Ying Zhang
Ying Zhang
I passed in a set, the error message is "columns not found" which confused me for a while, until I realize I have to pass a list. Raising on the...
tmp1 = rt.Dataset({'f': [1,1,2,2], 'g': list('abcd'), 'h': [5,6,7,8]}) rt.putmask(tmp1['h'], tmp1.f == 1, tmp1.groupby(['f'])['h'].shift(-1)) (this gives me a segmentation fault)
This is not necessarily a bug, but it really surprised me as a user. Create a test dataset of 80Million rows, one column of float64, ds = rt.Dataset({'f' : [1.0...
```python import riptable as rt rt.__version__ ``` '1.0.54' ```python ds_tmp = rt.Dataset({'f': [1,2,3,4]}) ds_tmp2 = rt.Dataset({'date': 20210904, 'symbol': 'AAPL'}) struct_tmp = rt.Struct({'data': ds_tmp, 'data2': ds_tmp2}) ``` ```python struct_tmp ``` #Name...
```python rt.__version__ ``` '1.0.54' ```python tmp1 = rt.Dataset({'f': [1, np.nan], 'g': [2,3]}) tmp2 = rt.Dataset({'f': [np.nan], 'h': ['100']}) ``` ```python tmp1.merge2(tmp2, on='f') ``` # fgh01.000021nan3 [2 rows x 3 columns]...
when np.nan show up as a bound, I don't have a strong opinion whether "nothing happens" or we should "nullify the value". But it would be good if there is...
```python tmp = pd.DataFrame({'f': [1,2,3,4], 'g':[5,6,7,8]}) ``` ```python tmp.to_csv('test.csv', sep='\t', index=False) ``` ```python ## version 4 uses pandas, but, the 'sep' argument is lost tmp2 = rt.load_csv_as_dataset('test.csv', version=4, sep='\t') ```...
riptable version 1.0.54. 1. Wen you compare a col of type int and another col of type S-str, you get a FuturesWarnings. This is (not great but) good. ("Great" would...
I am surprised that boolean array doesn't support missing values. It is in numpy as well as riptable. (I was trying to take a stab at a home-brewed version of...