ultralytics 8.3.70 add data argument to Sony IMX500 export (#18852)
Signed-off-by: Glenn Jocher <glenn.jocher@ultralytics.com> Co-authored-by: UltralyticsAssistant <web@ultralytics.com> Co-authored-by: Ultralytics Assistant <135830346+UltralyticsAssistant@users.noreply.github.com> Co-authored-by: Francesco Mattioli <Francesco.mttl@gmail.com> Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
This commit is contained in:
Lakshantha Dissanayake
GitHub
parent
ec622230fb
commit
f38fa61ffc
6 changed files with 47 additions and 36 deletions
|
|
@ -60,13 +60,18 @@ Export a YOLOv8n model to OpenVINO format and run inference with the exported mo
|
|||
## Arguments
|
||||
|
||||
| Key | Value | Description |
|
||||
| --------- | ------------ | --------------------------------------------------------------------------- |
|
||||
| --------- | ------------ | ------------------------------------------------------------------------------------------- |
|
||||
| `format` | `'openvino'` | format to export to |
|
||||
| `imgsz` | `640` | image size as scalar or (h, w) list, i.e. (640, 480) |
|
||||
| `half` | `False` | FP16 quantization |
|
||||
| `int8` | `False` | INT8 quantization |
|
||||
| `batch` | `1` | [batch size](https://www.ultralytics.com/glossary/batch-size) for inference |
|
||||
| `dynamic` | `False` | allows dynamic input sizes |
|
||||
| `data` | `coco8.yaml` | Path to the dataset configuration file (default: `coco8.yaml`), essential for quantization. |
|
||||
|
||||
!!! note
|
||||
|
||||
When using `data` argument for quantization, please check [Dataset Guide](https://docs.ultralytics.com/datasets/detect) to learn more about the dataset format.
|
||||
|
||||
## Benefits of OpenVINO
|
||||
|
||||
|
|
|
|||
|
|
@ -90,10 +90,15 @@ yolov8n_imx_model
|
|||
When exporting a model to IMX500 format, you can specify various arguments:
|
||||
|
||||
| Key | Value | Description |
|
||||
| -------- | ------ | -------------------------------------------------------- |
|
||||
| -------- | ------------ | -------------------------------------------------------------- |
|
||||
| `format` | `imx` | Format to export to (imx) |
|
||||
| `int8` | `True` | Enable INT8 quantization for the model (default: `True`) |
|
||||
| `imgsz` | `640` | Image size for the model input (default: `640`) |
|
||||
| `data` | `coco8.yaml` | Path to the dataset configuration file (default: `coco8.yaml`) |
|
||||
|
||||
!!! note
|
||||
|
||||
When using `data` argument for quantization, please check [Dataset Guide](https://docs.ultralytics.com/datasets/detect) to learn more about the dataset format.
|
||||
|
||||
## Using IMX500 Export in Deployment
|
||||
|
||||
|
|
|
|||
|
|
@ -13,3 +13,4 @@
|
|||
| `nms` | `bool` | `False` | Adds Non-Maximum Suppression (NMS) to the CoreML export, essential for accurate and efficient detection post-processing. |
|
||||
| `batch` | `int` | `1` | Specifies export model batch inference size or the max number of images the exported model will process concurrently in `predict` mode. |
|
||||
| `device` | `str` | `None` | Specifies the device for exporting: GPU (`device=0`), CPU (`device=cpu`), MPS for Apple silicon (`device=mps`) or DLA for NVIDIA Jetson (`device=dla:0` or `device=dla:1`). |
|
||||
| `data` | `str` | `coco8.yaml` | Path to the dataset configuration file (default: `coco8.yaml`), essential for quantization. |
|
||||
|
|
|
|||
|
|
@ -1,18 +1,18 @@
|
|||
| Format | `format` Argument | Model | Metadata | Arguments |
|
||||
| ------------------------------------------------- | ----------------- | ----------------------------------------------- | -------- | --------------------------------------------------------------------------- |
|
||||
| ------------------------------------------------- | ----------------- | ----------------------------------------------- | -------- | ----------------------------------------------------------------------------------- |
|
||||
| [PyTorch](https://pytorch.org/) | - | `{{ model_name or "yolo11n" }}.pt` | ✅ | - |
|
||||
| [TorchScript](../integrations/torchscript.md) | `torchscript` | `{{ model_name or "yolo11n" }}.torchscript` | ✅ | `imgsz`, `optimize`, `nms`, `batch` |
|
||||
| [ONNX](../integrations/onnx.md) | `onnx` | `{{ model_name or "yolo11n" }}.onnx` | ✅ | `imgsz`, `half`, `dynamic`, `simplify`, `opset`, `nms`, `batch` |
|
||||
| [OpenVINO](../integrations/openvino.md) | `openvino` | `{{ model_name or "yolo11n" }}_openvino_model/` | ✅ | `imgsz`, `half`, `dynamic`, `int8`, `nms`, `batch` |
|
||||
| [TensorRT](../integrations/tensorrt.md) | `engine` | `{{ model_name or "yolo11n" }}.engine` | ✅ | `imgsz`, `half`, `dynamic`, `simplify`, `workspace`, `int8`, `nms`, `batch` |
|
||||
| [OpenVINO](../integrations/openvino.md) | `openvino` | `{{ model_name or "yolo11n" }}_openvino_model/` | ✅ | `imgsz`, `half`, `dynamic`, `int8`, `nms`, `batch`, `data` |
|
||||
| [TensorRT](../integrations/tensorrt.md) | `engine` | `{{ model_name or "yolo11n" }}.engine` | ✅ | `imgsz`, `half`, `dynamic`, `simplify`, `workspace`, `int8`, `nms`, `batch`, `data` |
|
||||
| [CoreML](../integrations/coreml.md) | `coreml` | `{{ model_name or "yolo11n" }}.mlpackage` | ✅ | `imgsz`, `half`, `int8`, `nms`, `batch` |
|
||||
| [TF SavedModel](../integrations/tf-savedmodel.md) | `saved_model` | `{{ model_name or "yolo11n" }}_saved_model/` | ✅ | `imgsz`, `keras`, `int8`, `nms`, `batch` |
|
||||
| [TF GraphDef](../integrations/tf-graphdef.md) | `pb` | `{{ model_name or "yolo11n" }}.pb` | ❌ | `imgsz`, `batch` |
|
||||
| [TF Lite](../integrations/tflite.md) | `tflite` | `{{ model_name or "yolo11n" }}.tflite` | ✅ | `imgsz`, `half`, `int8`, `nms`, `batch` |
|
||||
| [TF Lite](../integrations/tflite.md) | `tflite` | `{{ model_name or "yolo11n" }}.tflite` | ✅ | `imgsz`, `half`, `int8`, `nms`, `batch`, `data` |
|
||||
| [TF Edge TPU](../integrations/edge-tpu.md) | `edgetpu` | `{{ model_name or "yolo11n" }}_edgetpu.tflite` | ✅ | `imgsz` |
|
||||
| [TF.js](../integrations/tfjs.md) | `tfjs` | `{{ model_name or "yolo11n" }}_web_model/` | ✅ | `imgsz`, `half`, `int8`, `nms`, `batch` |
|
||||
| [PaddlePaddle](../integrations/paddlepaddle.md) | `paddle` | `{{ model_name or "yolo11n" }}_paddle_model/` | ✅ | `imgsz`, `batch` |
|
||||
| [MNN](../integrations/mnn.md) | `mnn` | `{{ model_name or "yolo11n" }}.mnn` | ✅ | `imgsz`, `batch`, `int8`, `half` |
|
||||
| [NCNN](../integrations/ncnn.md) | `ncnn` | `{{ model_name or "yolo11n" }}_ncnn_model/` | ✅ | `imgsz`, `half`, `batch` |
|
||||
| [IMX500](../integrations/sony-imx500.md) | `imx` | `{{ model_name or "yolov8n" }}_imx_model/` | ✅ | `imgsz`, `int8` |
|
||||
| [IMX500](../integrations/sony-imx500.md) | `imx` | `{{ model_name or "yolov8n" }}_imx_model/` | ✅ | `imgsz`, `int8`, `data` |
|
||||
| [RKNN](../integrations/rockchip-rknn.md) | `rknn` | `{{ model_name or "yolo11n" }}_rknn_model/` | ✅ | `imgsz`, `batch`, `name` |
|
||||
|
|
|
|||
|
|
@ -367,18 +367,18 @@
|
|||
"| [PyTorch](https://pytorch.org/) | - | `yolo11n.pt` | ✅ | - |\n",
|
||||
"| [TorchScript](https://docs.ultralytics.com/integrations/torchscript) | `torchscript` | `yolo11n.torchscript` | ✅ | `imgsz`, `optimize`, `batch` |\n",
|
||||
"| [ONNX](https://docs.ultralytics.com/integrations/onnx) | `onnx` | `yolo11n.onnx` | ✅ | `imgsz`, `half`, `dynamic`, `simplify`, `opset`, `batch` |\n",
|
||||
"| [OpenVINO](https://docs.ultralytics.com/integrations/openvino) | `openvino` | `yolo11n_openvino_model/` | ✅ | `imgsz`, `half`, `dynamic`, `int8`, `batch` |\n",
|
||||
"| [TensorRT](https://docs.ultralytics.com/integrations/tensorrt) | `engine` | `yolo11n.engine` | ✅ | `imgsz`, `half`, `dynamic`, `simplify`, `workspace`, `int8`, `batch` |\n",
|
||||
"| [OpenVINO](https://docs.ultralytics.com/integrations/openvino) | `openvino` | `yolo11n_openvino_model/` | ✅ | `imgsz`, `half`, `dynamic`, `int8`, `batch`, `data` |\n",
|
||||
"| [TensorRT](https://docs.ultralytics.com/integrations/tensorrt) | `engine` | `yolo11n.engine` | ✅ | `imgsz`, `half`, `dynamic`, `simplify`, `workspace`, `int8`, `batch`, `data` |\n",
|
||||
"| [CoreML](https://docs.ultralytics.com/integrations/coreml) | `coreml` | `yolo11n.mlpackage` | ✅ | `imgsz`, `half`, `int8`, `nms`, `batch` |\n",
|
||||
"| [TF SavedModel](https://docs.ultralytics.com/integrations/tf-savedmodel) | `saved_model` | `yolo11n_saved_model/` | ✅ | `imgsz`, `keras`, `int8`, `batch` |\n",
|
||||
"| [TF GraphDef](https://docs.ultralytics.com/integrations/tf-graphdef) | `pb` | `yolo11n.pb` | ❌ | `imgsz`, `batch` |\n",
|
||||
"| [TF Lite](https://docs.ultralytics.com/integrations/tflite) | `tflite` | `yolo11n.tflite` | ✅ | `imgsz`, `half`, `int8`, `batch` |\n",
|
||||
"| [TF Lite](https://docs.ultralytics.com/integrations/tflite) | `tflite` | `yolo11n.tflite` | ✅ | `imgsz`, `half`, `int8`, `batch`, `data` |\n",
|
||||
"| [TF Edge TPU](https://docs.ultralytics.com/integrations/edge-tpu) | `edgetpu` | `yolo11n_edgetpu.tflite` | ✅ | `imgsz` |\n",
|
||||
"| [TF.js](https://docs.ultralytics.com/integrations/tfjs) | `tfjs` | `yolo11n_web_model/` | ✅ | `imgsz`, `half`, `int8`, `batch` |\n",
|
||||
"| [PaddlePaddle](https://docs.ultralytics.com/integrations/paddlepaddle) | `paddle` | `yolo11n_paddle_model/` | ✅ | `imgsz`, `batch` |\n",
|
||||
"| [MNN](https://docs.ultralytics.com/integrations/mnn) | `mnn` | `yolo11n.mnn` | ✅ | `imgsz`, `batch`, `int8`, `half` |\n",
|
||||
"| [NCNN](https://docs.ultralytics.com/integrations/ncnn) | `ncnn` | `yolo11n_ncnn_model/` | ✅ | `imgsz`, `half`, `batch` |\n",
|
||||
"| [IMX500](https://docs.ultralytics.com/integrations/sony-imx500) | `imx` | `yolov8n_imx_model/` | ✅ | `imgsz`, `int8` |\n",
|
||||
"| [IMX500](https://docs.ultralytics.com/integrations/sony-imx500) | `imx` | `yolov8n_imx_model/` | ✅ | `imgsz`, `int8`, `data` |\n",
|
||||
"| [RKNN](https://docs.ultralytics.com/integrations/rockchip-rknn) | `rknn` | `yolo11n_rknn_model/` | ✅ | `imgsz`, `batch`, `name` |"
|
||||
],
|
||||
"metadata": {
|
||||
|
|
|
|||
|
|
@ -1,6 +1,6 @@
|
|||
# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
|
||||
|
||||
__version__ = "8.3.69"
|
||||
__version__ = "8.3.70"
|
||||
|
||||
import os
|
||||
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue