This is an automated email from the ASF dual-hosted git repository.
tqchen pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-tvm-site.git
The following commit(s) were added to refs/heads/master by this push:
new 455aeb1 Points docs to the new location
455aeb1 is described below
commit 455aeb1f33e3f7f4c9cf5b18aa082e46f54bdd58
Author: tqchen <[email protected]>
AuthorDate: Mon Mar 30 11:15:57 2020 -0700
Points docs to the new location
---
_posts/2018-07-12-vta-release-announcement.markdown | 2 +-
_posts/2018-08-10-DLPack-Bridge.md | 2 +-
_posts/2018-10-03-auto-opt-all.md | 6 +++---
_posts/2019-01-19-Golang.md | 4 ++--
_posts/2019-03-18-tvm-apache-announcement.md | 2 +-
_posts/2019-04-30-opt-cuda-quantized.md | 8 ++++----
community.md | 2 +-
7 files changed, 13 insertions(+), 13 deletions(-)
diff --git a/_posts/2018-07-12-vta-release-announcement.markdown
b/_posts/2018-07-12-vta-release-announcement.markdown
index c65db37..107cf10 100644
--- a/_posts/2018-07-12-vta-release-announcement.markdown
+++ b/_posts/2018-07-12-vta-release-announcement.markdown
@@ -149,5 +149,5 @@ VTA is a research project that came out of the SAML group,
which is generously s
## Get Started!
- TVM and VTA Github page can be found here:
[https://github.com/dmlc/tvm](https://github.com/dmlc/tvm).
-- You can get started with easy to follow [tutorials on programming VTA with
TVM](https://docs.tvm.ai/vta/tutorials/index.html).
+- You can get started with easy to follow [tutorials on programming VTA with
TVM](https://tvm.apache.org/docs//vta/tutorials/index.html).
- For more technical details on VTA, read our [VTA technical
report](https://arxiv.org/abs/1807.04188) on ArXiv.
\ No newline at end of file
diff --git a/_posts/2018-08-10-DLPack-Bridge.md
b/_posts/2018-08-10-DLPack-Bridge.md
index f85e7d9..fb4b2e2 100644
--- a/_posts/2018-08-10-DLPack-Bridge.md
+++ b/_posts/2018-08-10-DLPack-Bridge.md
@@ -95,7 +95,7 @@ schedule:
For brevity, we do not cover TVM's large collection of scheduling primitives
that we can use to optimize matrix multiplication. If you wish to make a custom
GEMM operator run _fast_ on your hardware device, a detailed tutorial can be
-found [here](https://docs.tvm.ai/tutorials/optimize/opt_gemm.html).
+found [here](https://tvm.apache.org/docs//tutorials/optimize/opt_gemm.html).
We then convert the TVM function into one that supports PyTorch tensors:
```python
diff --git a/_posts/2018-10-03-auto-opt-all.md
b/_posts/2018-10-03-auto-opt-all.md
index b37ce2e..5c13edf 100644
--- a/_posts/2018-10-03-auto-opt-all.md
+++ b/_posts/2018-10-03-auto-opt-all.md
@@ -190,9 +190,9 @@ for inference deployment. TVM just provides such a solution.
## Links
[1] benchmark:
[https://github.com/dmlc/tvm/tree/master/apps/benchmark](https://github.com/dmlc/tvm/tree/master/apps/benchmark)
-[2] Tutorial on tuning for ARM CPU:
[https://docs.tvm.ai/tutorials/autotvm/tune_nnvm_arm.html](https://docs.tvm.ai/tutorials/autotvm/tune_nnvm_arm.html)
-[3] Tutorial on tuning for Mobile GPU:
[https://docs.tvm.ai/tutorials/autotvm/tune_nnvm_mobile_gpu.html](https://docs.tvm.ai/tutorials/autotvm/tune_nnvm_mobile_gpu.html)
-[4] Tutorial on tuning for NVIDIA/AMD GPU:
[https://docs.tvm.ai/tutorials/autotvm/tune_nnvm_cuda.html](https://docs.tvm.ai/tutorials/autotvm/tune_nnvm_cuda.html)
+[2] Tutorial on tuning for ARM CPU:
[https://tvm.apache.org/docs//tutorials/autotvm/tune_nnvm_arm.html](https://tvm.apache.org/docs//tutorials/autotvm/tune_nnvm_arm.html)
+[3] Tutorial on tuning for Mobile GPU:
[https://tvm.apache.org/docs//tutorials/autotvm/tune_nnvm_mobile_gpu.html](https://tvm.apache.org/docs//tutorials/autotvm/tune_nnvm_mobile_gpu.html)
+[4] Tutorial on tuning for NVIDIA/AMD GPU:
[https://tvm.apache.org/docs//tutorials/autotvm/tune_nnvm_cuda.html](https://tvm.apache.org/docs//tutorials/autotvm/tune_nnvm_cuda.html)
[5] Paper about AutoTVM: [Learning to Optimize Tensor
Program](https://arxiv.org/abs/1805.08166)
[6] Paper about Intel CPU (by AWS contributors) : [Optimizing CNN Model
Inference on CPUs](https://arxiv.org/abs/1809.02697)
diff --git a/_posts/2019-01-19-Golang.md b/_posts/2019-01-19-Golang.md
index c025763..7825345 100644
--- a/_posts/2019-01-19-Golang.md
+++ b/_posts/2019-01-19-Golang.md
@@ -19,7 +19,7 @@ deploy deep learning models from a variety of frameworks to a
choice of hardware
The TVM import and compilation process generates a graph JSON, a module and a
params. Any application that
integrates the TVM runtime can load these compiled modules and perform
inference. A detailed tutorial of module
-import and compilation using TVM can be found at
[tutorials](https://docs.tvm.ai/tutorials/).
+import and compilation using TVM can be found at
[tutorials](https://tvm.apache.org/docs//tutorials/).
TVM now supports deploying compiled modules through Golang. Golang
applications can make use of this
to deploy the deep learning models through TVM. The scope of this blog is the
introduction of ```gotvm``` package,
@@ -51,7 +51,7 @@ Developers can make use of TVM to import and compile deep
learning models and ge
{:center}
<center> Import, Compile, Integrate and Deploy</center> <p></p>
-TVM [Compile Deep Learning
Models](https://docs.tvm.ai/tutorials/#compile-deep-learning-models) tutorials
+TVM [Compile Deep Learning
Models](https://tvm.apache.org/docs//tutorials/#compile-deep-learning-models)
tutorials
are available to compile models from all frameworks supported by the TVM
frontend. This compilation process
generates the artifacts required to integrate and deploy the model on a target.
diff --git a/_posts/2019-03-18-tvm-apache-announcement.md
b/_posts/2019-03-18-tvm-apache-announcement.md
index b90cd2e..6fe0f60 100644
--- a/_posts/2019-03-18-tvm-apache-announcement.md
+++ b/_posts/2019-03-18-tvm-apache-announcement.md
@@ -21,4 +21,4 @@ Besides the technical innovations, the community adopts an
open, welcoming and n
We would like to take this chance to thank the Allen School for supporting the
SAMPL team that gave birth to the TVM project. We would also like to thank the
Halide project which provided the basis for TVM’s loop-level IR and initial
code generation. We would like to thank our Apache incubator mentors for
introducing the project to Apache and providing useful guidance. Finally, we
would like to thank the TVM community and all of the organizations, as listed
above, that supported the deve [...]
-See also the [Allen School news about the transition
here](https://news.cs.washington.edu/2019/03/18/allen-schools-tvm-deep-learning-compiler-framework-transitions-to-apache/),
[TVM conference program slides and
recordings](https://sampl.cs.washington.edu/tvmconf/#about-tvmconf), and [our
community guideline here](https://docs.tvm.ai/contribute/community.html).
Follow us on Twitter: [@ApacheTVM](https://twitter.com/ApacheTVM).
+See also the [Allen School news about the transition
here](https://news.cs.washington.edu/2019/03/18/allen-schools-tvm-deep-learning-compiler-framework-transitions-to-apache/),
[TVM conference program slides and
recordings](https://sampl.cs.washington.edu/tvmconf/#about-tvmconf), and [our
community guideline
here](https://tvm.apache.org/docs//contribute/community.html). Follow us on
Twitter: [@ApacheTVM](https://twitter.com/ApacheTVM).
diff --git a/_posts/2019-04-30-opt-cuda-quantized.md
b/_posts/2019-04-30-opt-cuda-quantized.md
index a96f3b8..ecacd6e 100644
--- a/_posts/2019-04-30-opt-cuda-quantized.md
+++ b/_posts/2019-04-30-opt-cuda-quantized.md
@@ -44,7 +44,7 @@ To illustrate, in 2d convolution we accumulate along the
channel, the width, and
This is a typical use case of `dp4a`.
TVM uses tensorization to support calling external intrinsics.
We do not need to modify the original computation declaration; we use the
schedule primitive `tensorize` to replace the accumulation with `dp4a` tensor
intrinsic.
-More details of tensorization can be found in the
[tutorial](https://docs.tvm.ai/tutorials/language/tensorize.html).
+More details of tensorization can be found in the
[tutorial](https://tvm.apache.org/docs//tutorials/language/tensorize.html).
## Data Layout Rearrangement
One of the challenges in tensorization is that we may need to design special
computation logic to adapt to the requirement of tensor intrinsics.
@@ -87,7 +87,7 @@ We also do some manual tiling such as splitting axes by 4 or
16 to facilitate ve
In quantized 2d convolution, we design a search space that includes a set of
tunable options, such as the tile size, the axes to fuse, configurations of
loop unrolling and double buffering.
The templates of quantized `conv2d` and `dense` on CUDA are registered under
template key `int8`.
During automatic tuning, we can create tuning tasks for these quantized
operators by setting the `template_key` argument.
-Details of how to launch automatic optimization can be found in the [AutoTVM
tutorial](https://docs.tvm.ai/tutorials/autotvm/tune_relay_cuda.html).
+Details of how to launch automatic optimization can be found in the [AutoTVM
tutorial](https://tvm.apache.org/docs//tutorials/autotvm/tune_relay_cuda.html).
# General Workflow
@@ -109,7 +109,7 @@ Next, we use the relay quantization API to convert it to a
quantized model.
net = relay.quantize.quantize(net, params=params)
```
-Then, we use AutoTVM to extract tuning tasks for the operators in the model
and perform automatic optimization. The [AutoTVM
tutorial](https://docs.tvm.ai/tutorials/autotvm/tune_relay_cuda.html) provides
an example for this.
+Then, we use AutoTVM to extract tuning tasks for the operators in the model
and perform automatic optimization. The [AutoTVM
tutorial](https://tvm.apache.org/docs//tutorials/autotvm/tune_relay_cuda.html)
provides an example for this.
Finally, we build the model and run inference in the quantized mode.
```python
@@ -117,7 +117,7 @@ with relay.build_config(opt_level=3):
graph, lib, params = relay.build(net, target)
```
The result of `relay.build` is a deployable library.
-We can either run inference [on the
GPU](https://docs.tvm.ai/tutorials/frontend/from_mxnet.html#execute-the-portable-graph-on-tvm)
directly or deploy [on the remote
devices](https://docs.tvm.ai/tutorials/frontend/deploy_model_on_rasp.html#deploy-the-model-remotely-by-rpc)
via RPC.
+We can either run inference [on the
GPU](https://tvm.apache.org/docs//tutorials/frontend/from_mxnet.html#execute-the-portable-graph-on-tvm)
directly or deploy [on the remote
devices](https://tvm.apache.org/docs//tutorials/frontend/deploy_model_on_rasp.html#deploy-the-model-remotely-by-rpc)
via RPC.
# Benchmark
To verify the performance of the quantized operators in TVM, we benchmark the
performance of several popular network models including VGG-19, ResNet-50 and
Inception V3.
diff --git a/community.md b/community.md
index 703bbb1..7013cd4 100644
--- a/community.md
+++ b/community.md
@@ -59,7 +59,7 @@ Please reach out are interested working in aspects that are
not on the roadmap.
As a community project, we welcome contributions!
The package is developed and used by the community.
-<a href="https://docs.tvm.ai/contribute"; class="link-btn">TVM Contributor
Guideline</a>
+<a href="https://tvm.apache.org/docs//contribute"; class="link-btn">TVM
Contributor Guideline</a>
<br>
