Yes, the 5th column worked.
Just to show what Peter meant:
julia> using DataFrames, CSV
julia> data = """
1511805.54010775,-3631401.61267572,134,14.0385102411102,1.58792636637882
1511880.54010775,-3631401.61267572,135,14.0623467964249,1.63997535275912
1511955.54010775,-3631401.61267572,136,13.8401810265915,1.48364309144577
1511730.54010775,-3631476.61267572,248,14.3080113191324,1.8129535268525
1511805.54010775,-3631476.61267572,249,14.2710283891669,1.78019664985144
1511880.54010775,-3631476.61267572,250,14.2895778872987,1.79291534883922
1511955.54010775,-3631476.61267572,251,13.9846463218381,1.55426381397442
1512030.54010775,-3631476.61267572,252,13.1978960888864,1.12702376308933
1512105.54010775,-3631476.61267572,253,12.0766192975667,0.867366055025797
"""
"1511805.54010775,-3631401.61267572,134,14.0385102411102,1.58792636637882\n1511880.54010775,-3631401.61267572,135,14.0623467964249,1.63997535275912\n1511955.54010775,-3631401.61267572,136,13.8401810265915,1.48364309144577\n1511730.54010775,-3631476.61267572,248,14.3080113191324,1.8129535268525\n1511805.54010775,-3631476.61267572,249,14.2710283891669,1.78019664985144\n1511880.54010775,-3631476.61267572,250,14.2895778872987,1.79291534883922\n1511955.54010775,-3631476.61267572,251,13.9846463218381,1.55426381397442\n1512030.54010775,-3631476.61267572,252,13.1978960888864,1.12702376308933\n1512105.54010775,-3631476.61267572,253,12.0766192975667,0.867366055025797\n"
julia> CSV.read(IOBuffer(data), DataFrame; delim = ",", header = false)
9×5 DataFrame
Row │ Column1 Column2 Column3 Column4 Column5
│ Float64 Float64 Int64 Float64 Float64
─────┼───────────────────────────────────────────────────
1 │ 1.51181e6 -3.6314e6 134 14.0385 1.58793
2 │ 1.51188e6 -3.6314e6 135 14.0623 1.63998
3 │ 1.51196e6 -3.6314e6 136 13.8402 1.48364
4 │ 1.51173e6 -3.63148e6 248 14.308 1.81295
5 │ 1.51181e6 -3.63148e6 249 14.271 1.7802
6 │ 1.51188e6 -3.63148e6 250 14.2896 1.79292
7 │ 1.51196e6 -3.63148e6 251 13.9846 1.55426
8 │ 1.51203e6 -3.63148e6 252 13.1979 1.12702
9 │ 1.51211e6 -3.63148e6 253 12.0766 0.867366
Here I’m just reading the data from an IOBuffer, but equally I can write this out to a file and read back in from disk:
# df is the DataFrame I read in above
julia> CSV.write("test.csv", df)
"test.csv"
julia> CSV.read("test.csv", DataFrame; delim = ",")
# Same result as above (although the file written to disk has a header now, so I removed header=false here
In my function readdata() I have:
df = DataFrame(a= [1, 2], b = [3, 4])
CSV.write(“file.csv”, df)
df_new = CSV(“file.csv”, DataFrame)
Data = CSV.read(filename, DataFrame; delim=‘,’, header=false)
When I run this I get a lot of text from ‘abstractdataframe.jl’ where one line is highligjhted:
“syntax df[column] is not supported use df[!, column] instead.”
‘filename’ refers to the datafile I still want to import into a DataFrame called ‘Data’
Please make sure to enclose your code in triple backticks ``` to ensure it’s legible.
The error you are reporting cannot be caused by the lines of code you posted, as you are not indexing a DataFrame. This is how to get the error:
julia> df = DataFrame(rand(2, 2), :auto)
2×2 DataFrame
Row │ x1 x2
│ Float64 Float64
─────┼────────────────────
1 │ 0.600275 0.241023
2 │ 0.542208 0.758831
julia> df[:x1]
ERROR: ArgumentError: syntax df[column] is not supported use df[!, column] instead
which is not what you’re doing in your code. When I run your code I get:
julia> df_new = CSV("file.csv", DataFrame)
ERROR: MethodError: objects of type Module are not callable
from the third line of what you posted - you are doing CSV() which is trying to use CSV (the name of a module) as a function.
I think you have much more fundamental problems then how many columns you are reading in, your earlier posts in this thread about getting comments back from your code, and the error you just posted which is unconnected to your code suggest that you’re not really doing what you think you’re doing.
2 Likes
Thank you Nils,
The first three line of the code that I posted were suggested by somebody else of this forum. I do not understand what their purpose is. From what you say I understand it’s better to dselete them.
When I do that I get the message again that “syntax df[column] is not supported use df[!, column instead]”.
Could that be caused by the next line in my code:
x = Data[:1]
I’ve been using this with julia-1.0
Yes. That is an expected error.
Someone above mentioned that this is an expected error already.
My code is now:
Data = CSV.read(filename, DataFrame; delim=',', header =false)
df = Data
x = df[!, 1]
y = df[!, 2]
# nr = df[!, 3] grid point identifier
z_pred = df[!, 4]
s2 = df[!, 5]
N = length(x)
println("Grid size : ", N)
And it WORKED. What a relief!
I tghank you guys for all your patience and help.
1 Like
well done 