OverflowError: cannot convert float infinity to integer about quantstats HOT 22 CLOSED

encountering the same issue. Can kindly explain how you resolved it? thanks

from quantstats.

Try cleaning up your data before using it, using:

import numpy as np
import quantstats as qs

your_data = your_data.replace([np.inf, -np.inf, -0], 0)

qs.plots.snapshot(your_data)

Let me know if that works out 😉

from quantstats.

Not worked.

There are no infinity values in the data

from quantstats.

Can you re-upload your data? I seem to be corrupted...

from quantstats.

https://drive.google.com/open?id=1XN3wfOiFFGVJop6SkephKzUeUKjhqK6H

from quantstats.

Upgrade to the latest version and try again.

$ pip install quantstats -U --no-cache-dir

from quantstats.

Still same error

from quantstats.

Running this code works ok for me:

import pandas as pd
import quantstats as qs

df = pd.read_csv('test.csv', names=['date', 'price'], 
                 index_col=['date'], parse_dates=['date'])[100:]
df['returns'] = df['price'].pct_change().fillna(0)

qs.plots.snapshot(df['returns'])

Result is below:

from quantstats.

If you're using jupyter - try reloading retarting the notebook kernel and import the libraries again to make sure you're using the latest version.

from quantstats.

Not working in: qs.reports.full()

from quantstats.

Can you post the entire code you're using? I've tested every method and eveything works.

from quantstats.

`df = pd.read_csv('test.csv', names=['date', 'price'], index_col=['date'], parse_dates=['date'])[100:]
df['returns'] = df['price'].pct_change().fillna(0)

qs.plots.snapshot(df['returns']) # Works

qs.reports.full(df['returns'] , "SPY") # return error`

ValueError: cannot convert float NaN to integer

from quantstats.

Can you share which version of pandas and numpy are you using?

If you check the picture it won't able to plot Rolling Beta to Benchmark

from quantstats.

Upgrade to the latest version and try again. Again :)

$ pip install quantstats -U --no-cache-dir

from quantstats.

I upgraded again. Restart the jupyter notebook and still the same error.

By error it looks there is some error in plot_rolling_beta function

from quantstats.

mmin = min([-100, int(beta.min()*100)])
In this line beta.min() returning NaN so it cannot convert NaN to integer. Can you please check?

from quantstats.

from quantstats.

I run the pip install quantstats -U --no-cache-dir

still its 0.0.19

from quantstats.

I fixed it out. Thank you very much :)

from quantstats.

👍🏻

from quantstats.

👍🏻

from quantstats.

BTW, the error still occurs, I'm investigating further and will provide more details when I have them.

from quantstats.