diff --git a/CHANGELOG.md b/CHANGELOG.md index 349d3efff8..1078a2549b 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -232,4 +232,74 @@ Signed-off-by: Naren Dasan Signed-off-by: Naren Dasan +# v0.1.0 (2020-10-23) + + +### Bug Fixes + +* added some fixes, trt/jit output still mismatches ([723ac1d](https://github.com/NVIDIA/TRTorch/commit/723ac1d)) +* added test cases to explicitly check hidden/cell state outputs ([d7c3164](https://github.com/NVIDIA/TRTorch/commit/d7c3164)) +* cleaned up logic, added case where bias doesn't exist for LSTM cell converter ([a3e1093](https://github.com/NVIDIA/TRTorch/commit/a3e1093)) +* **//core/conversion/evaluator:** Custom to IValue that handles int[] ([68c934a](https://github.com/NVIDIA/TRTorch/commit/68c934a)) +* **//docker:** Workaround only shared libraries being available in ([50c7eda](https://github.com/NVIDIA/TRTorch/commit/50c7eda)) +* **//py:** Fix long description section of setup.py ([efd2099](https://github.com/NVIDIA/TRTorch/commit/efd2099)) +* **//tests:** Add stride to complete tensors ([af5d28e](https://github.com/NVIDIA/TRTorch/commit/af5d28e)) +* **//tests/accuracy:** Fix int8 accuracy test for new PTQ api ([a53bea7](https://github.com/NVIDIA/TRTorch/commit/a53bea7)) +* **//tests/core/converters/activations:** Complete tensors in prelu test ([0e90f78](https://github.com/NVIDIA/TRTorch/commit/0e90f78)) +* **docsrc:** Update docsrc container for bazel 3.4.1 ([4eb53b5](https://github.com/NVIDIA/TRTorch/commit/4eb53b5)) + + +* fix(Windows)!: Fix dependency resolution for local builds ([858d8c3](https://github.com/NVIDIA/TRTorch/commit/858d8c3)) +* chore!: Update dependencies to PyTorch 1.6.0 ([8eda27d](https://github.com/NVIDIA/TRTorch/commit/8eda27d)) +* chore!: Bumping version numbers to 0.1.0 ([b84c90b](https://github.com/NVIDIA/TRTorch/commit/b84c90b)) +* refactor(//core)!: Introducing a binding convention that will address ([5a105c6](https://github.com/NVIDIA/TRTorch/commit/5a105c6)) +* refactor!: Renaming extra info to compile spec to be more consistent ([b8fa228](https://github.com/NVIDIA/TRTorch/commit/b8fa228)) + + +### Features + +* **//core/conversion/converters:** LSTMCell converter ([8c61248](https://github.com/NVIDIA/TRTorch/commit/8c61248)) +* **//core/conversion/var:** created ITensorOrFreeze() method, to replace functionality of Var::ITensor() ([2ccf8d0](https://github.com/NVIDIA/TRTorch/commit/2ccf8d0)) +* **//core/converters:** Add power layer conversion support and minor README edits ([a801506](https://github.com/NVIDIA/TRTorch/commit/a801506)) +* **//core/lowering:** Add functionalization pass to replace implace ([90a9ed6](https://github.com/NVIDIA/TRTorch/commit/90a9ed6)), closes [#30](https://github.com/NVIDIA/TRTorch/issues/30) +* **//docker:** Adding CUDA11 based container for Ampere support ([970d775](https://github.com/NVIDIA/TRTorch/commit/970d775)) +* started working on lstm_cell converter ([546d790](https://github.com/NVIDIA/TRTorch/commit/546d790)) +* **//py:** Initial compiliant implementation of the to_backend api for ([59113cf](https://github.com/NVIDIA/TRTorch/commit/59113cf)) +* **//third_party/tensorrt:** Add back TensorRT static lib in a cross ([d3c2e7e](https://github.com/NVIDIA/TRTorch/commit/d3c2e7e)) +* **aten::prelu:** Basic prelu support ([8bc4369](https://github.com/NVIDIA/TRTorch/commit/8bc4369)) +* **aten::prelu:** Implement the multi-channel version of prelu and ([c066581](https://github.com/NVIDIA/TRTorch/commit/c066581)) +* finished logic for LSTM cell, now to test ([a88cfaf](https://github.com/NVIDIA/TRTorch/commit/a88cfaf)) + + +### BREAKING CHANGES + +* Users on Windows trying to use cuDNN 8 must manually +configure third_party/cudnn/local/BUILD to use cuDNN 8. + +Signed-off-by: Naren Dasan +Signed-off-by: Naren Dasan +* Support for Python 3.5 is being dropped with this +update + +Signed-off-by: Naren Dasan +Signed-off-by: Naren Dasan +* Version is being bumped to version 0.1.0a0 to target +PyTorch 1.6.0 + +Signed-off-by: Naren Dasan +Signed-off-by: Naren Dasan +* This changes the "ABI" of compiled TRTorch programs and +the runtime and breaks backwards compatability between the runtime in +0.1.0+ and programs compiled pre-0.1.0 + +Signed-off-by: Naren Dasan +Signed-off-by: Naren Dasan +* This changes the top level api for setting the +specification for compilation, a simple find and replace should allow +users to port forward + +Signed-off-by: Naren Dasan +Signed-off-by: Naren Dasan + + diff --git a/docsrc/RELEASE_CHECKLIST.md b/docsrc/RELEASE_CHECKLIST.md index 5513f25742..48d85cf6ef 100644 --- a/docsrc/RELEASE_CHECKLIST.md +++ b/docsrc/RELEASE_CHECKLIST.md @@ -32,14 +32,14 @@ will result in a minor version bump and sigificant bug fixes will result in a pa - Reset `docsrc/conf.py` version - `make html` - Generate changelog - - `conventional-changelog -p angular -s CHANGELOG.md -t -a` + - `conventional-changelog -p angular -s -i CHANGELOG.md -t -a` 4. Once PR is merged tag commit and start creating release on GitHub - Paste in Milestone information and Changelog information into release notes - Generate libtrtorch.tar.gz for the following platforms: - x86_64 cxx11-abi - x86_64 pre-cxx11-abi - TODO: Add cxx11-abi build for aarch64 when a manylinux container for aarch64 exists - - Generate Python packages for Python 3.5/3.6/3.7/3.8 for x86_64 + - Generate Python packages for Python 3.6/3.7/3.8 for x86_64 - TODO: Build a manylinux container for aarch64 - `docker run -it -v$(pwd)/..:/workspace/TRTorch build_trtorch_wheel /bin/bash /workspace/TRTorch/py/build_whl.sh` generates all wheels - To build container `docker build -t build_trtorch_wheel .` \ No newline at end of file diff --git a/docsrc/conf.py b/docsrc/conf.py index 7c0805947f..bcf829d6e4 100644 --- a/docsrc/conf.py +++ b/docsrc/conf.py @@ -124,6 +124,7 @@ 'master_doc': True, "version_info": { "master": "https://nvidia.github.io/TRTorch/", + "v0.1.0": "https://nvidia.github.io/TRTorch/v0.1.0/", "v0.0.3": "https://nvidia.github.io/TRTorch/v0.0.3/", "v0.0.2": "https://nvidia.github.io/TRTorch/v0.0.2/", "v0.0.1": "https://nvidia.github.io/TRTorch/v0.0.1/", diff --git a/py/setup.py b/py/setup.py index 4f0db95f63..94cc398212 100644 --- a/py/setup.py +++ b/py/setup.py @@ -16,7 +16,7 @@ dir_path = os.path.dirname(os.path.realpath(__file__)) -__version__ = '0.1.0a0' +__version__ = '0.1.0' CXX11_ABI = False