How to import data from attatched csv file?

Question

Prabhanshu Chandra 2020-8-12

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/578535-how-to-import-data-from-attatched-csv-file

评论： Jeremy Hughes 2020-8-12

采纳的回答： Cris LaPierre

data.csv

In the attatched csv file there is a lot of clutter other than data, first 11 lines are not needed, so I used

T=readtable('data.csv','HeaderLines',11);

but in the data also, every data point has its unit GHz attached to it, which I dont want to import into my array, how can I stop that GHz from being imported. What more lines of code needs to be added. For ex: output from above readtable function is :

Var1 Var2

_____________ ______

{'1.000 GHz'} 200

{'1.100 GHz'} 200

{'1.200 GHz'} 200

{'1.300 GHz'} 200

{'1.400 GHz'} 200

{'1.500 GHz'} 200

{'1.600 GHz'} 199.99

{'1.700 GHz'} 199.99

{'1.800 GHz'} 199.99

{'1.900 GHz'} 199.99

{'2.000 GHz'} 199.99

It is taking var1 as string because of the presence of GHz, please someone help me remove that. Also explain how can I name change names of these var1 and var2 by using appropritae argument is readtable function.

I noticed another problem, in the var2 column the data is till 3 decimal places, but imported data is automatically rounded off to 2 decimal place, i don't want it to be rounded off. How to avoid that?

Thankyou in advance.

2 个评论
显示无隐藏无

Cris LaPierre 2020-8-12

There's no attached csv file.

Prabhanshu Chandra 2020-8-12

So sorry Chris, I added the question again and forgot to add to this one, now its attached.

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Cris LaPierre 2020-8-12

1
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/578535-how-to-import-data-from-attatched-csv-file#answer_478737

编辑：Cris LaPierre 2020-8-12

在 MATLAB Online 中打开

You could try using the Format name-value pair. You might need to read about it in textscan for the specifics of how it works. This allows you to specify the format of the delimited values in each row.

Space gets treated as a delimiter here. Since there is a space between the number and GHz, it can be read in independent of the number and actually be ignored. Since we don't have your csv, I can't test it for your data, but try somethign like this.

T=readtable('data.csv','HeaderLines',11,"Format",'%f%*s%f')

From the sample data you shared, this is the result I get.

T = 
    Var1     Var2 
    ____    ______
      200
1        200
2        200
3        200
4        200
5        200
6     199.99
7     199.99
8     199.99
9     199.99
   199.99

5 个评论
显示 3更早的评论隐藏 3更早的评论

Cris LaPierre 2020-8-12

在 MATLAB Online 中打开

Have you tested it with the actual data file? Remember, I didn't have the data file when I ran the sample code. I'd made up the data using what you showed in your original post, which was 200 for the first 6. When I rerun the exact same code on data.csv, I see internally MATLAB has read in the data correctly.

Inspect your variable in the Variable Editor (double click on the variable name in the Workspace to open it in the Variable Editor). When I do, I see desired resolution. It is only rounding the display.

               199.998000000000
10000000000000	199.997000000000
20000000000000	199.997000000000
30000000000000	199.996000000000
40000000000000	199.995000000000
50000000000000	199.995000000000
60000000000000	199.994000000000
70000000000000	199.993000000000
80000000000000	199.992000000000
90000000000000	199.992000000000
               199.991000000000
10000000000000	199.990000000000
20000000000000	199.989000000000
...

To change the varable names, you could do the following:

T5.Properties.VariableNames={'newVar1','newVar2'}

Cris LaPierre 2020-8-12

在 MATLAB Online 中打开

Glad to help. One more follow up. It's only rounding with you display the entire table. Try displaying just var2 and it will print without rounding.

Prabhanshu Chandra 2020-8-12

Oh Okay, Thankyou so much.

请先登录，再进行评论。

Answer 2

Jeremy Hughes 2020-8-12

1
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/578535-how-to-import-data-from-attatched-csv-file#answer_478824

在 MATLAB Online 中打开

In R2020a,

T = readtable('data.csv','TrimNonNumeric',true,'NumHeaderLines',11);

or in earlier releases,

opts = detectImportOptions('data.csv','TrimNonNumeric',true,'NumHeaderLines',11);
T = readtable('data.csv',opts);

5 个评论
显示 3更早的评论隐藏 3更早的评论

Prabhanshu Chandra 2020-8-12

Or maybe combined with TrimNonNumeric its acting that way.

Since we have already trimmed non numeric part we dont have to trim header again, we just have to trim numeric header. Thats why by using 1 as argument it works fine, because there only 1 numeric line in header part of my csv file.

Jeremy Hughes 2020-8-12

在 MATLAB Online 中打开

Oh, I see, never mind. Not a bug exactly--just a quirk of CSV data. The first row begins with a double quote character.

CSV spec says that should be one field, containing the newlines and all... so there is only one line being interpreted by the parser. The first "row" considering quotes looks like this

"freq\nReferences : ['R']\nDependency : [freq]\nNum. Points : [91]\nMatrix Size : scalar\nType : Real", ...
"R\nReferences : []\nDependency : [freq]\nNum. Points : [91]\nMatrix Size : scalar\nType : Real"

If you leave off the NumHeaderLines it treats this as a row, and stripes away everything but the 91. and I see:

T = readtable('data.csv','TrimNonNumeric',true); head(T)
ans =
  8×2 table
    Var1     Var2 
    ____    ______
     91         91
      1        200
    1.1        200
    1.2        200
    1.3        200
    1.4        200
    1.5        200
    1.6     199.99

请先登录，再进行评论。

How to import data from attatched csv file?

2 个评论
显示无隐藏无

采纳的回答

5 个评论
显示 3更早的评论隐藏 3更早的评论

更多回答（1 个）

5 个评论
显示 3更早的评论隐藏 3更早的评论

另请参阅

类别

标签

产品

版本

Community Treasure Hunt

How to import data from attatched csv file?

2 个评论 显示 无隐藏 无

采纳的回答

5 个评论 显示 3更早的评论隐藏 3更早的评论

更多回答（1 个）

5 个评论 显示 3更早的评论隐藏 3更早的评论

另请参阅

类别

标签

产品

版本

Community Treasure Hunt

2 个评论
显示无隐藏无

5 个评论
显示 3更早的评论隐藏 3更早的评论

5 个评论
显示 3更早的评论隐藏 3更早的评论