Dheeraj Peri

QAT model saving bug : KeyError: '__inference_depthwise_conv2d_layer_call_fn_126

19

**Describe the bug** Please download the scripts to reproduce from : https://drive.google.com/drive/folders/15cajAZ9sAZ2Uyix8sDVSYku6QCqDCec7?usp=sharing Command to run : python sample_qat.py. I have a simple model with input layer and a depthwise conv2d...

Issue with loading quantization aware trained model

13

**Describe the bug** Unable to load the saved model after applying quantization aware training. **System information** TensorFlow version (installed from source or binary): 2.2 TensorFlow Model Optimization version (installed from...

feat(//tools/perf): Refactor perf_run.py, add fx2trt backend support, usage via CLI arguments

1

# Description Refactor perf_run.py, add fx2trt backend support, usage via CLI arguments. Added a `benchmark.sh` which is used for internal perf regression testing. ## Type of change - New feature...

## Bug Description /* The following test is ambigious and somehow works in TRT 8.2, which might have a bug. This FP16 model has inputs and weights configured to be...

## Bug Description If a tensorrt calibrator implemented using TRT python API is passed to Torch-TRT compilation, it fails due to the object not being recognized. Error : ``` ======================================================================...

✨[Feature] Do not remove nodes which are in-place operations (classified in TRT segments) in fallback

3

**Is your feature request related to a problem? Please describe.** Currently for eg: if aten::append is in prim::If block, it gets removed because it is classified as TRT segment. No...

✨[Feature] Dynamic shape support in Automatic fallback

3

**Is your feature request related to a problem? Please describe.** Support dynamic shapes in the input via automatic fallback. **Describe the solution you'd like** Currently the compilation fails when automatic...

Signed-off-by: Dheeraj Peri # Description Please include a summary of the change and which issue is fixed. Please also include relevant motivation and context. List any dependencies that are required...

feat: Add min, max ranges to mark_dynamic API

13

Fixes https://github.com/pytorch/pytorch/issues/115137 This PR adds: - mark_dynamic API will accept `min`, `max` values to create a bounded constraint on the dim. - test case in test_misc.py which checks if `ConstraintViolationError`...

# Description Please include a summary of the change and which issue is fixed. Please also include relevant motivation and context. List any dependencies that are required for this change....