+1
--
Reply to this email directly or view it on GitHub:
https://github.com/apache/tvm/issues/10471#issuecomment-1058491019
You are receiving this because you are subscribed to this thread.
Message ID:
I spent a lot of time optimizing the sort/argsort kernel for GPUs, we get
pretty good performance on GPUs from multiple vendors that competes with those
vendor's hand tuned libraries.
If these TIR kernels are well optimized, they shouldn't end up being the
bottleneck in models.
---
[Visi
TE is a limited declarative programming model, it's not possible to write
operations that do data-dependent indexing in TE.
Anything that's sort/scatter related needs to be written directly in the more
imperative TIR.
---
[Visit
Topic](https://discuss.tvm.apache.org/t/autoscheduler-do-we
+1
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/apache/tvm/issues/7991#issuecomment-833699197
I've been looking at the PR and some of the discussion, and I thought I'd bring
my thoughts back jto this RFC, it seems like a better place for broader design
thoughts.
First, thanks for the RFC, @ziheng. There is definitely wy too much
boilerplate in TVM right now, and finding ways to st
Agreed, 100%. I've been using a `yapf` configuration file I got from @comaniac,
but I'm happy to standardize.
---
[Visit
Topic](https://discuss.tvm.apache.org/t/rfc-introduce-automatic-formatting-of-python-code/7843/2)
to respond.
You are receiving this because you enabled mailing list m
Thanks, @icemelon9 and the community!
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/apache/incubator-tvm/pull/5886#issuecomment-647736567
https://github.com/apache/incubator-tvm/pull/5826
---
[Visit Topic](https://discuss.tvm.ai/t/dynamic-ops-in-relay/6909/16) to respond.
You are receiving this because you enabled mailing list mode.
To unsubscribe from these emails, [click
here](https://discuss.tvm.ai/email/unsubscribe/5e76
I agree with @tqchen that making `float` throw a warning is a good option,
it's an ambiguous declaration, even in C, it means different things on
different systems and in different contexts. Being more precise is probably
better.
---
[Visit
Topic](https://discuss.tvm.ai/t/discuss-the-me
@kevinthesun @tqchen Are you guys agreeing to A1?
@lixiaoquan I haven't put much thought into dynamic rank. I'm not sure how we
would produce it with the current opset, do you have any use cases in mind?
---
[Visit Topic](https://discuss.tvm.ai/t/dynamic-ops-in-relay/6909/8) to respond.
Y
Frameworks and state of the art models are moving more and more toward
dynamism, where the shapes of tensors in a model are calculated at runtime,
either from the shapes of inputs or from the values of inputs.
There are a number of efforts underway in TVM to better support dynamic models,
inc
:) Want to open that POC as a PR?
---
[Visit Topic](https://discuss.tvm.ai/t/patternlang-match-constant-nodes/6835/4)
to respond.
You are receiving this because you enabled mailing list mode.
To unsubscribe from these emails, [click
here](https://discuss.tvm.ai/email/unsubscribe/2c7ad226
Hi Cody,
There are some examples in the unit tests of matching const nodes with specific
values:
https://github.com/apache/incubator-tvm/blob/a072da0588c542757d2815832b7f010f530b2428/tests/python/relay/test_dataflow_pattern.py#L685-L759
Adding ConstantPattern with an optional value is probabl
@comaniac @matt-arm @masahi
Sorry for the delay! The Meetup this morning ate more of my time than I
expected.
https://github.com/apache/incubator-tvm/pull/5646
---
[Visit
Topic](https://discuss.tvm.ai/t/byoc-use-pattern-language-to-create-composite-functions/6727/28)
to respond.
You
Hmm. Okay. I'll add a check callback to the partition pass.
I think that leaves us with three levels of possible user control in level of
priority:
1. Use the Pattern Language for any constraints possible. We might be able to
add shape checking to the language, we can't use an external codegen
@comaniac @matt-arm
I've extending the AttrPattern to handle attributes of CallNodes and
FunctionNodes here: https://github.com/apache/incubator-tvm/pull/5637
Would you mind taking a look and letting me know if you think that would be
sufficiently powerful for solution B?
---
[Visit
Top
I see. This is easily doable with the Rewriter API, since you can do an
arbitrary callback, but it doesn't have the niceties of the Parition pass.
The current AttrPattern is only matching Op attributes, but that's fairly easy
to extend.
I think I prefer doing this with the AttrPattern, it's a
[quote="matt-arm, post:13, topic:6727"]
Looking forward slightly, I do wonder whether it will start making sense to
instead of creating composite functions just directly insert the annotations
around the match. This would give us an opportunity to link up ‘composite’
matches with single ops as
[quote="matt-arm, post:13, topic:6727"]
Do you envisage this having a similar API to the existing pass? In particular
it’s important for us to retain the added ‘check’ function to confirm whether a
match is valid.
[/quote]
@matt-arm Can you point me at the API? I have something like this inte
We'd still need something like this:
`use pattern language to match fused_conv, and let codegen decide whether it is
conv+bias or conv+bias+relu when the codegen is parsing a composite function
with fused_conv attribute.`
But it would be analyzing strings instead of graphs.
---
[Visit
T
While I wasn't paying attention to the discussion, I implemented Cody's
original ask here:
https://github.com/apache/incubator-tvm/pull/5627
@masahi, what do you think of stringing together the types of all of the
nodes/names of all of the called ops in the partitioned function, in
topologic
The currently implementation wont, thinking over it more, but I'll work on an
update that adds that functionality and a user-specified set of attributes
---
[Visit
Topic](https://discuss.tvm.ai/t/byoc-use-pattern-language-to-create-composite-functions/6727/4)
to respond.
You are receivin
I'd be happy to extend the parition pass to accept an optional map of
attributes. I think it's a relatively small change to the pass, if you guys
think that would be the easiest option.
Something I thought of over the last few days: If some patterns in your list
are larger (say conv-bias-relu
In particular:
1. The Dominator pattern adds some fuzziness to the pattern matching that is
non-standard for graph pattern matchers. It provides a way to find something
like a convolution op, who's output is then used by several elementwise ops,
that are then combined together. This kind of fe
@yzhliu I've hit a couple of issue that would be completely solved by this
change :). Any chance you've made progress? If not, would you be offended if I
gave it a shot over the next few weeks?
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or
Hi All,
Thanks for all of the comments. The PR is complete, I've added the Partitioner
we were planning on doing. Since some of the API's (especially around the
dominator pattern) are somewhat novel, I'd appreciate anyone who can take a few
minutes to look over the PR.
Matthew
---
[Visi
I think the PR is ready for review as V1 of the pattern language, it contains
some documentation, testing, and the language itself, the matcher, and
pattern-based expression rewriting.
V2 will include graph partitioning and verification to better support bring
your own code gen, starting in o
Hi @jonso,
You can take a look at the matcher here:
https://github.com/apache/incubator-tvm/pull/5231
Need to add a little more documentation, and I'm working on a second stage
pattern-based rewriter now.
---
[Visit
Topic](https://discuss.tvm.ai/t/rfc-relay-program-matching-for-relay-pt
Okay, I've implemented these APIs and converted a couple of passes to use them,
please take a look and review. https://github.com/apache/incubator-tvm/pull/4886
@haichen @MarisaKirisame I'm noticing that the passes with these edge cases all
seem to be doing more complicated things than the sim
Hmm, that's a good point, the other thing I'm currently working on is the
[Pattern
Language](https://discuss.tvm.ai/t/rfc-relay-program-matching-for-relay-pt-1-a-pattern-language/5833),
and I'm noticing some similar kinds of inflexibilities with that. Will think
about how to make this a litt
30 matches
Mail list logo
