SHJ

sort_value issue might be due to the python version mismatch, please modify accordingly. Data integration and subsequent usage in sj_rsi_nifty100_zipline.ipynb are 2 different things. For this particular blog, I ingested...

Hi @Marigold, Appreciate your response however I am still facing the issue. Please see the below error. ``` LINK : fatal error LNK1181: cannot open input file 'm.lib' error: command...

It is already there however still the error. ``` (base) C:\windows\system32>pip install cvxopt Requirement already satisfied: cvxopt in c:\programdata\anaconda3\lib\site-packages (1.3.0) (base) C:\windows\system32> ```

@Marigold, Any idea? Appreciate your help. Thanks

Many thanks!! I could able to install however it is conflicting with seaborn==0.11.2, Below is the code snippet regarding conflicts. ``` Installing collected packages: seaborn Attempting uninstall: seaborn Found existing...

After you updated cvxopt 1.3.0, I install universal-portfolios and that downgraded some of the packages such as seaborn. I upgraded these packages afterward and it works. Thanks!

Thank you!! adding `solver_params `and `scale_objective, scale_constraints` helps however it is slow. You may want to check out implementation here if that helps to improve speed - https://github.com/dcajasn/Riskfolio-Lib/blob/master/examples/Tutorial%2037%20-%20OWA%20Portfolio%20Optimization.ipynb

Thank you for the response! In Riskfolio when you use `rm='GMD'` with `owa `weights it is faster. ``` owa_w = rp.owa_gmd(len(X)) w = port.owa_optimization(obj=obj,owa_w=owa_w,rf=rf,l=l) ```

hi, any update on this as I am also facing a similar issue? thanks

Please convert the index of returns data frame usinig tz_localize('Etc/UCT') ``` portfolio_returns.index = pd.to_datetime(portfolio_returns.index).tz_localize('utc') out_of_sample = portfolio_returns.index[-100] pf.create_full_tear_sheet(portfolio_returns, live_start_date=out_of_sample) ```