Pandas object could not convert string to float-CodePudding

I got some depth image data in csv format.

I want to convert the data into float type so I can reshape them to image

and I get this error

my code is like this (I excluded the variables that includes my directory)

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt


# The name of the column I want to convert is "data"
df = pd.read_csv(img)
dat_float = df['data'].apply(lambda x: x.replace(',','').replace('[','').replace(']','')).astype('float')
print(dat_float)

My dataframe is like this

since this is hard to see so I pull out the data I need

The target to change to float from object is the data (right one to the 'seq')

CodePudding user response：

import pandas as pd

# create sample data
data = "0;[0,0,13,3,2,251,2,252]"
df = pd.read_csv(io.StringIO(data), sep=";", header=None)
df.columns = ['seq', 'data']

# convert object to list
df.data=df.data.apply(lambda s: [float(x.strip(' []')) for x in s.split(',')])

# optional: explode to rows
df.explode('data')

CodePudding user response：

For the part that you need to turn your string into float, here's how you can do it. I just make myself a dataframe with a list of strings.

data = {'seq': [1,2],
        'data':[['0','0','0','0','0'],['0','0','0','0','0']]}
df=pd.DataFrame(data)

   seq             data
0    1  [0, 0, 0, 0, 0]
1    2  [0, 0, 0, 0, 0]

And by using list comprehension, you can get what you need.

df['data']=[[float(x) for x in i] for i in df['data']]

   seq                       data
0    1  [0.0, 0.0, 0.0, 0.0, 0.0]
1    2  [0.0, 0.0, 0.0, 0.0, 0.0]