mlir-tensorrt-v0.1.34: Update Version.cmake (#260)

Updates MLIR-TRT version to
0.1.34.

Signed-off-by: yizhuoz004 <[email protected]>

mlir-tensorrt-v0.1.33

[test] Corrects checks in layer metadata callback test

mlir-tensorrt-v0.1.32

Make `tp.Layernorm` 1:1 with torch (#119)

- Make the `normalized_shape` argument 1:1 with Torch. Previously, our
API only supported taking a single integer for the normalization
dimension. Now, a list of integers can be provided.
- Add integration tests for `tp.LayerNorm` and `tp.GroupNorm`

tripy-v0.0.2

Reduces container size, fixes Tripy dependencies, bumps version (#62)

- Greatly reduces the size of the development container by using a
smaller base image and cutting out many unnecessary packages.
    
- Adds TRT as a direct dependency of Tripy since in `plugin.py`, we need
to load it to determine the major version of `libnvinfer_plugin`.
    
- Removes CHANGELOG as we can automatically generate release notes via
GitHub releases.
    
- Removes JAX as a test dependency and the tests that use it. We were
previously using it to test interoperability using DLPack, but we
already test that with torch/cupy/numpy. Adding JAX provides no
additional coverage and unnecessarily bloats the container.

- Bumps version to 0.0.2

mlir-tensorrt-v0.1.31

[mlir-tensorrt] NFC: bump version to 0.1.31

Signed-off-by: Christopher Bate <[email protected]>

tripy-v0.0.1

Contributors (alphabetical):

@akhilg-nv
@azhurkevich
@jhalakpatel
@markkraay
@Mgluhovskoi
@parthchadha
@pranavm-nvidia
@SimengLiu-nv
@slyubomirsky
@yizhuoz004

mlir-tensorrt-v0.1.29

Contributors

@christopherbate
@shelkesagar29
@jhalakpatel
@jinkyuk-nv
@yekerr
@pranavm-nvidia