def test_heavy(self):
# Sync all S3 objects
from beep.utils.memprof import cache_all_kitware_data
cache_all_kitware_data()
paths = os.listdir(os.path.join(S3_CACHE, "D3Batt_Data_publication"))
paths = [path for path in paths if not "Metadata" in path]
paths = [
os.path.join(S3_CACHE, "D3Batt_Data_publication", path)
for path in paths
]
# Run validation on everything
v = SimpleValidator()
validate_record = v.validate_from_paths(paths,
record_results=True,
skip_existing=True)
df = pd.DataFrame(v.validation_records)
df = df.transpose()
print(df)
print("{} valid, {} invalid".format(
len([x for x in df.validated if x]),
len([x for x in df.validated if not x])))
invalid_runs = df[np.logical_not(df.validated)]
print(invalid_runs)
self.assertEqual(len(invalid_runs), 3)
def test_project_name(self):
with ScratchDir("."):
os.environ["BEEP_PROCESSING_DIR"] = os.getcwd()
os.mkdir("data-share")
os.mkdir(os.path.join("data-share", "validation"))
v = SimpleValidator()
paths = [
"FastCharge_000000_CH29.csv",
"FastCharge_000025_CH8.csv",
"PredictionDiagnostics_000151_test.052",
]
paths = [os.path.join(TEST_FILE_DIR, path) for path in paths]
validate_record = v.validate_from_paths(paths,
record_results=True,
skip_existing=False)
self.assertEqual(
validate_record["FastCharge_000000_CH29.csv"]["method"],
"schema-arbin-lfp.yaml",
)
self.assertEqual(
validate_record["FastCharge_000025_CH8.csv"]["method"],
"schema-arbin-lfp.yaml",
)
self.assertEqual(
validate_record["PredictionDiagnostics_000151_test.052"]
["method"],
"schema-maccor-2170.yaml",
)
def test_file_incomplete(self):
path = "FastCharge_000025_CH8.csv"
path = os.path.join(TEST_FILE_DIR, path)
v = SimpleValidator()
df = pd.read_csv(path, index_col=0)
validity, reason = v.validate(df)
self.assertFalse(validity)
self.assertEqual(
reason, "cycle_index needs to reach at least 1 "
"for processing, instead found:value=0.0")
def test_validation_maccor(self):
path = "xTESLADIAG_000019_CH70.070"
path = os.path.join(TEST_FILE_DIR, path)
v = SimpleValidator(schema_filename=os.path.join(
VALIDATION_SCHEMA_DIR, "schema-maccor-2170.yaml"))
v.allow_unknown = True
header = pd.read_csv(path, delimiter='\t', nrows=0)
df = pd.read_csv(path, delimiter='\t', skiprows=1)
df['State'] = df['State'].astype(str)
df['current'] = df['Amps']
validity, reason = v.validate(df)
self.assertTrue(validity)
def test_validate_from_paths_maccor(self):
paths = [os.path.join(TEST_FILE_DIR, "xTESLADIAG_000019_CH70.070")]
# Run validation on everything
v = SimpleValidator()
validate_record = v.validate_from_paths(paths,
record_results=True,
skip_existing=False)
df = pd.DataFrame(v.validation_records)
df = df.transpose()
self.assertEqual(df.loc["xTESLADIAG_000019_CH70.070", "method"],
"simple_maccor")
self.assertEqual(df.loc["xTESLADIAG_000019_CH70.070", "validated"],
True)
def test_invalidation_maccor(self):
path = "PredictionDiagnostics_000109_tztest.010"
path = os.path.join(TEST_FILE_DIR, path)
v = SimpleValidator(schema_filename=os.path.join(
VALIDATION_SCHEMA_DIR, "schema-maccor-2170.yaml"))
v.allow_unknown = True
header = pd.read_csv(path, delimiter='\t', nrows=0)
print(header)
df = pd.read_csv(path, delimiter='\t', skiprows=1)
df['State'] = df['State'].astype(str)
df['current'] = df['Amps']
print(df.dtypes)
validity, reason = v.validate(df)
print(validity, reason)
self.assertFalse(validity)
def test_monotonic_cycle_index_maccor(self):
path = "PredictionDiagnostics_000109_tztest.010"
path = os.path.join(TEST_FILE_DIR, path)
v = SimpleValidator(schema_filename=os.path.join(
VALIDATION_SCHEMA_DIR, "schema-maccor-2170.yaml"))
v.allow_unknown = True
header = pd.read_csv(path, delimiter="\t", nrows=0)
print(header)
df = pd.read_csv(path, delimiter="\t", skiprows=1)
df["State"] = df["State"].astype(str)
df["current"] = df["Amps"]
df.loc[df["Cyc#"] == 89, "Cyc#"] = 0
validity, reason = v.validate(df)
self.assertFalse(validity)
self.assertEqual(
reason, "cyc# needs to be monotonically increasing for processing")
def test_basic(self):
path = "2017-05-09_test-TC-contact_CH33.csv"
path = os.path.join(TEST_FILE_DIR, path)
v = SimpleValidator()
df = pd.read_csv(path, index_col=0)
validity, reason = v.validate(df)
self.assertFalse(validity)
self.assertEqual(
reason, "Column cycle_index: integer type check failed "
"at index 0 with value nan")
# Test bigger file, with float/numeric type
path = "2017-08-14_8C-5per_3_47C_CH44.csv"
path = os.path.join(TEST_FILE_DIR, path)
v = SimpleValidator()
df = pd.read_csv(path, index_col=0)
v.schema['cycle_index']['schema']['type'] = 'float'
validity, reason = v.validate(df)
self.assertTrue(validity)
self.assertEqual(reason, '')
v.schema['cycle_index']['schema']['type'] = 'numeric'
validity, reason = v.validate(df)
self.assertTrue(validity)
self.assertEqual(reason, '')
# Test good file
path = "2017-12-04_4_65C-69per_6C_CH29.csv"
path = os.path.join(TEST_FILE_DIR, path)
v = SimpleValidator()
df = pd.read_csv(path, index_col=0)
validity, reason = v.validate(df)
self.assertTrue(validity)
self.assertEqual(reason, '')
# Alter the schema on-the-fly to induce error
v.schema['discharge_capacity']['schema']['max'] = 1.8
validity, reason = v.validate(df)
self.assertFalse(validity)
# Cut off reasons to prevent floating format discrepancies
self.assertEqual(
reason[:80], "discharge_capacity is higher than allowed max 1.8 "
"at index 11154: value=1.801418")
# Alter the schema on-the-fly to move on to the next errors
v.schema['discharge_capacity']['schema']['max'] = 2.1
v.schema['step_time'] = {"schema": {"min": 0.0, "type": "float"}}
validity, reason = v.validate(df)
self.assertFalse(validity)
self.assertEqual(
reason[:70], 'step_time is lower than allowed min 0.0 '
'at index 104416:value=-450.945')
# Alter schema once more to recover validation
del v.schema['step_time']['schema']['min']
validity, reason = v.validate(df)
self.assertTrue(validity)
self.assertEqual(reason, '')
