GTG Lightening Talk
Summit Suen
GO
Data
Science
？
implementation of DataFrame
It is the thing that makes R and Python so important in the field of Data Science.
什麼
是DataFrame
[ ]
<int>
<bool>
<float>
<string>
其實
就是
表格啦
github.com/kniren/gota
I/O
df := dataframe.New(
series.New([]string{"b", "a"}, series.String, "COL.1"),
series.New([]int{1, 2}, series.Int, "COL.2"),
series.New([]float64{3.0, 4.0}, series.Float, "COL.3"),
)I/O
df := dataframe.LoadRecords(
[][]string{
[]string{"A", "B", "C", "D"},
[]string{"a", "4", "5.1", "true"},
[]string{"k", "5", "7.0", "true"},
[]string{"k", "4", "6.0", "true"},
[]string{"a", "2", "7.1", "false"},
},
)
fmt.Println(df)
[4x4] DataFrame
A B C D
0: a 4 5.100000 true
1: k 5 7.000000 true
2: k 4 6.000000 true
3: a 2 7.100000 false
<string> <int> <float> <bool>I/O
jsonStr := `[{"COL.2":1,"COL.3":3},
{"COL.1":5,"COL.2":2,"COL.3":2},
{"COL.1":6,"COL.2":3,"COL.3":1}]`
df := dataframe.ReadJSON(strings.NewReader(jsonStr))
fmt.Println(df)
[3x3] DataFrame
COL.1 COL.2 COL.3
0: NaN 1 3
1: 5 2 2
2: 6 3 1
<int> <int> <int>Subset
// Row selection
sub := df.Subset([]int{0, 2})
[2x4] DataFrame
A B C D
0: a 4 5.100000 true
2: k 5 7.000000 true
<string> <int> <float> <bool>
// Column selection
sel := df.Select([]string{"A", "C"})
[4x4] DataFrame
A C
0: a 5.100000
1: k 7.000000
2: k 6.000000
3: a 7.100000
<string> <float> Filter
fil := df.Filter(
dataframe.F{"A", series.Eq, "a"},
dataframe.F{"B", series.Greater, 4},
)
[3x4] DataFrame
A B C D
0: a 4 5.100000 true
1: k 5 7.000000 true
2: a 2 7.100000 false
<string> <int> <float> <bool>
fil2 := fil.Filter(
dataframe.F{"D", series.Eq, true},
)
[2x4] DataFrame
A B C D
0: a 4 5.100000 true
1: k 5 7.000000 true
<string> <int> <float> <bool>Apply
https://github.com/kniren/gota/issues/13
Benchmark
https://github.com/kniren/gota/issues/16
// storing the elements of a Series from []Element to Element
type intElement struct {
e *int
}
type intElement struct {
e int
nan bool
}Benchmark
https://github.com/kniren/gota/issues/16
And⋯⋯?
-
Mutate
-
Arrange
-
Join
-
Chaining
-
Split/Group (developing)
-
Combine (developing)