TAO TF1 DetectNet_v2 dataset_convert COCO typo

nkaaf · November 3, 2023, 8:32am

Description

I found an typo in the example configuration of the COCO config for dataset_convert. The table with the parameters mentions the image_dir_names parameter, and the example config file includes img_dir_names. Correct is image_dir_names.

Steps To Reproduce

https://docs.nvidia.com/tao/tao-toolkit/text/object_detection/detectnet_v2.html#configuration-file-for-dataset-converter

AakankshaS · November 27, 2023, 7:43am

Hi,
Can you try running your model with trtexec command, and share the “”–verbose"" log in case if the issue persist

You can refer below link for all the supported operators list, in case any operator is not supported you need to create a custom plugin to support that operation

github.com

onnx/onnx-tensorrt/blob/main/docs/operators.md

<!--- SPDX-License-Identifier: Apache-2.0 -->

# Supported ONNX Operators

TensorRT 8.6 supports operators up to Opset 17. Latest information of ONNX operators can be found [here](https://github.com/onnx/onnx/blob/master/docs/Operators.md)

TensorRT supports the following ONNX data types: DOUBLE, FLOAT32, FLOAT16, INT8, and BOOL

> Note: There is limited support for INT32, INT64, and DOUBLE types. TensorRT will attempt to cast down INT64 to INT32 and DOUBLE down to FLOAT, clamping values to `+-INT_MAX` or `+-FLT_MAX` if necessary.

See below for the support matrix of ONNX operators in ONNX-TensorRT.

## Operator Support Matrix

| Operator                  | Supported  | Supported Types | Restrictions                                                                                                           |
|---------------------------|------------|-----------------|------------------------------------------------------------------------------------------------------------------------|
| Abs                       | Y          | FP32, FP16, INT32 |
| Acos                      | Y          | FP32, FP16 |
| Acosh                     | Y          | FP32, FP16 |
| Add                       | Y          | FP32, FP16, INT32 |

This file has been truncated. show original

Also, request you to share your model and script if not shared already so that we can help you better.

Meanwhile, for some common errors and queries please refer to below link:

Thanks!

AakankshaS · November 29, 2023, 8:39am

Hi @nkaaf ,
Moving to TAO Forum

Morganh · November 29, 2023, 9:03am

Thanks for the catching. Yes, you are right. Also, it matches https://github.com/NVIDIA/tao_tensorflow1_backend/blob/c7a3926ddddf3911842e057620bceb45bb5303cc/nvidia_tao_tf1/cv/detectnet_v2/dataio/coco_converter_lib.py#L45.
Will improve the document.

system · December 24, 2023, 10:40am

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.

Topic		Replies	Views
Trtexec convert onnx to engine fails TAO Toolkit	14	1462	October 30, 2023
TAO Toolkit: Problem with dataset conversion using COCO format TAO Toolkit	4	848	January 30, 2023
Driveworks dnn custom model for objects detection (yolo) DRIVE AGX Xavier General driveworks	13	2215	August 5, 2022
Detectron2: faster inferencing TensorRT	2	1500	April 29, 2022
Excute tao model detectnet_v2 train but Failed TAO Toolkit tao	5	301	June 4, 2024
TAO TF1 DetectNet_v2 Dataset_Converter Typo TAO Toolkit documentation , tao	3	715	December 24, 2023
TAO DetectNet_v2 doc KITTI Converter Parameter TAO Toolkit documentation , tao	4	742	February 27, 2024
Running trtexec on Google's EfficientDet-D4 in ONNX Format TensorRT	9	1428	June 11, 2021
Tao training - Visualise inference after training provides 98% accuracy, however, after model export to TensorRT, the inference result is 0% TAO Toolkit	5	683	March 12, 2022
TensorRT Inference form a .etlt model on Python TAO Toolkit tensorrt	7	1347	November 16, 2021

TAO TF1 DetectNet_v2 dataset_convert COCO typo

Description

Steps To Reproduce

Related topics