junrushao1994 opened a new pull request #8114:
URL: https://github.com/apache/tvm/pull/8114
This PR is part of the TensorIR upstreaming effort (#7527), stage M2a.
In this PR, we implemented cached flag verification, which will check the
correctness of the following flags if the
MarisaKirisame merged pull request #8111:
URL: https://github.com/apache/tvm/pull/8111
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service,
MarisaKirisame commented on pull request #8111:
URL: https://github.com/apache/tvm/pull/8111#issuecomment-846592984
Thanks @YuchenJin @junrushao1994 !
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to
This is an automated email from the ASF dual-hosted git repository.
marisa pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/tvm.git
The following commit(s) were added to refs/heads/main by this push:
new ebf80cb [Docs] Update stale links (#8111)
ebf80cb is
tqchen closed issue #8033:
URL: https://github.com/apache/tvm/issues/8033
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact
zackcquic commented on a change in pull request #7952:
URL: https://github.com/apache/tvm/pull/7952#discussion_r637521199
##
File path: python/tvm/ir/instrument.py
##
@@ -0,0 +1,154 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contributor license
zackcquic removed a comment on pull request #7952:
URL: https://github.com/apache/tvm/pull/7952#issuecomment-845250654
Hi @areusch:
Could you elaborate more on "after_pass with a result notation"?
What kind of result and use scenario you expected?
That will be very helpful
zackcquic commented on a change in pull request #7952:
URL: https://github.com/apache/tvm/pull/7952#discussion_r637521526
##
File path: src/ir/instrument.cc
##
@@ -0,0 +1,314 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license
xqdan commented on pull request #8079:
URL: https://github.com/apache/tvm/pull/8079#issuecomment-846551249
@zackcquic please take another look
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to
zackcquic commented on a change in pull request #7952:
URL: https://github.com/apache/tvm/pull/7952#discussion_r637521860
##
File path: include/tvm/ir/instrument.h
##
@@ -0,0 +1,127 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor
PhilippvK commented on pull request #7401:
URL: https://github.com/apache/tvm/pull/7401#issuecomment-846545791
Hey @t-vi, the idea of a fallback for unsupported TorchScript Ops is great.
I am currently pursuing a similar approach for unsupported (and custom) TFLite
Ops.
I also
zackcquic commented on pull request #7952:
URL: https://github.com/apache/tvm/pull/7952#issuecomment-846553197
### Collector of pass instrumentation instance
**Following table shows the class and API that collects a set of pass
instrumentations in different projects:**
Note:
-
zackcquic commented on a change in pull request #7952:
URL: https://github.com/apache/tvm/pull/7952#discussion_r637520285
##
File path: include/tvm/ir/instrument.h
##
@@ -0,0 +1,127 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor
zackcquic commented on pull request #7952:
URL: https://github.com/apache/tvm/pull/7952#issuecomment-846553311
@tkonolige @areusch @altanh @tqchen Changes are done.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
zackcquic commented on a change in pull request #7952:
URL: https://github.com/apache/tvm/pull/7952#discussion_r637521086
##
File path: python/tvm/ir/instrument.py
##
@@ -0,0 +1,154 @@
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contributor license
zackcquic commented on pull request #7952:
URL: https://github.com/apache/tvm/pull/7952#issuecomment-846538241
> > For the reason not to overload superclass, I don't know, but I have a
guess, maybe need someone to confirm:
> > Currently, only FunctionPass, ModulePass, SequentialPass and
zhuzilin edited a comment on pull request #8085:
URL: https://github.com/apache/tvm/pull/8085#issuecomment-846697224
@tkonolige I've added the comment and rebased the code to main branch to fix
the shape limit in `uniform`. Could you take another look? Thank you~
--
This is an automated
junrushao1994 commented on pull request #8114:
URL: https://github.com/apache/tvm/pull/8114#issuecomment-846631033
CC: @tqchen @jroesch @comaniac @icemelon9 @zhiics
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
zhuzilin commented on pull request #8085:
URL: https://github.com/apache/tvm/pull/8085#issuecomment-846697224
@tkonolige I've added the comment and rebase the code to main branch to fix
the shape limit in `uniform`. Could you take another look? Thank you~
--
This is an automated message
tqchen commented on pull request #8115:
URL: https://github.com/apache/tvm/pull/8115#issuecomment-846610260
cc @tkonolige @junrushao1994
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the
tqchen opened a new pull request #8115:
URL: https://github.com/apache/tvm/pull/8115
Fixes https://github.com/apache/tvm/issues/7856
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the
Beya2019 opened a new pull request #8116:
URL: https://github.com/apache/tvm/pull/8116
Fixed some syntax errors in the hybrid_script document.
```
for i in range(5):
s = 0 # declaration, this s will be a 1-array in lowered IR
for j in range(5):
s += a[i, j] #
gromero commented on pull request #8086:
URL: https://github.com/apache/tvm/pull/8086#issuecomment-846736014
Hi @areusch and @leandron PTAL. In that v2 I addressed your review comments,
added 3 pytest tests to check tvmc + MLF, and rebased the code resolving the
conflicts with recent
23 matches
Mail list logo