Thanks, that makes sense. I was thinking that while calibration, you could use
different attributes for `simulated_quantize` and `simulated_dequantize` ops.
In the callback of calibrating an operator, one can simulate the affine space
and argue about scales and zero points. But for capturing r
Also, as part of the standardization of QNN, we could ensure that all QNN
"compute" ops go from `int8 -> int8` . I believe that `qnn.conv2d` is the only
QNN op that outputs an accumulation dtype, so we could change `qnn.conv2d` to
take in bias in addition to the data and weight.
---
[Visi
Here are my two cents before diving into the detail code review.
1. At the first glance most implementations, including the Relay passes, were
done in Python. It would be better to implement them in C++ for better
performance.
2. The term and namespace "subgraph" is improper and confusing.
3.
@tqchen @comaniac @areusch @tmoreau89 @zhiics @giuseros.
---
[Visit
Topic](https://discuss.tvm.apache.org/t/rfc-compute-graph-pipeline-with-new-subgraph-executor/9839/2)
to respond.
You are receiving this because you enabled mailing list mode.
To unsubscribe from these emails, [click
he
[quote="electriclilies, post:21, topic:9775, full:true"]
@mikeseven
Yes, the goal is to create a fully quantized graph, and we do recognize that
this transformation will change the output of the graph. For this reason, we're
not going to present the rewrite as a Relay pass. And I definitely agr
[quote="anijain2305, post:20, topic:9775"]
I am trying to understand why we need `qnn.conv2d*` (* represents operator
along the lines of `qnn.simulated_conv2d`) during calibration. The only reason
would be if you want to propagate the error from previous operators while
**calibrating** current
@mikeseven
Yes, the goal is to create a fully quantized graph, and we do recognize that
this transformation will change the output of the graph. For this reason, we're
not going to present the rewrite as a Relay pass. And I definitely agree that
we should let there be user-defined handling.
A
Thanks @comaniac @zhiics @giuseros @JaydeepIMG @JakeStevens for the review and
discussion.
The compute pipeline runtime/executor PR already get
submitted(https://github.com/apache/tvm/pull/7892) , I also propose a new
RFC(https://discuss.tvm.apache.org/t/rfc-compute-graph-pipeline-with-new-s
>This is a follow up RFC for
>https://discuss.tvm.apache.org/t/compute-graph-pipeline/8957
>PR https://github.com/apache/tvm/pull/7892
# **Split relay graph into Subgraph then doing Subgraph Pipeline :RFC**
In this RFC, we present a new framework for subgraph pipelining that first
split rela
The new [RFC repository](https://github.com/apache/tvm-rfcs) is online, and we
have a [final draft of the RFC process ready for
review](https://github.com/apache/tvm-rfcs/pull/2). Please take a moment to
review the process and raise any issues or questions so that we can address
them before w
### What is a rolling buffer?
A rolling buffer (at least for the purposes of this RFC) is a buffer where one
of the dimensions should be addressed via modulo arithmetic. This gives it a
'wrap-around' behaviour which makes it self-overwriting. This means they are
effectively just higher dimens
@siju-samuel @FrozenGene you may also be interested in this proposal.
---
[Visit
Topic](https://discuss.tvm.apache.org/t/rfc-tflite-frontend-create-models-for-frontend-testing-by-directly-writing-tflite-buffers/9811/3)
to respond.
You are receiving this because you enabled mailing list mo
@anijain2305 @tqchen @dmitriy-arm
---
[Visit
Topic](https://discuss.tvm.apache.org/t/rfc-tflite-frontend-create-models-for-frontend-testing-by-directly-writing-tflite-buffers/9811/2)
to respond.
You are receiving this because you enabled mailing list mode.
To unsubscribe from these email
I apologize for the long delay.
Thanks @electriclilies and team for nicely written RFC. I support the idea.
Reading through the comments, it seems that many of us are in agreement about
the AutoQ and its reliance on QNN extension. The mentioned pain points mostly
revolve around
* The inconsi
14 matches
Mail list logo
