Learn DataFrames.jl

学习Julia

Julia

Published

June 19, 2023

1 导入库

using CSV
using DataFrames

2 读取数据

df = CSV.read("../data/203st_idw_interp_0.05deg.csv", DataFrame)
# df = CSV.read("I:/order_data/Xujw/203st_idw_interp_0.05deg.csv", DataFrame)
df # copy(df) 复制df

203×7 DataFrame

178 rows omitted

Row	site	lon	lat	Eh	V	Ni	Ba
	String15	Float64	Float64	Float64	Float64	Float64	Float64
1	19G26	111.8	29.07	33.2	0.08	1.55	126.99
2	22G-80	111.93	28.94	163.6	0.45	2.77	71.83
3	19G22	111.95	28.98	-48.0	0.17	1.18	161.81
4	22G-83	111.98	29.11	64.8	0.25	0.59	136.21
5	22G-81	112.03	29.04	143.1	0.97	1.83	124.54
6	20DT-19G	112.07	29.09	26.0	0.853692	1.28	173.21
7	19G21	112.14	28.87	143.0	0.06	1.16	5.54
8	20DT-17G	112.14	29.0	-60.8	0.590868	0.7	258.56
9	22G-82	112.2	29.08	96.1	0.36	1.04	183.63
10	22G-90	112.2	29.66	-88.1	0.06	1.8	167.24
11	22G-88	112.24	29.27	115.5	0.51	2.29	20.98
12	22G-85	112.26	29.18	90.5	0.310593	4.08	476.34
13	DT02G	112.26	29.59	-29.3	0.53	2.16	234.38
⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮
192	11JH031	113.65	30.17	-54.3	0.11	0.69	279.53
193	11JH161	113.69	30.09	-54.3	0.08	0.63	213.16
194	22G-67	113.7	30.15	-54.3	0.47	5.19	136.8
195	11JH057	113.7	30.39	-81.3826	0.15	0.97	457.99
196	11JH182	113.72	30.05	-54.3	0.11	0.42	119.42
197	22G-66	113.75	30.29	-97.3	0.82	4.77	332.1
198	11JH183	113.77	30.07	-54.8787	0.07	0.44	155.31
199	11JH185	113.8	30.1	-54.7215	0.36	0.81	101.32
200	11JH180	113.84	30.06	-61.5239	0.12	0.57	258.18
201	22G-68	113.87	30.14	-55.6	0.2	4.7	163.95
202	18G-40	113.95	30.17	-149.0	0.3	2.15	514.11
203	22G-69	114.0	30.13	-99.0	0.288669	1.38	0.08

3 取出某一行

3.1 方法1

df.V # df."V"

203-element Vector{Float64}:
 0.08
 0.45
 0.17
 0.25
 0.97
 0.85369167324573
 0.06
 0.590868123957716
 0.36
 0.06
 0.51
 0.310592788968676
 0.53
 ⋮
 0.11
 0.08
 0.47
 0.15
 0.11
 0.82
 0.07
 0.36
 0.12
 0.2
 0.3
 0.288668959788242

3.2 方法2

df[:, :V] # 这种情况下取值更改不会改变之前的数据框内容

203-element Vector{Float64}:
 0.08
 0.45
 0.17
 0.25
 0.97
 0.85369167324573
 0.06
 0.590868123957716
 0.36
 0.06
 0.51
 0.310592788968676
 0.53
 ⋮
 0.11
 0.08
 0.47
 0.15
 0.11
 0.82
 0.07
 0.36
 0.12
 0.2
 0.3
 0.288668959788242

4 列名

names(df, AbstractString) # 筛选出不同数据类型的列名
names(df, Float64)

6-element Vector{String}:
 "lon"
 "lat"
 "Eh"
 "V"
 "Ni"
 "Ba"

5 创建一个0行但包含全部列名的df

empty(df) # empty!()直接让目标df清空

0×7 DataFrame

Row	site	lon	lat	Eh	V	Ni	Ba
	String15	Float64	Float64	Float64	Float64	Float64	Float64

6 查看df基本信息

df维度：

size(df) # 维度
nrow(df) # ncol()

统计信息，类似R语言summary

describe(df, cols=4:7)

4×7 DataFrame

Row	variable	mean	min	median	max	nmissing	eltype
	Symbol	Float64	Float64	Float64	Float64	Int64	DataType
1	Eh	-54.8574	-278.3	-72.2	177.2	0	Float64
2	V	0.344441	0.0	0.259441	1.88	0	Float64
3	Ni	2.46611	0.0	2.14	8.39	0	Float64
4	Ba	253.259	0.01	195.07	1752.0	0	Float64

7 筛选df

df[:, [:lon, :lat]]

203×2 DataFrame

178 rows omitted

Row	lon	lat
	Float64	Float64
1	111.8	29.07
2	111.93	28.94
3	111.95	28.98
4	111.98	29.11
5	112.03	29.04
6	112.07	29.09
7	112.14	28.87
8	112.14	29.0
9	112.2	29.08
10	112.2	29.66
11	112.24	29.27
12	112.26	29.18
13	112.26	29.59
⋮	⋮	⋮
192	113.65	30.17
193	113.69	30.09
194	113.7	30.15
195	113.7	30.39
196	113.72	30.05
197	113.75	30.29
198	113.77	30.07
199	113.8	30.1
200	113.84	30.06
201	113.87	30.14
202	113.95	30.17
203	114.0	30.13

8 条件筛选

9 常规

df[:, Not([:site, :lon, :lat])]

203×4 DataFrame

178 rows omitted

Row	Eh	V	Ni	Ba
	Float64	Float64	Float64	Float64
1	33.2	0.08	1.55	126.99
2	163.6	0.45	2.77	71.83
3	-48.0	0.17	1.18	161.81
4	64.8	0.25	0.59	136.21
5	143.1	0.97	1.83	124.54
6	26.0	0.853692	1.28	173.21
7	143.0	0.06	1.16	5.54
8	-60.8	0.590868	0.7	258.56
9	96.1	0.36	1.04	183.63
10	-88.1	0.06	1.8	167.24
11	115.5	0.51	2.29	20.98
12	90.5	0.310593	4.08	476.34
13	-29.3	0.53	2.16	234.38
⋮	⋮	⋮	⋮	⋮
192	-54.3	0.11	0.69	279.53
193	-54.3	0.08	0.63	213.16
194	-54.3	0.47	5.19	136.8
195	-81.3826	0.15	0.97	457.99
196	-54.3	0.11	0.42	119.42
197	-97.3	0.82	4.77	332.1
198	-54.8787	0.07	0.44	155.31
199	-54.7215	0.36	0.81	101.32
200	-61.5239	0.12	0.57	258.18
201	-55.6	0.2	4.7	163.95
202	-149.0	0.3	2.15	514.11
203	-99.0	0.288669	1.38	0.08

df[:, Between(:site, :lat)]

203×3 DataFrame

178 rows omitted

Row	site	lon	lat
	String15	Float64	Float64
1	19G26	111.8	29.07
2	22G-80	111.93	28.94
3	19G22	111.95	28.98
4	22G-83	111.98	29.11
5	22G-81	112.03	29.04
6	20DT-19G	112.07	29.09
7	19G21	112.14	28.87
8	20DT-17G	112.14	29.0
9	22G-82	112.2	29.08
10	22G-90	112.2	29.66
11	22G-88	112.24	29.27
12	22G-85	112.26	29.18
13	DT02G	112.26	29.59
⋮	⋮	⋮	⋮
192	11JH031	113.65	30.17
193	11JH161	113.69	30.09
194	22G-67	113.7	30.15
195	11JH057	113.7	30.39
196	11JH182	113.72	30.05
197	22G-66	113.75	30.29
198	11JH183	113.77	30.07
199	11JH185	113.8	30.1
200	11JH180	113.84	30.06
201	22G-68	113.87	30.14
202	18G-40	113.95	30.17
203	22G-69	114.0	30.13

9.1 正则表达式

df[:, r"l"]

203×2 DataFrame

178 rows omitted

Row	lon	lat
	Float64	Float64
1	111.8	29.07
2	111.93	28.94
3	111.95	28.98
4	111.98	29.11
5	112.03	29.04
6	112.07	29.09
7	112.14	28.87
8	112.14	29.0
9	112.2	29.08
10	112.2	29.66
11	112.24	29.27
12	112.26	29.18
13	112.26	29.59
⋮	⋮	⋮
192	113.65	30.17
193	113.69	30.09
194	113.7	30.15
195	113.7	30.39
196	113.72	30.05
197	113.75	30.29
198	113.77	30.07
199	113.8	30.1
200	113.84	30.06
201	113.87	30.14
202	113.95	30.17
203	114.0	30.13

10 转换函数

共五种方法：

combine
select
transformselect

导入分组统计所需包

using Statistics

求V的平均值

combine(df, :V => mean => :mean_V)

1×1 DataFrame

Row	mean_V
	Float64
1	0.344441

筛选lon

select(df, :lon => mean => :mean_lon)

203×1 DataFrame

178 rows omitted

Row	mean_lon
	Float64
1	112.868
2	112.868
3	112.868
4	112.868
5	112.868
6	112.868
7	112.868
8	112.868
9	112.868
10	112.868
11	112.868
12	112.868
13	112.868
⋮	⋮
192	112.868
193	112.868
194	112.868
195	112.868
196	112.868
197	112.868
198	112.868
199	112.868
200	112.868
201	112.868
202	112.868
203	112.868

随便选

select(df, r"l", "site", :V)

203×4 DataFrame

178 rows omitted

Row	lon	lat	site	V
	Float64	Float64	String15	Float64
1	111.8	29.07	19G26	0.08
2	111.93	28.94	22G-80	0.45
3	111.95	28.98	19G22	0.17
4	111.98	29.11	22G-83	0.25
5	112.03	29.04	22G-81	0.97
6	112.07	29.09	20DT-19G	0.853692
7	112.14	28.87	19G21	0.06
8	112.14	29.0	20DT-17G	0.590868
9	112.2	29.08	22G-82	0.36
10	112.2	29.66	22G-90	0.06
11	112.24	29.27	22G-88	0.51
12	112.26	29.18	22G-85	0.310593
13	112.26	29.59	DT02G	0.53
⋮	⋮	⋮	⋮	⋮
192	113.65	30.17	11JH031	0.11
193	113.69	30.09	11JH161	0.08
194	113.7	30.15	22G-67	0.47
195	113.7	30.39	11JH057	0.15
196	113.72	30.05	11JH182	0.11
197	113.75	30.29	22G-66	0.82
198	113.77	30.07	11JH183	0.07
199	113.8	30.1	11JH185	0.36
200	113.84	30.06	11JH180	0.12
201	113.87	30.14	22G-68	0.2
202	113.95	30.17	18G-40	0.3
203	114.0	30.13	22G-69	0.288669

改列名

select(df, :lon => :new_lon, :)

203×8 DataFrame

178 rows omitted

Row	new_lon	site	lon	lat	Eh	V	Ni	Ba
	Float64	String15	Float64	Float64	Float64	Float64	Float64	Float64
1	111.8	19G26	111.8	29.07	33.2	0.08	1.55	126.99
2	111.93	22G-80	111.93	28.94	163.6	0.45	2.77	71.83
3	111.95	19G22	111.95	28.98	-48.0	0.17	1.18	161.81
4	111.98	22G-83	111.98	29.11	64.8	0.25	0.59	136.21
5	112.03	22G-81	112.03	29.04	143.1	0.97	1.83	124.54
6	112.07	20DT-19G	112.07	29.09	26.0	0.853692	1.28	173.21
7	112.14	19G21	112.14	28.87	143.0	0.06	1.16	5.54
8	112.14	20DT-17G	112.14	29.0	-60.8	0.590868	0.7	258.56
9	112.2	22G-82	112.2	29.08	96.1	0.36	1.04	183.63
10	112.2	22G-90	112.2	29.66	-88.1	0.06	1.8	167.24
11	112.24	22G-88	112.24	29.27	115.5	0.51	2.29	20.98
12	112.26	22G-85	112.26	29.18	90.5	0.310593	4.08	476.34
13	112.26	DT02G	112.26	29.59	-29.3	0.53	2.16	234.38
⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮
192	113.65	11JH031	113.65	30.17	-54.3	0.11	0.69	279.53
193	113.69	11JH161	113.69	30.09	-54.3	0.08	0.63	213.16
194	113.7	22G-67	113.7	30.15	-54.3	0.47	5.19	136.8
195	113.7	11JH057	113.7	30.39	-81.3826	0.15	0.97	457.99
196	113.72	11JH182	113.72	30.05	-54.3	0.11	0.42	119.42
197	113.75	22G-66	113.75	30.29	-97.3	0.82	4.77	332.1
198	113.77	11JH183	113.77	30.07	-54.8787	0.07	0.44	155.31
199	113.8	11JH185	113.8	30.1	-54.7215	0.36	0.81	101.32
200	113.84	11JH180	113.84	30.06	-61.5239	0.12	0.57	258.18
201	113.87	22G-68	113.87	30.14	-55.6	0.2	4.7	163.95
202	113.95	18G-40	113.95	30.17	-149.0	0.3	2.15	514.11
203	114.0	22G-69	114.0	30.13	-99.0	0.288669	1.38	0.08

11 尝试管道符

参考学习链接：Comparing dplyr vs DataFrames.jl

using Pipe

@time @pipe df |>
  combine(_, :V => (x -> x) => :V_mean, :) |>
  (df -> begin
    df.new_col = df.V
    df
  end) |>
  filter(:V_mean => v -> v >0, _) |>
  sort(_, :new_col)

  0.172564 seconds (818.78 k allocations: 40.826 MiB, 5.39% gc time, 99.71% compilation time: 1% of which was recompilation)

193×9 DataFrame

168 rows omitted

Row	V_mean	site	lon	lat	Eh	V	Ni	Ba	new_col
	Float64	String15	Float64	Float64	Float64	Float64	Float64	Float64	Float64
1	0.01	12JH093	112.85	29.94	-218.5	0.01	2.09	184.509	0.01
2	0.02	18S-25G	112.3	29.77	-88.1	0.02	0.22	112.0	0.02
3	0.04	12JH076	112.58	30.29	-90.0	0.04	2.47	170.86	0.04
4	0.04	18G-05	113.51	29.84	87.5	0.04	0.31	59.45	0.04
5	0.05	22G-86	112.43	29.39	-7.6	0.05	1.47	160.13	0.05
6	0.05	22G-73	112.81	29.42	-78.5	0.05	3.49	120.7	0.05
7	0.05	18G-11	113.26	29.61	-83.2	0.05	0.06	275.26	0.05
8	0.0558746	22G-70	113.56	29.88	-50.4	0.0558746	8.39	58.55	0.0558746
9	0.06	19G21	112.14	28.87	143.0	0.06	1.16	5.54	0.06
10	0.06	22G-90	112.2	29.66	-88.1	0.06	1.8	167.24	0.06
11	0.06	19G14	112.69	29.15	-81.0	0.06	0.99	120.31	0.06
12	0.07	18G-19	112.63	29.78	-14.3	0.07	0.45	510.65	0.07
13	0.07	18G-25	112.64	29.95	-72.0	0.07	1.31	58.46	0.07
⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮	⋮
182	0.92	22G-31	112.81	30.18	-75.2	0.92	4.45	237.92	0.92
183	0.92	22G-34	112.92	30.42	-75.9	0.92	4.13	235.87	0.92
184	0.97	22G-81	112.03	29.04	143.1	0.97	1.83	124.54	0.97
185	0.97	22G-03	112.5	29.95	-88.0	0.97	4.46	117.86	0.97
186	1.09	12JH078	112.6	30.25	-16.5	1.09	2.63	170.86	1.09
187	1.09	22G-18	112.95	29.64	-21.4	1.09	5.42	175.97	1.09
188	1.2	22G-13	112.96	29.75	-77.5	1.2	5.28	205.88	1.2
189	1.27	12JH145	113.17	29.71	-1.6	1.27	2.51	166.341	1.27
190	1.36	22G-01	112.53	29.85	-141.1	1.36	8.17	517.61	1.36
191	1.44	22G-42	112.98	30.4	-93.8	1.44	4.32	314.24	1.44
192	1.48	22G-22	112.77	29.62	125.8	1.48	1.52	51.14	1.48
193	1.88	22G-91	112.39	29.66	177.2	1.88	4.19	164.61	1.88