Yusuke Minami

icon indicating a website link https://www.linkedin.com/in/yusukeminami/

icon company Shopee Singapore icon location Singapore icon location Algorithm Engineer (Formerly Data Scientist) in Singapore. Author of ML Python packages: PipelineX (Pipeline for eXperimentation) and CausalLift (marketing).

Results 29 comments of Yusuke Minami

ModuleNotFoundError: No module named 'kedro.io.transformers'

Hi @Peccer , Thank you for reporting the issue. As you pointed out, PipelineX does not support kedro 0.18.x yet. Could you install kedro 0.17.x ? ``` pip install kedro==0.17.7...

Add option to skip enumeration by element types

Sure, sounds good to me. Could you please implement it?

Add option to skip enumeration by element types

Thank you! I look forward to it.

Recover deleted cells

Thank you! I look forward to it.

Recover deleted cells

Worked in v2.0.1! Thanks @joouha !

ModuleNotFoundError: No module named 'kedro.io.transformers'

I released pipelinex 0.7.6 which does not import modules missing in kedro 0.18.x to avoid the error.

[FR] join 2 CSV files

Thank you, @dcmoura ! Proper JOINs would not be easy from scratch, but I would suggest using pandas like this: ```python import pandas as pd left_df = pd.read_csv(left_file, dtype=str, keep_default_na=False)...

[FR] join 2 CSV files

Currently, I use pandas to join CSV files, which means I need to prepare and manage Python scripts. If spyql supports joining CSV files, I can save more time.

[FR] join 2 CSV files

Thank you for your suggestion, @dcmoura ! 1) Hmmm, unfortunately, that seems a bit complicated to me. Hope syntax similar to SQL can be used, as supported by `clickhouse local`...

[FR] join 2 CSV files

Advantages of SPyQL over clickhouse local for my use case are: 1. SpyQL can be installed easily by pip 2. SpyQL supports Python objects/functions (e.g. pathlib.Path)