deepstream-import-vision-model

Files (56)

• .gitattributes
• .gitignore
• BENCHMARK.md
• CHANGELOG.md
• README.md
• SKILL.md
• openai.yaml · agents/
• evals.json · evals/
• install.ps1
• install.sh
• README.md · references/
• engine-build.md · references/
• model-acquire.md · references/
• pipeline-run.md · references/
• report-generation.md · references/
• windows.md · references/
• benchmark-ds.sh · scripts/deepstream/
• ds-kitti-dump.sh · scripts/deepstream/
• ds-perf-run.sh · scripts/deepstream/
• ds-single-stream.sh · scripts/deepstream/
• ds-sweep.sh · scripts/deepstream/
• extract-frame.sh · scripts/deepstream/
• dsrun.sh · scripts/
• benchmark-trtexec.sh · scripts/engine/
• cleanup.sh · scripts/model/
• config-to-labels.py · scripts/model/
• hf-download-config.sh · scripts/model/
• hf-list-files.sh · scripts/model/
• inspect-onnx.py · scripts/model/
• make-static-batch-onnx.py · scripts/model/
• ngc-download.sh · scripts/model/
• ngc-list-files.sh · scripts/model/
• resolve-engine.sh · scripts/model/
• safetensors-to-onnx.sh · scripts/model/
• safetensors_to_onnx.py · scripts/model/
• preflight.sh · scripts/
• generate-benchmark-charts.py · scripts/report/
• latex-pdf-wrap.tex · scripts/report/
• md-to-html-pdf.py · scripts/report/
• md-to-pdf.sh · scripts/report/
• mermaid-puppeteer-root.json · scripts/report/
• mermaid-puppeteer.json · scripts/report/
• pandoc-wrap-tables.lua · scripts/report/
• render-mermaid-for-pdf.py · scripts/report/
• report-style.css · scripts/report/
• requirements.txt · scripts/
• setup.sh
• skill-card.md
• skill.oms.sig
• README.md · tests/
• __init__.py · tests/
• test_evals_contract.py · tests/
• test_hardened_scripts.py · tests/
• test_onnx_export_contract.py · tests/
• test_powershell_installer.py · tests/
• test_skill_model_intake.py · tests/

SKILL.md (first lines)

# DeepStream Import Vision Model

When this skill is active, **read the relevant reference document before starting each phase**. Do not rely on memory — reference documents contain exact script paths, bash variable conventions, log filename contracts, and critical parsing rules.

**Current scope:** Object detection models only. Fail fast on classification, segmentation, or other architectures detected in `config.json`.

## Model choice — always offer two options

Before preflight, browsing, downloads, or file creation, present exactly these two choices. Do not
start with only an open-ended model-source prompt. If the user's request already clearly selects a
model, confirm the matching choice instead of asking redundantly.

### 1. Default model (recommended)

Use the validated Hugging Face RT-DETR model:

```yaml
model_id: PekingU/rtdetr_r50vd
source: huggingface
task: object-detection
precision_preference: fp16
```

### 2. Custom object-detection model

Ask for one supported source:

- Hugging Face model ID (`organization/model`) or full model URL.
- NVIDIA NGC catalog model URL including its version.

Explain that the skill currently rejects classification, segmentation, and other non-detection
architectures after inspecting `config.json`. Do not invent or silently substitute a model when the
custom source is missing or unsupported.

For a dry run, present the same two choices and simulate discovery, build, benchmark, and report
stages without browsing, downloading, launching Docker, writing files, or starting processes.

## Pipeline Overview

| Step | Phase | Reference | What it does |
|------|-------|-----------|--------------|
| 1–3 | Model Acquire | [references/model-acquire.md](references/model-acquire.md) | Browse HF/NGC, detect format, download ONNX or export SafeTensors |
| 4–5 | Engine Build  | [references/engine-build.md](references/engine-build.md) | Build dynamic TRT engine, run trtexec BS=1 and BS=MAX_BS |
| 6–7 | DS Pipeline   | [references/pipeline-run.md](references/pipeline-run.md) | Custom bbox parser, nvinfer config, single-stream + multi-stream benchmarks |
| 8   | Report        | [references/report-generation.md](references/report-generation.md) | 5 charts, HTML, PDF benchmark report |

Run the full pipeline autonomously without pausing for confirmation at each step.

## Runs entirely through Docker (no host packages)

**Every step runs INSIDE the DeepStream container.** The host needs only **Docker + the NVIDIA
driver** — no host python/venv/torch/trtexec/make/wkhtmltopdf. This works identically on Linux and
**Windows** (Docker Desktop + WSL2 backend, required for `--gpus`). The per-shell bind-mount
token is the only OS difference — `-v "$PWD":/work` (bash), `-v "${PWD}:/work"` (PowerShell),
`-v "%cd%:/work"` (cmd); full guide in [references/windows.md](references/windows.md). All venv/ONNX/
engine/parser/config/report artifacts live under the mounted working root and persist between the
ephemeral `--rm` containers.

## Pre-flight — bootstrap + verify (through the container)

**1. One-time bootstrap** — builds `build/.venv_optimum` (torch/onnx/onnxruntime/report deps; the
venv name is historical, optimum is no longer used) +
installs `wkhtmltopdf`, all in-container. From the working root:
```bash
docker run --rm -it --gpus all --shm-size=16g -v "$PWD":/work -w /work \
  --entrypoint bash nvcr.io/nvidia/deepstream:9.1-triton-multiarch \
  .claude/skills/deepstream-import-vision-model/setup.sh
```

**2. Preflight** — GPU + venv + trtexec, run THROUGH the container (container-mode auto-detects):
```bash
docker run --rm --gpus all -v "$PWD":/work -w /work \
  --entrypoint bash nvcr.io/nvidia/deepstream:9.1-triton-multiarch \
  .claude/skills/deepstream-import-vision-model/scripts/preflight.sh   # proceed only on PASS
```

**Every subsequent phase runs the same way** — issue the model's commands via
`docker run … --entrypoint bash … -lc '<commands>'` (or the
`.claude/skills/deepstream-import-vision-model/scripts/dsrun.sh` wrapper:
`bash .claude/skills/deepstream-import-vision-model/scripts/dsrun.sh '<in-container command>'`),
using `PY=build/.venv_optimum/bin/python` and
`trtexec` at `/usr/src/tensorrt/bin/trtexec` inside the container. `deepstream-app`,
`gst-launch-1.0`, and `/opt/nvidia/deepstream/…` sample paths all exist **in** the image.
TensorRT build+runtime share one image, so there is **no version skew** (the concern the old
"build on the host" rule tried to avoid — see [references/engine-build.md](references/engine-build.md)).
`sample_720p.mp4` ships in the image; set `DS_VIDEO` only to override.

## Mandatory Output Structure

Create once `MODEL_NAME` is known (Step 1). Never dump files flat.

```
models/{model_name}/
  model/           <- ONNX file(s)
  parser/          <- .cpp, Makefile, .so
  config/          <- nvinfer config, ds-app config, labels.txt
  scripts/         <- run helper scripts
  benchmarks/
    engines/       <- _dynamic_b{MAX_BS}.engine, timing.cache, build logs
    b1/            <- trtexec BS=1 log
    b{MAX_BS}/     <- trtexec BS=MAX_BS log
    ds/            <- DS benchmark logs
  reports/         <- benchmark_report.md, .html, .pdf, benchmark_data.json
    charts/        <- chart_*.png (5 charts)
  samples/         <- output .mp4 or .ogv (theoraenc fallback), test frames
    kitti_output/  <- KITTI detection .txt files
```

```bash
mkdir -p models/$MODEL_NAME/{model,parser,config,scripts,benchmarks/engines,benchmarks/ds,reports/charts,samples/kitti_output}
```

## Critical Rules

1. **Engine naming** — always `{model}_dynamic_b{MAX_BS}.engine`. Never bare `model_dynamic.engine`.
2. **batch_size == num_streams** — in DS runs, `batch-size` and stream count are always equal.
3. **Log filenames are fixed** — `trtexec_b1.log`, `trtexec_b${MAX_BS}.log`, `ds_s${N}_run1.log`, `ds_s${N}_run2.log`. No timestamps. Report generation reads exact paths.
4. **Parser zero-init** — always `NvDsInferObjectDetectionInfo obj = {};`. Required for DS 9.1 OBB support; bare `obj;` leaves `rotation_angle` uninitialized, causing tilted bounding boxes.
5. **KITTI validation gate** — do NOT proceed to Step 7 if KITTI frame count is zero or detection rate < 90%.

From NVIDIA/skills, CC-BY-4.0 AND Apache-2.0.

More from NVIDIA/skills