Hi,
hi, I am following ACE’s Quick Start Guide but, during step 2.C of the deployment I have an error that I can’t solve
GPU Type: 1660ti
Nvidia Driver Version: 560.35.03
Operating System + Version: Ubuntu 24.10
Python Version (if applicable): 3.12.7
model-utils-speech | ==========================
model-utils-speech | === Riva Speech Skills ===
model-utils-speech | ==========================
model-utils-speech |
model-utils-speech | NVIDIA Release (build 93062467)
model-utils-speech | Copyright (c) 2016-2023, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
model-utils-speech |
model-utils-speech | Copyright (c) 2018-2023, NVIDIA CORPORATION & AFFILIATES. All rights reserved.
model-utils-speech |
model-utils-speech | TensorRT SDK | NVIDIA Developer
model-utils-speech |
model-utils-speech | Various files include modifications (c) NVIDIA CORPORATION & AFFILIATES. All rights reserved.
model-utils-speech |
model-utils-speech | This container image and its contents are governed by the NVIDIA Deep Learning Container License.
model-utils-speech | By pulling and using the container, you accept the terms and conditions of this license:
model-utils-speech | https://developer.nvidia.com/ngc/nvidia-deep-learning-container-license
model-utils-speech |
model-utils-speech | To install Python sample dependencies, run /opt/tensorrt/python/python_setup.sh
model-utils-speech |
model-utils-speech | To install the open-source samples corresponding to this TensorRT release version
model-utils-speech | run /opt/tensorrt/install_opensource.sh. To build the open source parsers,
model-utils-speech | plugins, and samples for current top-of-tree on master or a different branch,
model-utils-speech | run /opt/tensorrt/install_opensource.sh -b
model-utils-speech | See GitHub - NVIDIA/TensorRT: NVIDIA® TensorRT™ is an SDK for high-performance deep learning inference on NVIDIA GPUs. This repository contains the open source components of TensorRT. for more information.
model-utils-speech |
model-utils-speech | 2024-10-24 08:41:14,266 [INFO] Writing Riva model repository to ‘/data/models’…
model-utils-speech | 2024-10-24 08:41:14,266 [INFO] The riva model repo target directory is /data/models
model-utils-speech | 2024-10-24 08:42:55,211 [INFO] Using obey-precision pass with fp16 TRT
model-utils-speech | 2024-10-24 08:42:55,212 [INFO] Using tensorrt with fp16
model-utils-speech | 2024-10-24 08:42:55,214 [INFO] Extract_binaries for nn → /data/models/riva-trt-parakeet-1.1b-en-US-asr-streaming-am-streaming/1
model-utils-speech | 2024-10-24 08:42:55,214 [INFO] extracting {‘onnx’: (‘nemo.collections.asr.models.ctc_bpe_models.EncDecCTCModelBPE’, ‘model_graph.onnx’)} → /data/models/riva-trt-parakeet-1.1b-en-US-asr-streaming-am-streaming/1
model-utils-speech | 2024-10-24 08:43:10,476 [INFO] Printing copied artifacts:
model-utils-speech | 2024-10-24 08:43:10,477 [INFO] {‘onnx’: ‘/data/models/riva-trt-parakeet-1.1b-en-US-asr-streaming-am-streaming/1/model_graph.onnx’}
model-utils-speech | 2024-10-24 08:43:10,477 [INFO] Building TRT engine from ONNX file /data/models/riva-trt-parakeet-1.1b-en-US-asr-streaming-am-streaming/1/model_graph.onnx
model-utils-speech | [10/24/2024-08:43:31] [TRT] [W] onnx2trt_utils.cpp:374: Your ONNX model has been generated with INT64 weights, while TensorRT does not natively support INT64. Attempting to cast down to INT32.
model-utils-speech | [10/24/2024-08:43:32] [TRT] [W] onnx2trt_utils.cpp:400: One or more weights outside the range of INT32 was clamped
model-utils-speech | [10/24/2024-08:43:38] [TRT] [E] ModelImporter.cpp:535: Parse was called with a non-empty network definition
model-utils-speech | 2024-10-24 08:43:38,868 [INFO] Mixed-precision net: 12587 layers, 12587 tensors, 1 outputs…
model-utils-speech | 2024-10-24 08:43:39,226 [INFO] Mixed-precision net: 0 layers / 0 outputs fixed
model-utils-speech | 2024-10-24 08:50:56,791 [ERROR] ACE Agent Model Utils failed with error : Failed to generate Triton model plans
model-utils-speech exited with code 1