.\
replace("1t1c", "1c").replace("2t1c", "1c").\
replace("2t2c", "2c").replace("4t2c", "2c").\
replace("4t4c", "4c").replace("8t4c", "4c")
tbl_dict[tst_name_mod] = {
"name": name,
"ref-data": list(),
"cmp-data": list()
}
try:
# TODO: Re-work when NDRPDRDISC tests are not used
if table["include-tests"] == "MRR":
tbl_dict[tst_name_mod]["ref-data"]. \
append(tst_data["result"]["receive-rate"].avg)
elif table["include-tests"] == "PDR":
if tst_data["type"] == "PDR":
tbl_dict[tst_name_mod]["ref-data"]. \
append(tst_data["throughput"]["value"])
elif tst_data["type"] == "NDRPDR":
tbl_dict[tst_name_mod]["ref-data"].append(
tst_data["throughput"]["PDR"]["LOWER"])
elif table["include-tests"] == "NDR":
if tst_data["type"] == "NDR":
tbl_dict[tst_name_mod]["ref-data"]. \
append(tst_data["throughput"]["value"])
elif tst_data["type"] == "NDRPDR":
tbl_dict[tst_name_mod]["ref-data"].append(
tst_data["throughput"]["NDR"]["LOWER"])
else:
continue
except TypeError:
pass # No data in output.xml for this test
for job, builds in table["compare"]["data"].items():
for build in builds:
for tst_name, tst_data in data[job][str(build)].iteritems():
tst_name_mod = tst_name.replace("-ndrpdrdisc", ""). \
replace("-ndrpdr", "").replace("-pdrdisc", ""). \
replace("-ndrdisc", "").replace("-pdr", ""). \
replace("-ndr", "").\
replace("1t1c", "1c").replace("2t1c", "1c").\
replace("2t2c", "2c").replace("4t2c", "2c").\
replace("4t4c", "4c").replace("8t4c", "4c")
if "across topologies" in table["title"].lower():
tst_name_mod = tst_name_mod.replace("2n1l-", "")
try:
# TODO: Re-work when NDRPDRDISC tests are not used
if table["include-tests"] == "MRR":
tbl_dict[tst_name_mod]["cmp-data"]. \
append(tst_data["result"]["receive-rate"].avg)
elif table["include-tests"] == "PDR":
if tst_data["type"] == "PDR":
tbl_dict[tst_name_mod]["cmp-data"]. \
append(tst_data["throughput"]["value"])
elif tst_data["type"] == "NDRPDR":
tbl_dict[tst_name_mod]["cmp-data"].append(
tst_data["throughput"]["PDR"]["LOWER"])
elif table["include-tests"] == "NDR":
if tst_data["type"] == "NDR":
tbl_dict[tst_name_mod]["cmp-data"]. \
append(tst_data["throughput"]["value"])
elif tst_data["type"] == "NDRPDR":
tbl_dict[tst_name_mod]["cmp-data"].append(
tst_data["throughput"]["NDR"]["LOWER"])
else:
continue
except KeyError:
pass
except TypeError:
tbl_dict.pop(tst_name_mod, None)
if history:
for item in history:
for job, builds in item["data"].items():
for build in builds:
for tst_name, tst_data in data[job][str(
build)].iteritems():
tst_name_mod = tst_name.replace("-ndrpdrdisc", ""). \
replace("-ndrpdr", "").replace("-pdrdisc", ""). \
replace("-ndrdisc", "").replace("-pdr", ""). \
replace("-ndr", "").\
replace("1t1c", "1c").replace("2t1c", "1c").\
replace("2t2c", "2c").replace("4t2c", "2c").\
replace("4t4c", "4c").replace("8t4c", "4c")
if "across topologies" in table["title"].lower():
tst_name_mod = tst_name_mod.replace("2n1l-", "")
if tbl_dict.get(tst_name_mod, None) is None:
continue
if tbl_dict[tst_name_mod].get("history", None) is None:
tbl_dict[tst_name_mod]["history"] = OrderedDict()
if tbl_dict[tst_name_mod]["history"].get(
item["title"], None) is None:
tbl_dict[tst_name_mod]["history"][item["title"]] = \
list()
try:
# TODO: Re-work when NDRPDRDISC tests are not used
if table["include-tests"] == "MRR":
tbl_dict[tst_name_mod]["history"][
item["title"]].append(
tst_data["result"]["receive-rate"].avg)
elif table["include-tests"] == "PDR":
if tst_data["type"] == "PDR":
tbl_dict[tst_name_mod]["history"][
item["title"]].\
append(tst_data["throughput"]["value"])
elif tst_data["type"] == "NDRPDR":
tbl_dict[tst_name_mod]["history"][
item["title"]].append(
tst_data["throughput"]["PDR"]
["LOWER"])
elif table["include-tests"] == "NDR":
if tst_data["type"] == "NDR":
tbl_dict[tst_name_mod]["history"][
item["title"]].\
append(tst_data["throughput"]["value"])
elif tst_data["type"] == "NDRPDR":
tbl_dict[tst_name_mod]["history"][
item["title"]].append(
tst_data["throughput"]["NDR"]
["LOWER"])
else:
continue
except (TypeError, KeyError):
pass
tbl_lst = list()
for tst_name in tbl_dict.keys():
item = [
tbl_dict[tst_name]["name"],
]
if history:
if tbl_dict[tst_name].get("history", None) is not None:
for hist_data in tbl_dict[tst_name]["history"].values():
if hist_data:
item.append(round(mean(hist_data) / 1000000, 2))
item.append(round(stdev(hist_data) / 1000000, 2))
else:
item.extend([None, None])
else:
item.extend([None, None])
data_t = tbl_dict[tst_name]["ref-data"]
if data_t:
item.append(round(mean(data_t) / 1000000, 2))
item.append(round(stdev(data_t) / 1000000, 2))
else:
item.extend([None, None])
data_t = tbl_dict[tst_name]["cmp-data"]
if data_t:
item.append(round(mean(data_t) / 1000000, 2))
item.append(round(stdev(data_t) / 1000000, 2))
else:
item.extend([None, None])
if item[-4] is not None and item[-2] is not None and item[-4] != 0:
item.append(int(relative_change(float(item[-4]), float(item[-2]))))
if len(item) == len(header):
tbl_lst.append(item)
# Sort the table according to the relative change
tbl_lst.sort(key=lambda rel: rel[-1], reverse=True)
# Generate csv tables:
csv_file = "{0}.csv".format(table["output-file"])
with open(csv_file, "w") as file_handler:
file_handler.write(header_str)
for test in tbl_lst:
file_handler.write(",".join([str(item) for item in test]) + "\n")
convert_csv_to_pretty_txt(csv_file, "{0}.txt".format(table["output-file"]))
def table_performance_comparison_mrr(table, input_data):
"""Generate the table(s) with algorithm: table_performance_comparison_mrr
specified in the specification file.
:param table: Table to generate.
:param input_data: Data to process.
:type table: pandas.Series
:type input_data: InputData
"""
logging.info(" Generating the table {0} ...".format(table.get(
"title", "")))
# Transform the data
logging.info(" Creating the data set for the {0} '{1}'.".format(
table.get("type", ""), table.get("title", "")))
data = input_data.filter_data(table, continue_on_error=True)
# Prepare the header of the tables
try:
header = [
"Test case",
"{0} Throughput [Mpps]".format(table["reference"]["title"]),
"{0} stdev [Mpps]".format(table["reference"]["title"]),
"{0} Throughput [Mpps]".format(table["compare"]["title"]),
"{0} stdev [Mpps]".format(table["compare"]["title"]), "Change [%]"
]
header_str = ",".join(header) + "\n"
except (AttributeError, KeyError) as err:
logging.error(
"The model is invalid, missing parameter: {0}".format(err))
return
# Prepare data to the table:
tbl_dict = dict()
for job, builds in table["reference"]["data"].items():
for build in builds:
for tst_name, tst_data in data[job][str(build)].iteritems():
if tbl_dict.get(tst_name, None) is None:
name = "{0}-{1}".format(
tst_data["parent"].split("-")[0],
"-".join(tst_data["name"].split("-")[1:]))
tbl_dict[tst_name] = {
"name": name,
"ref-data": list(),
"cmp-data": list()
}
try:
tbl_dict[tst_name]["ref-data"].\
append(tst_data["result"]["receive-rate"].avg)
except TypeError:
pass # No data in output.xml for this test
for job, builds in table["compare"]["data"].items():
for build in builds:
for tst_name, tst_data in data[job][str(build)].iteritems():
try:
tbl_dict[tst_name]["cmp-data"].\
append(tst_data["result"]["receive-rate"].avg)
except KeyError:
pass
except TypeError:
tbl_dict.pop(tst_name, None)
tbl_lst = list()
for tst_name in tbl_dict.keys():
item = [
tbl_dict[tst_name]["name"],
]
data_t = tbl_dict[tst_name]["ref-data"]
if data_t:
item.append(round(mean(data_t) / 1000000, 2))
item.append(round(stdev(data_t) / 1000000, 2))
else:
item.extend([None, None])
data_t = tbl_dict[tst_name]["cmp-data"]
if data_t:
item.append(round(mean(data_t) / 1000000, 2))
item.append(round(stdev(data_t) / 1000000, 2))
else:
item.extend([None, None])
if item[1] is not None and item[3] is not None and item[1] != 0:
item.append(int(relative_change(float(item[1]), float(item[3]))))
if len(item) == 6:
tbl_lst.append(item)
# Sort the table according to the relative change
tbl_lst.sort(key=lambda rel: rel[-1], reverse=True)
# Generate tables:
# All tests in csv:
tbl_names = [
"{0}-1t1c-full{1}".format(table["output-file"],
table["output-file-ext"]),
"{0}-2t2c-full{1}".format(table["output-file"],
table["output-file-ext"]),
"{0}-4t4c-full{1}".format(table["output-file"],
table["output-file-ext"])
]
for file_name in tbl_names:
logging.info(" Writing file: '{0}'".format(file_name))
with open(file_name, "w") as file_handler:
file_handler.write(header_str)
for test in tbl_lst:
if file_name.split("-")[-2] in test[0]: # cores
test[0] = "-".join(test[0].split("-")[:-1])
file_handler.write(",".join([str(item)
for item in test]) + "\n")
# All tests in txt:
tbl_names_txt = [
"{0}-1t1c-full.txt".format(table["output-file"]),
"{0}-2t2c-full.txt".format(table["output-file"]),
"{0}-4t4c-full.txt".format(table["output-file"])
]
for i, txt_name in enumerate(tbl_names_txt):
logging.info(" Writing file: '{0}'".format(txt_name))
convert_csv_to_pretty_txt(tbl_names[i], txt_name)
def table_performance_comparison(table, input_data):
"""Generate the table(s) with algorithm: table_performance_comparison
specified in the specification file.
:param table: Table to generate.
:param input_data: Data to process.
:type table: pandas.Series
:type input_data: InputData
"""
logging.info(" Generating the table {0} ...".format(table.get(
"title", "")))
# Transform the data
logging.info(" Creating the data set for the {0} '{1}'.".format(
table.get("type", ""), table.get("title", "")))
data = input_data.filter_data(table, continue_on_error=True)
# Prepare the header of the tables
try:
header = [
"Test case",
]
history = table.get("history", None)
if history:
for item in history:
header.extend([
"{0} Throughput [Mpps]".format(item["title"]),
"{0} Stdev [Mpps]".format(item["title"])
])
header.extend([
"{0} Throughput [Mpps]".format(table["reference"]["title"]),
"{0} Stdev [Mpps]".format(table["reference"]["title"]),
"{0} Throughput [Mpps]".format(table["compare"]["title"]),
"{0} Stdev [Mpps]".format(table["compare"]["title"]), "Change [%]"
])
header_str = ",".join(header) + "\n"
except (AttributeError, KeyError) as err:
logging.error(
"The model is invalid, missing parameter: {0}".format(err))
return
# Prepare data to the table:
tbl_dict = dict()
for job, builds in table["reference"]["data"].items():
for build in builds:
for tst_name, tst_data in data[job][str(build)].iteritems():
if tbl_dict.get(tst_name, None) is None:
name = "{0}-{1}".format(
tst_data["parent"].split("-")[0],
"-".join(tst_data["name"].split("-")[1:]))
tbl_dict[tst_name] = {
"name": name,
"ref-data": list(),
"cmp-data": list()
}
try:
tbl_dict[tst_name]["ref-data"].\
append(tst_data["throughput"]["value"])
except TypeError:
pass # No data in output.xml for this test
for job, builds in table["compare"]["data"].items():
for build in builds:
for tst_name, tst_data in data[job][str(build)].iteritems():
try:
tbl_dict[tst_name]["cmp-data"].\
append(tst_data["throughput"]["value"])
except KeyError:
pass
except TypeError:
tbl_dict.pop(tst_name, None)
if history:
for item in history:
for job, builds in item["data"].items():
for build in builds:
for tst_name, tst_data in data[job][str(
build)].iteritems():
if tbl_dict.get(tst_name, None) is None:
continue
if tbl_dict[tst_name].get("history", None) is None:
tbl_dict[tst_name]["history"] = OrderedDict()
if tbl_dict[tst_name]["history"].get(
item["title"], None) is None:
tbl_dict[tst_name]["history"][item["title"]] = \
list()
try:
tbl_dict[tst_name]["history"][item["title"]].\
append(tst_data["throughput"]["value"])
except (TypeError, KeyError):
pass
tbl_lst = list()
for tst_name in tbl_dict.keys():
item = [
tbl_dict[tst_name]["name"],
]
if history:
if tbl_dict[tst_name].get("history", None) is not None:
for hist_data in tbl_dict[tst_name]["history"].values():
if hist_data:
item.append(round(mean(hist_data) / 1000000, 2))
item.append(round(stdev(hist_data) / 1000000, 2))
else:
item.extend([None, None])
else:
item.extend([None, None])
data_t = tbl_dict[tst_name]["ref-data"]
if data_t:
item.append(round(mean(data_t) / 1000000, 2))
item.append(round(stdev(data_t) / 1000000, 2))
else:
item.extend([None, None])
data_t = tbl_dict[tst_name]["cmp-data"]
if data_t:
item.append(round(mean(data_t) / 1000000, 2))
item.append(round(stdev(data_t) / 1000000, 2))
else:
item.extend([None, None])
if item[-4] is not None and item[-2] is not None and item[-4] != 0:
item.append(int(relative_change(float(item[-4]), float(item[-2]))))
if len(item) == len(header):
tbl_lst.append(item)
# Sort the table according to the relative change
tbl_lst.sort(key=lambda rel: rel[-1], reverse=True)
# Generate tables:
# All tests in csv:
tbl_names = [
"{0}-ndr-1t1c-full{1}".format(table["output-file"],
table["output-file-ext"]),
"{0}-ndr-2t2c-full{1}".format(table["output-file"],
table["output-file-ext"]),
"{0}-ndr-4t4c-full{1}".format(table["output-file"],
table["output-file-ext"]),
"{0}-pdr-1t1c-full{1}".format(table["output-file"],
table["output-file-ext"]),
"{0}-pdr-2t2c-full{1}".format(table["output-file"],
table["output-file-ext"]),
"{0}-pdr-4t4c-full{1}".format(table["output-file"],
table["output-file-ext"])
]
for file_name in tbl_names:
logging.info(" Writing file: '{0}'".format(file_name))
with open(file_name, "w") as file_handler:
file_handler.write(header_str)
for test in tbl_lst:
if (file_name.split("-")[-3] in test[0] and # NDR vs PDR
file_name.split("-")[-2] in test[0]): # cores
test[0] = "-".join(test[0].split("-")[:-1])
file_handler.write(",".join([str(item)
for item in test]) + "\n")
# All tests in txt:
tbl_names_txt = [
"{0}-ndr-1t1c-full.txt".format(table["output-file"]),
"{0}-ndr-2t2c-full.txt".format(table["output-file"]),
"{0}-ndr-4t4c-full.txt".format(table["output-file"]),
"{0}-pdr-1t1c-full.txt".format(table["output-file"]),
"{0}-pdr-2t2c-full.txt".format(table["output-file"]),
"{0}-pdr-4t4c-full.txt".format(table["output-file"])
]
for i, txt_name in enumerate(tbl_names_txt):
logging.info(" Writing file: '{0}'".format(txt_name))
convert_csv_to_pretty_txt(tbl_names[i], txt_name)
# Selected tests in csv:
input_file = "{0}-ndr-1t1c-full{1}".format(table["output-file"],
table["output-file-ext"])
with open(input_file, "r") as in_file:
lines = list()
for line in in_file:
lines.append(line)
output_file = "{0}-ndr-1t1c-top{1}".format(table["output-file"],
table["output-file-ext"])
logging.info(" Writing file: '{0}'".format(output_file))
with open(output_file, "w") as out_file:
out_file.write(header_str)
for i, line in enumerate(lines[1:]):
if i == table["nr-of-tests-shown"]:
break
out_file.write(line)
output_file = "{0}-ndr-1t1c-bottom{1}".format(table["output-file"],
table["output-file-ext"])
logging.info(" Writing file: '{0}'".format(output_file))
with open(output_file, "w") as out_file:
out_file.write(header_str)
for i, line in enumerate(lines[-1:0:-1]):
if i == table["nr-of-tests-shown"]:
break
out_file.write(line)
input_file = "{0}-pdr-1t1c-full{1}".format(table["output-file"],
table["output-file-ext"])
with open(input_file, "r") as in_file:
lines = list()
for line in in_file:
lines.append(line)
output_file = "{0}-pdr-1t1c-top{1}".format(table["output-file"],
table["output-file-ext"])
logging.info(" Writing file: '{0}'".format(output_file))
with open(output_file, "w") as out_file:
out_file.write(header_str)
for i, line in enumerate(lines[1:]):
if i == table["nr-of-tests-shown"]:
break
out_file.write(line)
output_file = "{0}-pdr-1t1c-bottom{1}".format(table["output-file"],
table["output-file-ext"])
logging.info(" Writing file: '{0}'".format(output_file))
with open(output_file, "w") as out_file:
out_file.write(header_str)
for i, line in enumerate(lines[-1:0:-1]):
if i == table["nr-of-tests-shown"]:
break
out_file.write(line)
# Append zero to the remaining group
if len(group_rev_cnt_list) < no_of_groups:
group_rev_cnt_list = assign_zero(class_group_list,
group_rev_cnt_list)
for row in group_rev_cnt_list: # group_rev_cnt_list is a 2D-list
group_rev_no.append(row[0])
group_rev_count.append(row[1])
# Output for this part
if is_null:
stat = "Not a single group made any revisions this week. Please consider encouraging students to contribute more actively.</p>"
else:
avg = mean(group_rev_count)
sd = stdev(group_rev_count)
stat = "In this class, the average number of revisions per group is " + str(
avg
) + ". Following is a brief analysis of weekly performance of the class.</p>"
print(
str(sum(group_rev_count)) + " " + str(avg) + " " + str(sd))
# This part is to get the Best 3 and Worst 3 groups in a class by comparing their revision counts
# No need to fetch data from db here, @group_rev_cnt_list contains all group numbers with there revision counts.
group_rev_cnt_list.sort(
key=lambda x: x[1]) #order by 2nd column @revision_count AS
num = 3
len_group_rev_cnt_list = len(group_rev_cnt_list)
if len_group_rev_cnt_list < num:
num = len_group_rev_cnt_list
xs = list()
for code in votes:
xs.append(votes[code][dim])
part.append(votes[code]['particip'])
plt.scatter(xs, part)
plt.show()
### Participation-related analysis
print "Histogram of participation diff"
diff = list()
for code in jt:
diff.append(tables[1][code]['particip'] - tables[0][code]['particip'])
m = utils.mean(diff)
st = utils.stdev(diff)
print "Mean= %f , and stdev= %f " % (m, st)
#plt.hist(diff, bins=range(-30, 30))
#plt.show()
pbound = 98
geq = lambda v: v['particip'] >= pbound
print "Tables in 7O with >%d pct particip: %d" % (pbound, len(utils.filter_by(tables[0], geq)))
print "Tables in 14A with >%d pct particip: %d" % (pbound, len(utils.filter_by(tables[1], geq)))
#delta = round(m + 3*st, 2)
delta = 10
print "Tables in which there is more than %.2f pct participation difference:" % delta
codes = utils.compare_by(
tables,