qs.reports.html(), ValueError: cannot convert float NaN to integer

[nistrup]

Hi, I'm experiencing problems simply using the example from the README to make a report, ie.:

%matplotlib inline
import quantstats as qs

# extend pandas functionality with metrics, etc.
qs.extend_pandas()

# fetch the daily returns for a stock
stock = qs.utils.download_returns('FB')

qs.reports.html(stock, "SPY)

This results in

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-112-5ce6901da6d9> in <module>
----> 1 qs.reports.html(stock, benchmark)

~/WORKSPACE/venv/lib/python3.8/site-packages/quantstats/reports.py in html(returns, benchmark, rf, grayscale, title, output, compounded)
     56     tpl = tpl.replace('{{v}}', __version__)
     57 
---> 58     mtrx = metrics(returns=returns, benchmark=benchmark,
     59                    rf=rf, display=False, mode='full',
     60                    sep=True, internal="True",

~/WORKSPACE/venv/lib/python3.8/site-packages/quantstats/reports.py in metrics(returns, benchmark, rf, display, mode, sep, compounded, **kwargs)
    297 
    298     # return df
--> 299     dd = _calc_dd(df, display=(display or "internal" in kwargs))
    300 
    301     metrics = _pd.DataFrame()

~/WORKSPACE/venv/lib/python3.8/site-packages/quantstats/reports.py in _calc_dd(df, display)
    573                 by='max drawdown', ascending=True
    574             )['max drawdown'].values[0] / pct,
--> 575             'Longest DD Days': str(round(ret_dd.sort_values(
    576                 by='days', ascending=False)['days'].values[0])),
    577             'Avg. Drawdown %': ret_dd['max drawdown'].mean() / pct,

ValueError: cannot convert float NaN to integer

Any help would be much appreciated.

[orshemtov]

I've encountered the same issue

[TradeProphet]

same here

[Onepamopa]

And how long are the backtests you guys running? Cause I've not seen any such errors but my tests are >10 years.

[TradeProphet]

Less than 1 year, I can send you the returns for testing in case you are interested Alon

…

On Thu, Dec 31, 2020, 10:57 Onepamopa ***@***.***> wrote: And how long are the backtests you guys running? Cause I've not seen any such errors but my tests are >10 years. — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#58 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AIGPXPYQRK42PXFTNFHKU6LSXQ4HTANCNFSM4VKZFLWQ> .

[gmjohns]

has anyone found a solution to this?

[TradeProphet]

Developed my own tearsheet, took one day, works like a charm

…

On Mon, Jan 4, 2021, 23:49 Garrett Johnson ***@***.***> wrote: has anyone found a solution to this? — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#58 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AIGPXPZZANOJTVFRA7YIQ7LSYIZXJANCNFSM4VKZFLWQ> .

[ranaroussi]

Hi @nistrup - I can't seem to reproduce this. Can you please post the returns csv here?

Thanks!

[Onepamopa]

@ranaroussi it does occur, so does the "Inf" problem as well, I had to adjust several functions and replace Inf & so forth with 0.
stats.py rolling_greeks:
result = _pd.DataFrame(index=returns.index, data={
"beta": beta,
"alpha": alpha
}).fillna(0)
result = result.replace([_np.inf, -_np.inf, -0], 0) # Replace any Inf's with 0 (taken from to_drawdown_series())
return result

stats.py ulcer_index:
dd = dd.replace([_np.inf, -_np.inf, -0], 0) # Replace any Inf's with 0 (taken from to_drawdown_series())

result = _np.sqrt(_np.divide((dd**2).sum(), returns.shape[0] - 1))
return result

The problem is not with _prepare_returns - I checked - it doesn't have any NaN or Inf, but they appear after some calculation (placed in a new dataframe).

[Onepamopa]

I can post the returns dataframe (as sent to quantstats), if that'll help.
I get it from a "cash_market" analyzer in backtrader and convert it as:
rets = self.analyzers.getbyname(self.testId).get_analysis()
#self.log.debug(rets)
df_values = pd.DataFrame(rets).T # .T = transpose
df_values = df_values.iloc[:, 1]
returns = qs.utils.to_returns(df_values, rf=0.)
returns.index = pd.to_datetime(returns.index)

returns is then sent to qs.reports.html(returns=returns, . . . . . . .)

[ranaroussi]

@Onepamopa - It will help :) Please post the returns dataframe here. Thanks!

[TradeProphet]

Guys, leave me out of the correspondence please, thanks

…

On Tue, Jan 5, 2021, 14:03 Ran Aroussi ***@***.***> wrote: @Onepamopa <https://github.com/Onepamopa> - It will help :) Please post the returns dataframe here. Thanks! — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#58 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AIGPXPZX6SHREXTMIWFULJ3SYL5ZDANCNFSM4VKZFLWQ> .

[Onepamopa]

@ranaroussi Here's the dataframe after calling to_returns()
df.txt

[gmjohns]

@ranaroussi I'll include mine as well. This is obviously before converting to returns:
nlv.txt

[ranaroussi]

Please upgrade to 0.0.26 by running pip install quantstats -U and let me know if at least some of the problems were fixed. I was able to create a tearsheet with your data without problems using the new verison

[Onepamopa]

I'll check tomorrow and let you know.

[gmjohns]

for me the latest version fixed the error stated in this original issue.... still not producing a report tho:

   3209             if (isinstance(converted_limit, Real)
   3210                     and not np.isfinite(converted_limit)):
-> 3211                 raise ValueError("Axis limits cannot be NaN or Inf")
   3212             return converted_limit
   3213 

ValueError: Axis limits cannot be NaN or Inf

I'll try to debug this, could be something simple

[oudingfan]

for me the latest version fixed the error stated in this original issue.... still not producing a report tho:

   3209             if (isinstance(converted_limit, Real)
   3210                     and not np.isfinite(converted_limit)):
-> 3211                 raise ValueError("Axis limits cannot be NaN or Inf")
   3212             return converted_limit
   3213 

ValueError: Axis limits cannot be NaN or Inf

I'll try to debug this, could be something simple

quantstats/quantstats/_plotting/core.py

Line 89 in a76c1e3

if isinstance(benchmark, _pd.Series):

Add benchmark = _utils.group_returns(benchmark, benchmark.index.year, True) after this line works for me.