Liu Bao

> `|` refers to bitwise or. Try using `or`: > > ``` > print(numexpr.evaluate("10 or (a >> 5)")) > ``` > > Does that solve your problem? This code works....

The issue is caused by my incorrect usage. But, the document and the error are misleading: https://numexpr.readthedocs.io/projects/NumExpr3/en/latest/user_guide.html#supported-operators NumExpr supports the set of operators listed below: - Logical operators: &, |,...

Just clone the repo, open & run the `pycode_similar.py` in pycharm.

Sorry for the incorrect document. The usage should be: ``` python import pycode_similar pycode_similar.detect([data[0], data[0]],diff_method=pycode_similar.UnifiedDiff) ```

The `ExecutionInfo.result()` does not have to call `asyncio.wrap_future`. I will fix it.

Thanks for reporting this bug. Currently, the advanced third-party frameworks of mars learn (the libs in `/mars/learn/contrib/`) are not supported by ray backend. You can try the mars backend or...

> Can we add other scheduler like `Slurm` here using execution API? @fyrestone It seems that this is a deployment problem, not an execution issue.

@chaokunyang @wjsi It seems that the `tblib` has compatibility issue with `RayError`?

Could you try this? ```python import ray import time import mars import mars.tensor as mt import mars.dataframe as md def ray_init(): while True: try: print(ray.init(address='xx.xx.xx.xx:9999', log_to_driver=True, ignore_reinit_error=True, )) break except...

> @fyrestone Thanks for reply. I've tried the new code and the **'pid'** error seems to be fixed. But another error raises at `print(df.describe().execute())` > > ``` > Traceback (most...