As mentioned in this blog post Accelerating Inference with Sparsity Using the NVIDIA Ampere Architecture and NVIDIA TensorRT | NVIDIA Technical Blog, I tried to use ngc to download their uploaded models.
However, when I use command ngc registry model download-version “nvidia/resnext101_32x8d_sparse_onnx:1” to download it exits after a few seconds and reports this error：
Getting files to download…
⠹ ━━━━━━━━━━ • 0.0/338.3 • Remaining: -:–:-- • ? • Elapsed: 0:00:01 • Total: 1 - Completed: 0 - Failed: 0
MiB Transient error StatusCode.DEADLINE_EXCEEDED encountered while exporting traces to prod.otel.kaizen.nvidia.
⠋ ━━━━━━━━━━ • 0.0/338.3 • Remaining: -:–:-- • ? • Elapsed: 0:00:08 • Total: 1 - Completed: 0 - Failed: 0
MiB
Download status: COMPLETED
Downloaded local path model: /n_m_sparsity/resnext101_32x8d_sparse_onnx_v1
Total files downloaded: 0
Total transferred: 0 B
Started at: 2024-07-24 02:53:36
Completed at: 2024-07-24 02:53:45
Duration taken: 8s
What should I do about it?