[TVM Discuss] [RFC] Canonicalizing AutoTVM Log Format

2020-06-19 Thread An Wang via TVM Discuss
@jroesch @merrymercy @tqchen Here are my own takes: 2. I prefer the more readable but longer format. 3. I prefer to remove `kwargs` as it's unused for cleanliness. 4. I'm unsure how useful the idea of `code_hash` is for developers currently. 5. I think "Log" is more canonical in the tvm com

[TVM Discuss] [RFC] Canonicalizing AutoTVM Log Format

2020-06-19 Thread An Wang via TVM Discuss
@mdw-octoml I don't think there's currently enough interest to justify adding Protobuf as a dependency in TVM. TVM users are used to readable json for their autotvm logs. If there is more interest from the broader community, we can revisit this. @tqchen I feel we may not need to re-discuss th

[TVM Discuss] [RFC] Canonicalizing AutoTVM Log Format

2020-06-22 Thread An Wang via TVM Discuss
My thoughts are that the suggested change to add Python structure shouldn't necessarily depend on what the log format will look like, so I don't think there is a need to wait for the Ansor log format. (I imagine Ansor coders have their hands full, and that they'd prefer to consider polish late

[TVM Discuss] [RFC] Canonicalizing AutoTVM Log Format

2020-06-23 Thread An Wang via TVM Discuss
I've thought about this some more, and I'm changing my stance with respect to ProtoBuf. While adding a Python class schema is a less invasive change than introducing ProtoBuf and allows us to stick to the current log format exactly, protos do have the added benefit of being language-neutral. A

[TVM Discuss] [RFC] Canonicalizing AutoTVM Log Format

2020-06-24 Thread An Wang via TVM Discuss
I think the main benefit of keeping the ProtoBuf opaque is avoiding the unnecessary effort of fleshing out a schema that will change very soon. However, since I have a full specification described here already, I prefer to go ahead with it, unless there other concerns I have missed. I sugges

[TVM Discuss] [RFC] Canonicalizing AutoTVM Log Format

2020-06-29 Thread An Wang via TVM Discuss
addressing @mdw-octoml's points: - I will add a comment addressing the semantics of the dtype field in the proto. - I will further refine the spec to avoid Any. I originally included google.protobuf.Any to capture the current tuple argument semantics, which seemingly supports arbitrary nesting

[TVM Discuss] [RFC] Canonicalizing AutoTVM Log Format

2020-06-29 Thread An Wang via TVM Discuss
@comaniac ~~I will change `workload` to `task`.~~ Since ansor is not op-based, I think it makes sense to keep the `workload` syntax to prepare for ansor's log format changes. I agree that the list-based representation of arguments is less than ideal -- currently it's hard to understand the

[TVM Discuss] [RFC] Canonicalizing AutoTVM Log Format

2020-06-29 Thread An Wang via TVM Discuss
I see. Thanks for clarifying @comaniac, I agree with your comments. Addressing @merrymercy's points: - One possible solution to the redundancy of repeating items such as target string would be to encode something like this: `message AutoTVMLogs{ string target; repeated AutoTVMLog; ...} ` whe