Outputs "NaN" after "DataFrame columns" function?

Outputs "NaN" after "DataFrame columns" function? - Printable Version

+- Python Forum (https://python-forum.io)
+-- Forum: Python Coding (https://python-forum.io/forum-7.html)
+--- Forum: Data Science (https://python-forum.io/forum-44.html)
+--- Thread: Outputs "NaN" after "DataFrame columns" function? (/thread-32145.html)

Outputs "NaN" after "DataFrame columns" function? - epsilon - Jan-23-2021

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from scipy.signal import argrelextrema

data = pd.read_csv("Data/EURUSD60.csv", delimiter="\t")

print(data)

data = pd.DataFrame(data=data, columns=["Date", "Open", "High", "Low", "Close", "Volume"])

print(data)

and output is:

I'm trying some strategies and calculations on prices. Previously I was working on windows (I was using excel for the python file) then I installed manjaro and encountered such problem with the csv file.

RE: Outputs "NaN" after "DataFrame columns" function? - Larz60+ - Jan-23-2021

NaN is missing data or None (can be used interchangeably).

RE: Outputs "NaN" after "DataFrame columns" function? - epsilon - Jan-24-2021

(Jan-23-2021, 10:53 PM)Larz60+ Wrote: NaN is missing data or None (can be used interchangeably).

Yes I know, but why ? :)As you can see in the picture, when there is data after the first "read_csv", NaN returns after naming the columns.

RE: Outputs "NaN" after "DataFrame columns" function? - Larz60+ - Jan-24-2021

First change line 10 to:
df = pd.DataFrame(data=data, columns=["Date", "Open", "High", "Low", "Close", "Volume"])
you were overwriting 'data'.

Then, to isolate the data, please change line 8 to:
print(f"\nCSV data:\n{data}")

and line 12 to:
print(f"\ndataframe:\n{df}")

Show results again

RE: Outputs "NaN" after "DataFrame columns" function? - epsilon - Jan-26-2021

(Jan-24-2021, 06:35 PM)Larz60+ Wrote: First change line 10 to:
df = pd.DataFrame(data=data, columns=["Date", "Open", "High", "Low", "Close", "Volume"])
you were overwriting 'data'.

Then, to isolate the data, please change line 8 to:
print(f"\nCSV data:\n{data}")

and line 12 to:
print(f"\ndataframe:\n{df}")

Show results again

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from scipy.signal import argrelextrema

data = pd.read_csv("Data/EURUSD60.csv", delimiter="\t")

print(f"\nCSV data:\n{data}")

df = pd.DataFrame(data=data, columns=["Date", "Open", "High", "Low", "Close", "Volume"])

print(f"\ndataframe:\n{df}")

I did, but the result is the same as in the picture

RE: Outputs "NaN" after "DataFrame columns" function? - Larz60+ - Jan-26-2021

The new output should contain the two print statements, which will isolate the output to the different sections of software. That was what I was interested in seeing.

If you don't see the print statements, something else is happening here.

RE: Outputs "NaN" after "DataFrame columns" function? - buran - Jan-26-2021

Looking at the screenshot, note that initial data dataframe does not have these columns. Confirmed by 99 rows x6 columns (index is 0 to 98)
index 0 is 2012-12-14 18:00, and the column names are values for 2012-12-24 17:00

I guess your csv file has no header and you need to specify columns when reading it.
Can you show your file (e.g. top 10 rows of it?)

or try

data = pd.read_csv("Data/EURUSD60.csv", delimiter="\t", names=["Date", "Open", "High", "Low", "Close", "Volume"])

RE: Outputs "NaN" after "DataFrame columns" function? - epsilon - Jan-27-2021

(Jan-26-2021, 07:19 PM)buran Wrote: Looking at the screenshot, note that initial data dataframe does not have these columns. Confirmed by 99 rows x6 columns (index is 0 to 98)
index 0 is 2012-12-14 18:00, and the column names are values for 2012-12-24 17:00

I guess your csv file has no header and you need to specify columns when reading it.
Can you show your file (e.g. top 10 rows of it?)

or try
data = pd.read_csv("Data/EURUSD60.csv", delimiter="\t", names=["Date", "Open", "High", "Low", "Close", "Volume"])

Yes, there is no title in the csv-file, as you said, I added a header(name=[..]) in "read_csv" and it's okay.
Thanks!