Introducing parallel variant of add_all_ta_features.

[such3r]

Suggest simple process-based all-set indicators calculation parallelizing. Gives ~2x speedup compared to the initial implementation.

[Harsh-Maheshwari]

@such3r
Getting this error when running

line 609, in _job
    instance = globals()[obj](**{k: target[v] if k in ohlcv_keys else v for k, v in kwargs.items()})
TypeError: __init__() got an unexpected keyword argument 'close'

The above exception was the direct cause of the following exception:

TypeError                                 Traceback (most recent call last)
Cell In[4], line 1
----> 1 _ = add_all_ta_features_parallel(df,open="open", high="high", low="low", close="close", volume="volume", fillna=False)

File ~/utils/parallel_ta.py:1068, in add_all_ta_features_parallel(df, open, high, low, close, volume, fillna, colprefix, vectorized, n_jobs)
    964     init += [
    965         # Money Flow Indicator
    966         (
   (...)
   1063         ),
   1064     ]
   1066 columns_order = df.columns.tolist() + [column for *_, jobs in init for column, _ in jobs]
-> 1068 results = Parallel(n_jobs=n_jobs)(delayed(_job)(df, *args) for args in init)
   1070 for columns in results:
   1071     for column, values in columns.items():

[such3r]

@Harsh-Maheshwari, could you, please, provide the input data you used?