Assistant Wang

By the way, kubernetes seems to support GPU management via Device plugins https://kubernetes.io/docs/tasks/manage-gpus/scheduling-gpus/#using-device-plugins. So why the gpu mode is not supported in kubernetes? Is it due to lack of standards,...

另，本题的推导证明过程不是非常严谨，正在重新推导中

您好，您的理解无误。另外，钢的定义在题目中已经写明，不是碳含量小于等于t%，而是碳含量等于t%。虽然根据公式解出将生铁炼成钢需要负数时间显然不合理，不能作为答案，然而当碳含量小于t%的情况，根据不同的诠释方式（开放世界诠释与封闭世界诠释），确实认为是钢和不是钢都是合理的，即本题在解出炼钢时间

Just for debugging, as far as I know, if you use `script` directive, snakemake will generate a self-contained script (`snakemake` object injected) at `.snakemake/script` (let's say `tmpxxx.yourscript.py`) and simply run...

If you really want the `snakemake` object when run script directly, there's a really hackish work-around: In your `yourscript.py` script, find and match script like `.snakemake/script/tmpxxx.{__file__}.py`, pick one you like...

> If you really want the `snakemake` object when run script directly, there's a really hackish work-around: > > In your `yourscript.py` script, find and match script like `.snakemake/script/tmpxxx.{__file__}.py`, pick...

and `from torch.utils.data import default_collate` raise an `ImportError` ``` ImportError: cannot import name 'default_collate' from 'torch.utils.data' ``` but `from torch.utils.data._utils.collate import default_collate` works. For pytorch 1.10. (no documention for `default_collate`...

Add unittest to reproduce and simply remove __reduce__ to fix it.