def test_melt(nulls, num_id_vars, num_value_vars, num_rows, dtype):
if dtype not in ["float32", "float64"] and nulls in ["some", "all"]:
pytest.skip(msg="nulls not supported in dtype: " + dtype)
pdf = pd.DataFrame()
id_vars = []
for i in range(num_id_vars):
colname = "id" + str(i)
data = np.random.randint(0, 26, num_rows).astype(dtype)
if nulls == "some":
idx = np.random.choice(
num_rows, size=int(num_rows / 2), replace=False
)
data[idx] = np.nan
elif nulls == "all":
data[:] = np.nan
pdf[colname] = data
id_vars.append(colname)
value_vars = []
for i in range(num_value_vars):
colname = "val" + str(i)
data = np.random.randint(0, 26, num_rows).astype(dtype)
if nulls == "some":
idx = np.random.choice(
num_rows, size=int(num_rows / 2), replace=False
)
data[idx] = np.nan
elif nulls == "all":
data[:] = np.nan
pdf[colname] = data
value_vars.append(colname)
gdf = cudf.from_pandas(pdf)
got = cudf_melt(frame=gdf, id_vars=id_vars, value_vars=value_vars)
got_from_melt_method = gdf.melt(id_vars=id_vars, value_vars=value_vars)
expect = pd.melt(frame=pdf, id_vars=id_vars, value_vars=value_vars)
# pandas' melt makes the 'variable' column of 'object' type (string)
# cuDF's melt makes it Categorical because it doesn't support strings
expect["variable"] = expect["variable"].astype("category")
assert_eq(expect, got)
assert_eq(expect, got_from_melt_method)
def test_melt(nulls, num_id_vars, num_value_vars, num_rows, dtype):
if dtype not in ['float32', 'float64'] and nulls in ['some', 'all']:
pytest.skip(msg='nulls not supported in dtype: ' + dtype)
pdf = pd.DataFrame()
id_vars = []
for i in range(num_id_vars):
colname = 'id' + str(i)
data = np.random.randint(0, 26, num_rows).astype(dtype)
if nulls == 'some':
idx = np.random.choice(num_rows,
size=int(num_rows / 2),
replace=False)
data[idx] = np.nan
elif nulls == 'all':
data[:] = np.nan
pdf[colname] = data
id_vars.append(colname)
value_vars = []
for i in range(num_value_vars):
colname = 'val' + str(i)
data = np.random.randint(0, 26, num_rows).astype(dtype)
if nulls == 'some':
idx = np.random.choice(num_rows,
size=int(num_rows / 2),
replace=False)
data[idx] = np.nan
elif nulls == 'all':
data[:] = np.nan
pdf[colname] = data
value_vars.append(colname)
gdf = DataFrame.from_pandas(pdf)
got = cudf_melt(frame=gdf, id_vars=id_vars, value_vars=value_vars)
got_from_melt_method = gdf.melt(id_vars=id_vars, value_vars=value_vars)
expect = pd.melt(frame=pdf, id_vars=id_vars, value_vars=value_vars)
# pandas' melt makes the 'variable' column of 'object' type (string)
# cuDF's melt makes it Categorical because it doesn't support strings
expect['variable'] = expect['variable'].astype('category')
pd.testing.assert_frame_equal(expect, got.to_pandas())
pd.testing.assert_frame_equal(expect, got_from_melt_method.to_pandas())