def test():
"""
Update操作是，将新DataFrame中凡是跟原DataFrame的Index和Column有重复的位置，只要
值不是 np.nan, 就用它取代原DataFrame中的值。
**此操作只考虑原DataFrame中有的Index和Column**。
"""
df1 = pd.DataFrame({"v": [1, 1]}, index=[0, 1])
"""
v
0 1
1 1
"""
df2 = pd.DataFrame({"v": [2, 2]}, index=[1, 2])
"""
v
1 2
2 2
"""
df = pd.DataFrame({"v": [0, 0, 0]}, index=[0, 1, 2])
df.update(df1)
assert_value_equal(df.v, [1, 1, 0])
"""
v
0 1
1 1
2 0
"""
df.update(df2)
assert_value_equal(df.v, [1, 2, 2])
"""
def test_sort_index():
df = pd.DataFrame(
{"v": [2, 3, 1]},
index=[2, 3, 1],
)
df = df.sort_index(axis=0)
assert_value_equal(df.index, [1, 2, 3])
assert_value_equal(df.v, [1, 2, 3])
def test_timeseries():
m, n = 2, 3
data = [[1, 2, 3], [4, 5, 6]]
df = pd.DataFrame(
data,
index=pd.date_range("2017-01-01", periods=m, freq="D"),
columns=list("ABC")
)
assert_value_equal(df, data)
def test_multiple_row():
# select second and third rows
row23 = df.iloc[1:3]
assert_value_equal(row23, [[4, 5, 6], [7, 8, 9]])
assert isinstance(row23, pd.DataFrame)
# select second and third rows
row23 = df.loc[2:3]
assert_value_equal(row23, [[4, 5, 6], [7, 8, 9]])
assert isinstance(row23, pd.DataFrame)
def test_specified_row():
# select second row, use .iloc (Integer Index Only)
row2 = df.iloc[1] # the seconds index
assert_value_equal(row2, [4, 5, 6])
assert isinstance(row2, pd.Series)
# select second row, use .loc (Value Index)
row2 = df.loc[2] # the index value is 2
assert_value_equal(row2, [4, 5, 6])
assert isinstance(row2, pd.Series)
def test_head_tail():
data = [[1, 2], [3, 4]]
df = pd.DataFrame(data)
assert_value_equal(df.head(1), [
[1, 2],
])
assert_value_equal(df.tail(1), [
[3, 4],
])
def test_dropna():
data = [[1, 2, 3], [None, 5, 6], [7, 8, 9]]
df = pd.DataFrame(
data,
index=[1, 2, 3],
columns=list("ABC"),
)
res = df.dropna(axis=0) # by row
assert_value_equal(res, [[1, 2, 3], [7, 8, 9]])
"""
A B C
1 1 2 3
3 7 8 9
"""
res = df.dropna(axis=1) # by column
assert_value_equal(res, [[2, 3], [5, 6], [8, 9]])
"""
def test_sort_index():
data = [
[1, 1, 1],
[1, 2, 2],
[1, 3, 3],
[2, 1, 4],
[2, 2, 5],
[2, 3, 6],
[3, 1, 7],
[3, 2, 8],
[3, 3, 9],
]
columns = ["k1", "k2", "v"]
random.shuffle(data)
df = pd.DataFrame(data, columns=columns)
res = df.sort_values([
"k1",
])
assert_value_equal(res.k1, [1, 1, 1, 2, 2, 2, 3, 3, 3])
"""
k1 k2 v
1 3 3
1 1 1
1 2 2
2 2 5
2 1 4
2 3 6
3 1 7
3 2 8
3 3 9
"""
res = df.sort_values([
"k2",
])
assert_value_equal(res.k2, [1, 1, 1, 2, 2, 2, 3, 3, 3])
"""
k1 k2 v
2 1 4
3 1 7
1 1 1
1 2 2
3 2 8
2 2 5
1 3 3
3 3 9
2 3 6
"""
res = df.sort_values(["k1", "k2"])
assert_value_equal(res.k1, [1, 1, 1, 2, 2, 2, 3, 3, 3])
assert_value_equal(res.k2, [1, 2, 3, 1, 2, 3, 1, 2, 3])
"""
def test_sort_column():
df = pd.DataFrame({
"b": [
2,
],
"c": [
3,
],
"a": [
1,
]
})
df.sort_index(axis=1)
assert_value_equal(df.columns, ["a", "b", "c"])
assert_value_equal(df.a, [
1,
])
assert_value_equal(df.b, [
2,
])
assert_value_equal(df.c, [
3,
])
def test():
df = pd.DataFrame([
[1, 2, None],
[None, 5, 6],
[7, 8, 9],
])
assert_value_equal(df.isnull(), [
[False, False, True],
[True, False, False],
[False, False, False],
])
# any na value
assert df.isnull().values.any() == True
# for each column, any na in each column
assert_value_equal(df.isnull().any(axis=0), [True, False, True])
# for each row, any na in each row
assert_value_equal(df.isnull().any(axis=1), [True, True, False])
def test():
"""
将两个Column相同的DataFrame连接起来，对于有相同的Index，使用 前面/后面 的哪一个。
"""
df1 = pd.DataFrame({"v": [1, 1]}, index=[0, 1])
"""
v
0 1
1 1
"""
df2 = pd.DataFrame({"v": [2, 2]}, index=[1, 2])
"""
v
1 2
2 2
"""
df = pd.concat([df1, df2])
assert_value_equal(df.v, [1, 1, 2, 2])
"""
v
0 1
1 1
1 2
2 2
"""
# Remove Duplicate Index, keep df1's
res = df[~df.index.duplicated(keep="first")]
assert_value_equal(res.v, [1, 1, 2])
"""
v
0 1
1 1
2 2
"""
# Remove Duplicate Index, keep df2's
res = df[~df.index.duplicated(keep="last")]
assert_value_equal(res.v, [1, 2, 2])
"""
def test_from_records():
data = [[1, 2], [3, 4]]
df = pd.DataFrame(data, columns=["c1", "c2"], index=["r1", "r2"])
assert_value_equal(df, data)
def test_from_row_list():
data = [{"c1": 1, "c2": 2}, {"c1": 3, "c2": 4}]
df = pd.DataFrame(data, columns=["c1", "c2"], index=["r1", "r2"])
assert_value_equal(df, [[1, 2], [3, 4]])
def test_from_dict():
data = {"c1": [1, 3], "c2": [2, 4]}
df = pd.DataFrame(data, columns=["c1", "c2"], index=["r1", "r2"])
assert_value_equal(df, [[1, 2], [3, 4]])