def test_unused_columns():
df = results.get_results(r,
row_types=["itervar"],
omit_unused_columns=False)
_assert_sequential_index(df)
# two replications of three measurements of a single experiment, and 19 columns in total
return df.shape == (6, 19)
def test_row_type_filter_2():
filtered = results.read_result_files(RESULT_FILES, "run =~ *General-0* AND module =~ Test.node1 AND name =~ foo1*")
df = results.get_results(filtered, row_types=["scalar", "attr"])
_assert_sequential_index(df)
# 2 times 3 rows for scalars (incl. value), and 3 times 2 rows for the vector, stats, and histogram (only attr)
# since we only filtered for row types, not result types, we get the attrs for the other kinds of results too, just not the results themselves
return df.shape == (12, 7)
def test_vector_time_limit_at_load_2():
filtered = results.read_result_files(
RESULT_FILES,
"type =~ vector AND run =~ General-0*",
vector_end_time=50.0)
df = results.get_results(filtered, row_types=["vector"])
_assert_sequential_index(df)
return df["vectime"].map(lambda a: (a < 50.0).all()).all()
def test_vectors_start_end_time_at_load():
trimmed = results.read_result_files(RESULT_FILES,
vector_start_time=40,
vector_end_time=60)
df = results.get_vectors(trimmed)
_assert_sequential_index(df)
_assert(sanitize_and_compare_csv(df, "vectors_start_end_time.csv"),
"content mismatch")
def test_runs_without_config_entries():
df = results.get_runs(r,
include_itervars=True,
include_runattrs=True,
include_param_assignments=True,
include_config_entries=False)
_assert_sequential_index(df)
_assert(sanitize_and_compare_csv(df, "runs_without_config_entries.csv"),
"content mismatch")
def test_statistics_with_all():
df = results.get_statistics(r,
include_attrs=True,
include_itervars=True,
include_runattrs=True,
include_param_assignments=True,
include_config_entries=True)
_assert_sequential_index(df)
_assert(sanitize_and_compare_csv(df, "statistics_with_all.csv"),
"content mismatch")
def test_param_assignments_with_all():
df = results.get_param_assignments(r,
include_itervars=True,
include_runattrs=True,
include_param_assignments=True,
include_config_entries=True)
_assert_sequential_index(df)
_assert(sanitize_and_compare_csv(df, "param_assignments_with_all.csv"),
"content mismatch")
_assert(
df.apply(lambda r: r[r["name"]] == r["value"], axis=1).all(),
"wrong join")
def test_parameters_with_attrs():
df = results.get_parameters(r, include_attrs=True)
_assert_sequential_index(df)
# these parameters don't have any attrs
_assert(sanitize_and_compare_csv(df, "parameters.csv"), "content mismatch")
def test_parameters():
df = results.get_parameters(r)
_assert_sequential_index(df)
_assert(sanitize_and_compare_csv(df, "parameters.csv"), "content mismatch")
def test_statistics_with_attrs():
df = results.get_statistics(r, include_attrs=True)
_assert_sequential_index(df)
_assert(sanitize_and_compare_csv(df, "statistics_with_attrs.csv"),
"content mismatch")
def test_runs_with_config_entries():
df = results.get_runs(r, include_config_entries=True)
_assert_sequential_index(df)
_assert(sanitize_and_compare_csv(df, "runs_with_config_entries.csv"),
"content mismatch")
def test_vectors_end_time():
trimmed = results.read_result_files(RESULT_FILES)
df = results.get_vectors(trimmed, end_time=80)
_assert_sequential_index(df)
_assert(sanitize_and_compare_csv(df, "vectors_end_time.csv"),
"content mismatch")
def test_scalars_with_param_assignments():
df = results.get_scalars(r, include_param_assignments=True)
_assert_sequential_index(df)
_assert(sanitize_and_compare_csv(df, "scalars_with_param_assignments.csv"),
"content mismatch")
def test_vector_data():
filtered = results.read_result_files(
RESULT_FILES, "type =~ vector AND run =~ General-0*")
df = results.get_results(filtered, row_types=["vector"])
_assert_sequential_index(df)
return df["vectime"].map(lambda a: a.shape == (100, )).all()
def test_row_type_filter_3():
filtered = results.read_result_files(RESULT_FILES, "type =~ param")
df = results.get_results(filtered, row_types=["attr"])
_assert_sequential_index(df)
# params don't have attrs
return df.empty
def test_row_type_filter_1():
df = results.get_results(r, row_types=["scalar"])
_assert_sequential_index(df)
# two recorded values from two sources of two submodules in all six runs
return df.shape == (48, 5)
def test_result_filter():
filtered = results.read_result_files(RESULT_FILES, "type =~ scalar")
df = results.get_results(filtered)
_assert_sequential_index(df)
# in all 6 runs: 20 lines of metadata, and 4 lines (1 scalar and 3 attrs) for all 8 scalars
return df.shape == (312, 7)
def test_itervar_count():
df = results.get_results(r, row_types=["itervar"])
_assert_sequential_index(df)
return df["type"].map(lambda t: t == "itervar").all() and df.shape == (6,
4)
def test_config_count():
df = results.get_results(r, row_types=["config"])
_assert_sequential_index(df)
return df["type"].map(lambda t: t == "config").all() and df.shape == (18,
4)
