diff --git a/hpvm/docs/getting-started.rst b/hpvm/docs/getting-started.rst index 0f1ee13cb8c54a5e9bccdc37c5afcbaf7837e537..cbf5d0b095c1035acf8bfcc7aaccf1d3cb0be3ce 100644 --- a/hpvm/docs/getting-started.rst +++ b/hpvm/docs/getting-started.rst @@ -63,8 +63,8 @@ Create a DNN `module` and load the checkpoint: Any `torch.nn.Module` can be similarly used, as long as they only contain the tensor operators supported in HPVM. -See "Supported Operators" in :doc:`PyTorch frontend <components/torch2hpvm>` -and :doc:`Keras frontend <components/keras-frontend>`. +See "Supported Operators" in :doc:`PyTorch frontend </components/torch2hpvm>` +and :doc:`Keras frontend </components/keras-frontend>`. Now we are ready to export the model. The main functioning class of `torch2hpvm` is `ModelExporter`: @@ -183,7 +183,7 @@ After the tuning finishes, the tuner will * save the HPVM config format (write-only) at ``./hpvm_confs.txt``. It is also possible to save the configuration in other formats -(see the :doc:`predtuner documentation <components/predtuner>`). +(see the :doc:`predtuner documentation </components/predtuner>`). Profiling the Configurations ---------------------------- @@ -228,4 +228,4 @@ An example of ``configs_profiled.png`` looks like this (proportion of your image This concludes the whole workflow of HPVM. For more detailed usages, check out the documentation of each component listed -:doc:`here <components/index>`. +:doc:`here </components/index>`. diff --git a/hpvm/docs/install.rst b/hpvm/docs/install.rst index 47b400d2be63dbbc6caa548384656dc9e4d27bd1..4876cf960c36a172b3f3896acb34eb0310934971 100644 --- a/hpvm/docs/install.rst +++ b/hpvm/docs/install.rst @@ -161,4 +161,4 @@ You can run tests similarly as how ``approxhpvm.py`` is compiled: for example, make -j<number of threads> check-hpvm-pass -runs ``check-hpvm-pass`` tests. See TODO for details on benchmarks and test cases. +runs ``check-hpvm-pass`` tests. See :doc:`/tests` for details on benchmarks and test cases. diff --git a/hpvm/projects/predtuner b/hpvm/projects/predtuner index aa28a41ca12b15af31e1d15b687367e27cde3878..a149e365170263666db764664ad8ed6b03f258d3 160000 --- a/hpvm/projects/predtuner +++ b/hpvm/projects/predtuner @@ -1 +1 @@ -Subproject commit aa28a41ca12b15af31e1d15b687367e27cde3878 +Subproject commit a149e365170263666db764664ad8ed6b03f258d3 diff --git a/hpvm/test/README.rst b/hpvm/test/README.rst index e770b05cb39796aab6deb480012470a553923323..66af5c999fd4c90df193f6ca22cc911419d66f40 100644 --- a/hpvm/test/README.rst +++ b/hpvm/test/README.rst @@ -120,9 +120,3 @@ Currently, there are 20 of them. These are: ``_cudnn`` suffix indicates the code is generated onto cuDNN functions. Otherwise they are generated to ``tensor_runtime`` DNN functions which are hand-written in CUDA. - -TODO: figure out how to -^^^^^^^^^^^^^^^^^^^^^^^ - - -#. Auto run Keras and PyTorch tests (generating, compiling and running all DNNs)