serving

What happened + What you expected to happen

When initializing a Ray Trainer, we provide a logdir argument, and the __init__ method of the Trainer stores it as a logdir class variable.

Then, when creating a Trainable with Trainer.to_tune_trainable(), it in-turn calls _create_tune_trainable(), which does not use self.logdir. So when tune_function is defined inside `_create_tu

... to make it easier to read Vespa documentation on an e-reader / offline

Vespa documentation is generated using Jekyll from .md and .html files, look into options for generating the artifact as part of site generation (there might be plugins we can use here)

The documentation for creating complex inference graphs should include what components can be connected to what, and how the overall graph should behave.

For example, all inference graphs need to end with a single node, whether its a combiner combining multiple inputs, or a transformer, or a model--the graph can't split and then never rejoin, etc.

I've noticed that there is an error in the code:
https://github.com/pytorch/serve/blob/a4d5090e114cdbeddf5077a817a8cd02d129159e/ts/model_loader.py#L92

There is an attempt to read the MANIFEST.json file, and there is use of the / charecter which used as a seperator in the Linux os, and in Windows there is use of the back-slash \, so this line creates always an invalid path so the loader d

在尝试本地服务化部署的时候，日志抛出数组越界异常

代码：

midped_batch = []
for idx in range(len(feed_batch)):
    predict_res, error_code, error_info = func_timeout.func_timeout(
        self._timeout,
        self.process,
        args=([feed_batch[idx]], typical_logid))
    #midped_batch[idx].append

The format of current master branch UDF output, eva-reuse and my SSD object detector is different. Need a sync format.