Hello, @fantasyRqg
Thanks for very good PR!! I try using your code to export one system-lib model
and I have one problem about model similarity score.
I try to export 2 model with `build_one_system_lib` and it work fine. However,
when I try to validate similarity score between exported TVM m
I don't think install from source needs any system privileges, even
PYTHON_PATH, you can use export method to make it visible.
---
[Visit
Topic](https://discuss.tvm.apache.org/t/conda-tvm-py-target-build-llvm-is-not-enabled/9346/7)
to respond.
You are receiving this because you enabled m
seems another way: conda env create --file conda/build-environment.yaml,
i am trying ... ...
---
[Visit
Topic](https://discuss.tvm.apache.org/t/conda-tvm-py-target-build-llvm-is-not-enabled/9346/6)
to respond.
You are receiving this because you enabled mailing list mode.
To unsubscribe f
Thanks, i understand what you mean. Because I am not the supervisor of my
company's server, so i use the conda environment and install the package. It
seems that current conda package is with this issue.
---
[Visit
Topic](https://discuss.tvm.apache.org/t/conda-tvm-py-target-build-llvm-is-
no, I mean in cmake.config, set llvm config and recompile libtvm.so from tvm
source code.
see
https://tvm.apache.org/docs/install/from_source.html#developers-get-source-from-github.
Install tvm from source, and export to PYTHON_PATH, and it will be visible to
conda env.
---
[Visit
Topic]
I use the way of conda package like below:
conda install -c conda-forge tvm-py
Do you mean that we can "set llvm toolchain during the compilation of
libtvm.so" when using the conda install?
---
[Visit
Topic](https://discuss.tvm.apache.org/t/conda-tvm-py-target-build-llvm-is-not-enabled/9
Hi, everyone ! I have a few questions about the following code in
`include/tvm/tir/stmt_functor.h`
```
template
TSelf& set_dispatch(FPointer f) { // NOLINT(*)
uint32_t tindex = TNode::RuntimeTypeIndex();
if (func_.size() <= tindex) {
func_.resize(tindex + 1, nullptr);
}
In pytorch, graph converts a mkldnn_graph(use to_mkldnn).
then compile the graph.
In tvm, is the same process as pytorch included in relay.build??
normal graph -> mkldnn graph.
if it's right, where can i find a graph(dense layer) applied with mkldnn.
Im already know that tvm only applies mkl
Hi, have you solved it ? I would like to use some dynamic computed loop counts
inside te.compute and don't know how to do it, I guess it is the same issue as
yours. Looking forward to your feedback, thanks.
---
[Visit
Topic](https://discuss.tvm.apache.org/t/does-tvm-support-dynamic-range-
Have you set llvm toolchain during the compilation of libtvm.so ?
---
[Visit
Topic](https://discuss.tvm.apache.org/t/conda-tvm-py-target-build-llvm-is-not-enabled/9346/2)
to respond.
You are receiving this because you enabled mailing list mode.
To unsubscribe from these emails, [click
h
Hi, appreciate your help,
I try to install TVM at conda environment and do the tutorial of " Quick Start
Tutorial for Compiling Deep Learning Models". I meet the issue about
`target.build.llvm` is not enabled like below:
https://github.com/conda-forge/libtvm-feedstock/issues/7
https://tvm.apa
Sorry about the confusion. You can still use mkldnn without autoTVM tuning. The
warning log only indicates that AutoTVM cannot find the profile results of
mkldnn kernel.
---
[Visit
Topic](https://discuss.tvm.apache.org/t/mkldnn-verbose-doesnt-work/9315/6) to
respond.
You are receiving t
12 matches
Mail list logo
