Sambit Mohapatra

Multi-process option to load sensor data

4

Hi, While using the API function to load data, I observe that the reading is quite slow, I think this is mainly due to the large number of annotations in...

Hello, I see the following: car_detection_ground AP: 63.954884 73.779037 74.295746 As I understand, the scores are ordered as easy, moderate, hard. If this this correct then is there something wrong...

auto_tune_lr + StochasticWeightAveraging returns error

1

### First check - [X] I'm sure this is a bug. - [X] I've added a descriptive title to this bug. - [x] I've provided clear instructions on how to...

Request for example of C++/Python inter operation use

3

Hello, First, a million thanks for your work!! It's not an overstatement to say - your framework works at least 300% better than Nvidia's stuff. That said, I am a...

🐛 [Bug] Encountered bug when using Torch-TensorRT - native compilation xavier AGX

1

## Bug Description After successful build of both C++ and Python interfaces I get this error when I try to import torch_tensorrt  ## To Reproduce Steps to reproduce the...

How to correctly format input for Fp16 inference using torch-tensorrt C++

4

## ❓ Question ## What you have already tried Hi, I am using the following to export a torch scripted model to Fp16 tensorrt which will then be used in...

Size of model and inference time is same as FP32 after calibration/quatization step.

8

## Description I was trying to follow along this: 1) https://github.com/NVIDIA/TensorRT/blob/master/tools/pytorch-quantization/examples/calibrate_quant_resnet50.ipynb 2) https://github.com/NVIDIA/TensorRT/blob/master/tools/pytorch-quantization/examples/finetune_quant_resnet50.ipynb As I understand, step 1 should result in a quantized INT8 model. So I should expect a...

How to get FP16 weights for optimization level O2?

3

**Describe the Bug** I followed the steps and trained using O2 optimization level. I saved using torch.save(). However, size of saved weights is same as FP32 training. Ideally FP16 weights...

Generating .so error : /usr/bin/ld: cannot find -lz

3

Hello, I have a special but simple test case: `@export def show_array(): @python def plot_random_matrix_100x100(): import numpy as np import matplotlib.pyplot as plt arr = np.random.random((100,100, 3)) plt.imshow(arr) plt.show()` -...

🐛 [Bug] Cannot compile with FP16, "RuntimeError: Expected bias to have type Half but got Float"

1

## Bug Description I am trying to compile a very simple network with Torch-Tensorrt with FP16. I get the following error message:  ## To Reproduce Steps to reproduce the...