search

jealous / stockstats

Supply a wrapper ``StockDataFrame`` based on the ``pandas.DataFrame`` with inline stock statistics/indicators support.
Other
1.29k stars 299 forks source link

copy should not modify data #124

Open neuraldevelopment opened 2 years ago

neuraldevelopment commented 2 years ago

Hi there,

thanks for your wonderful lib, really like it.

When using your lib together with other frameworks I run into the problem that a copy call in the other framework caused the column names to be renamed to lower case, which the other framework can't handle. I scanned your code and found that the copy function has this "feature" in it.

As I think a copy should not modify the data compared to the original one I introduced a lowerCase-Flag in wrap and retype causing copy to leave the column names as they are. With that modification I was able to run your StockDataFrame also with other frameworks like backtesting.py

Best regards, Neutro2

neuraldevelopment commented 2 years ago

... and sorry for the two commits, I forgot to delete a print type message and had already pushed my repo...

jealous commented 2 years ago

Looks like there are some format issues in the code. I change the column name to lower case because the calculation of some indicators requires specific columns like 'close' or 'high', 'low'. I am not sure if they still working when you have columns like 'Close' or 'High'.

neuraldevelopment commented 2 years ago

Hi Cedric,

Could understand when converting column names to lowercase, but on other side there are some libs that require capitalized names when using them. Really bad that there is no standard for the naming issue for the columns.

Best from Bavaria, Neutro2

From: Cedric Zhuang @.> Reply to: jealous/stockstats @.> Date: Tuesday, 4. October 2022 at 08:01 To: jealous/stockstats @.> Cc: Neutro2 @.>, Author @.***> Subject: Re: [jealous/stockstats] copy should not modify data (PR #124)

Looks like there are some format issues in the code. I change the column name to lower case because the calculation of some indicators requires specific columns like 'close' or 'high', 'low'. I am not sure if they still working when you have columns like 'Close' or 'High'.

— Reply to this email directly, view it on GitHubhttps://github.com/jealous/stockstats/pull/124#issuecomment-1266433875, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AZI6FFQRDVNVMJL4272VHNDWBPB2RANCNFSM6AAAAAAQ3WTI7U. You are receiving this because you authored the thread.Message ID: @.***>