From 39b037408f3de51f05ad343f971c0c76514ae96f Mon Sep 17 00:00:00 2001 From: Lakshantha Dissanayake Date: Wed, 5 Feb 2025 18:43:02 -0800 Subject: [PATCH] Add Export Arguments tables to all Export docs (#18952) Co-authored-by: UltralyticsAssistant Co-authored-by: Francesco Mattioli Co-authored-by: Glenn Jocher --- docs/en/integrations/coreml.md | 11 +++++++++++ docs/en/integrations/edge-tpu.md | 9 ++++++++- docs/en/integrations/mnn.md | 12 +++++++++++- docs/en/integrations/ncnn.md | 11 ++++++++++- docs/en/integrations/onnx.md | 13 +++++++++++++ docs/en/integrations/openvino.md | 25 ++++++++++++------------- docs/en/integrations/paddlepaddle.md | 10 +++++++++- docs/en/integrations/rockchip-rknn.md | 9 +++++++++ docs/en/integrations/sony-imx500.md | 20 ++++++++------------ docs/en/integrations/tensorrt.md | 15 +++++++++++++++ docs/en/integrations/tf-graphdef.md | 10 +++++++++- docs/en/integrations/tf-savedmodel.md | 13 ++++++++++++- docs/en/integrations/tfjs.md | 13 ++++++++++++- docs/en/integrations/tflite.md | 12 ++++++++++++ docs/en/integrations/torchscript.md | 10 ++++++++++ docs/en/macros/export-args.md | 2 +- 16 files changed, 162 insertions(+), 33 deletions(-) diff --git a/docs/en/integrations/coreml.md b/docs/en/integrations/coreml.md index 41ff20d1..284d27f1 100644 --- a/docs/en/integrations/coreml.md +++ b/docs/en/integrations/coreml.md @@ -105,6 +105,17 @@ Before diving into the usage instructions, be sure to check out the range of [YO yolo predict model=yolo11n.mlpackage source='https://ultralytics.com/images/bus.jpg' ``` +### Export Arguments + +| Argument | Type | Default | Description | +| -------- | ---------------- | -------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | +| `format` | `str` | `coreml` | Target format for the exported model, defining compatibility with various deployment environments. | +| `imgsz` | `int` or `tuple` | `640` | Desired image size for the model input. Can be an integer for square images or a tuple `(height, width)` for specific dimensions. | +| `half` | `bool` | `False` | Enables FP16 (half-precision) quantization, reducing model size and potentially speeding up inference on supported hardware. | +| `int8` | `bool` | `False` | Activates INT8 quantization, further compressing the model and speeding up inference with minimal [accuracy](https://www.ultralytics.com/glossary/accuracy) loss, primarily for edge devices. | +| `nms` | `bool` | `False` | Adds Non-Maximum Suppression (NMS), 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. | + For more details about the export process, visit the [Ultralytics documentation page on exporting](../modes/export.md). ## Deploying Exported YOLO11 CoreML Models diff --git a/docs/en/integrations/edge-tpu.md b/docs/en/integrations/edge-tpu.md index f8821e47..eabb093a 100644 --- a/docs/en/integrations/edge-tpu.md +++ b/docs/en/integrations/edge-tpu.md @@ -95,7 +95,14 @@ Before diving into the usage instructions, it's important to note that while all yolo predict model=yolo11n_full_integer_quant_edgetpu.tflite source='https://ultralytics.com/images/bus.jpg' ``` -For more details about supported export options, visit the [Ultralytics documentation page on deployment options](../guides/model-deployment-options.md). +### Export Arguments + +| Argument | Type | Default | Description | +| -------- | ---------------- | --------- | --------------------------------------------------------------------------------------------------------------------------------- | +| `format` | `str` | `edgetpu` | Target format for the exported model, defining compatibility with various deployment environments. | +| `imgsz` | `int` or `tuple` | `640` | Desired image size for the model input. Can be an integer for square images or a tuple `(height, width)` for specific dimensions. | + +For more details about the export process, visit the [Ultralytics documentation page on exporting](../modes/export.md). ## Deploying Exported YOLO11 TFLite Edge TPU Models diff --git a/docs/en/integrations/mnn.md b/docs/en/integrations/mnn.md index bd92c9a8..7c2aeaf9 100644 --- a/docs/en/integrations/mnn.md +++ b/docs/en/integrations/mnn.md @@ -66,7 +66,17 @@ Before diving into the usage instructions, it's important to note that while all yolo predict model='yolo11n.mnn' source='https://ultralytics.com/images/bus.jpg' ``` -For more details about supported export options, visit the [Ultralytics documentation page on deployment options](../guides/model-deployment-options.md). +### Export Arguments + +| Argument | Type | Default | Description | +| -------- | ---------------- | ------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | +| `format` | `str` | `mnn` | Target format for the exported model, defining compatibility with various deployment environments. | +| `imgsz` | `int` or `tuple` | `640` | Desired image size for the model input. Can be an integer for square images or a tuple `(height, width)` for specific dimensions. | +| `half` | `bool` | `False` | Enables FP16 (half-precision) quantization, reducing model size and potentially speeding up inference on supported hardware. | +| `int8` | `bool` | `False` | Activates INT8 quantization, further compressing the model and speeding up inference with minimal [accuracy](https://www.ultralytics.com/glossary/accuracy) loss, primarily for edge devices. | +| `batch` | `int` | `1` | Specifies export model batch inference size or the max number of images the exported model will process concurrently in `predict` mode. | + +For more details about the export process, visit the [Ultralytics documentation page on exporting](../modes/export.md). ### MNN-Only Inference diff --git a/docs/en/integrations/ncnn.md b/docs/en/integrations/ncnn.md index 9dc13f96..bba42562 100644 --- a/docs/en/integrations/ncnn.md +++ b/docs/en/integrations/ncnn.md @@ -97,7 +97,16 @@ Before diving into the usage instructions, it's important to note that while all yolo predict model='./yolo11n_ncnn_model' source='https://ultralytics.com/images/bus.jpg' ``` -For more details about supported export options, visit the [Ultralytics documentation page on deployment options](../guides/model-deployment-options.md). +### Export Arguments + +| Argument | Type | Default | Description | +| -------- | ---------------- | ------- | --------------------------------------------------------------------------------------------------------------------------------------- | +| `format` | `str` | `ncnn` | Target format for the exported model, defining compatibility with various deployment environments. | +| `imgsz` | `int` or `tuple` | `640` | Desired image size for the model input. Can be an integer for square images or a tuple `(height, width)` for specific dimensions. | +| `half` | `bool` | `False` | Enables FP16 (half-precision) quantization, reducing model size and potentially speeding up inference on supported hardware. | +| `batch` | `int` | `1` | Specifies export model batch inference size or the max number of images the exported model will process concurrently in `predict` mode. | + +For more details about the export process, visit the [Ultralytics documentation page on exporting](../modes/export.md). ## Deploying Exported YOLO11 NCNN Models diff --git a/docs/en/integrations/onnx.md b/docs/en/integrations/onnx.md index fbff328d..3f5eff4c 100644 --- a/docs/en/integrations/onnx.md +++ b/docs/en/integrations/onnx.md @@ -113,6 +113,19 @@ Before diving into the usage instructions, be sure to check out the range of [YO yolo predict model=yolo11n.onnx source='https://ultralytics.com/images/bus.jpg' ``` +### Export Arguments + +| Argument | Type | Default | Description | +| ---------- | ---------------- | ------- | ------------------------------------------------------------------------------------------------------------------------------------------- | +| `format` | `str` | `onnx` | Target format for the exported model, defining compatibility with various deployment environments. | +| `imgsz` | `int` or `tuple` | `640` | Desired image size for the model input. Can be an integer for square images or a tuple `(height, width)` for specific dimensions. | +| `half` | `bool` | `False` | Enables FP16 (half-precision) quantization, reducing model size and potentially speeding up inference on supported hardware. | +| `dynamic` | `bool` | `False` | Allows dynamic input sizes, enhancing flexibility in handling varying image dimensions. | +| `simplify` | `bool` | `True` | Simplifies the model graph with `onnxslim`, potentially improving performance and compatibility. | +| `opset` | `int` | `None` | Specifies the ONNX opset version for compatibility with different ONNX parsers and runtimes. If not set, uses the latest supported version. | +| `nms` | `bool` | `False` | Adds Non-Maximum Suppression (NMS), 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. | + For more details about the export process, visit the [Ultralytics documentation page on exporting](../modes/export.md). ## Deploying Exported YOLO11 ONNX Models diff --git a/docs/en/integrations/openvino.md b/docs/en/integrations/openvino.md index b3fd6a82..a7e834bc 100644 --- a/docs/en/integrations/openvino.md +++ b/docs/en/integrations/openvino.md @@ -57,21 +57,20 @@ Export a YOLOv8n model to OpenVINO format and run inference with the exported mo yolo predict model=yolov8n_openvino_model source='https://ultralytics.com/images/bus.jpg' ``` -## Arguments +## Export 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. | +| Argument | Type | Default | Description | +| --------- | ---------------- | ------------ | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | +| `format` | `str` | `openvino` | Target format for the exported model, defining compatibility with various deployment environments. | +| `imgsz` | `int` or `tuple` | `640` | Desired image size for the model input. Can be an integer for square images or a tuple `(height, width)` for specific dimensions. | +| `half` | `bool` | `False` | Enables FP16 (half-precision) quantization, reducing model size and potentially speeding up inference on supported hardware. | +| `int8` | `bool` | `False` | Activates INT8 quantization, further compressing the model and speeding up inference with minimal [accuracy](https://www.ultralytics.com/glossary/accuracy) loss, primarily for edge devices. | +| `dynamic` | `bool` | `False` | Allows dynamic input sizes, enhancing flexibility in handling varying image dimensions. | +| `nms` | `bool` | `False` | Adds Non-Maximum Suppression (NMS), 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. | +| `data` | `str` | `coco8.yaml` | Path to the [dataset](https://docs.ultralytics.com/datasets) 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. +For more details about the export process, visit the [Ultralytics documentation page on exporting](../modes/export.md). ## Benefits of OpenVINO diff --git a/docs/en/integrations/paddlepaddle.md b/docs/en/integrations/paddlepaddle.md index 77c6164b..5eae09a9 100644 --- a/docs/en/integrations/paddlepaddle.md +++ b/docs/en/integrations/paddlepaddle.md @@ -110,7 +110,15 @@ Before diving into the usage instructions, it's important to note that while all yolo predict model='./yolo11n_paddle_model' source='https://ultralytics.com/images/bus.jpg' ``` -For more details about supported export options, visit the [Ultralytics documentation page on deployment options](../guides/model-deployment-options.md). +### Export Arguments + +| Argument | Type | Default | Description | +| -------- | ---------------- | -------- | --------------------------------------------------------------------------------------------------------------------------------------- | +| `format` | `str` | `paddle` | Target format for the exported model, defining compatibility with various deployment environments. | +| `imgsz` | `int` or `tuple` | `640` | Desired image size for the model input. Can be an integer for square images or a tuple `(height, width)` for specific dimensions. | +| `batch` | `int` | `1` | Specifies export model batch inference size or the max number of images the exported model will process concurrently in `predict` mode. | + +For more details about the export process, visit the [Ultralytics documentation page on exporting](../modes/export.md). ## Deploying Exported YOLO11 PaddlePaddle Models diff --git a/docs/en/integrations/rockchip-rknn.md b/docs/en/integrations/rockchip-rknn.md index 087f942e..c8688975 100644 --- a/docs/en/integrations/rockchip-rknn.md +++ b/docs/en/integrations/rockchip-rknn.md @@ -91,6 +91,15 @@ For detailed instructions and best practices related to the installation process yolo export model=yolo11n.pt format=rknn name=rk3588 # creates '/yolo11n_rknn_model' ``` +### Export Arguments + +| Argument | Type | Default | Description | +| -------- | ---------------- | -------- | --------------------------------------------------------------------------------------------------------------------------------------- | +| `format` | `str` | `rknn` | Target format for the exported model, defining compatibility with various deployment environments. | +| `imgsz` | `int` or `tuple` | `640` | Desired image size for the model input. Can be an integer for square images or a tuple `(height, width)` for specific dimensions. | +| `batch` | `int` | `1` | Specifies export model batch inference size or the max number of images the exported model will process concurrently in `predict` mode. | +| `name` | `str` | `rk3588` | Specifies the Rockchip model (rk3588, rk3576, rk3566, rk3568, rk3562, rv1103, rv1106, rv1103b, rv1106b, rk2118) | + For more details about the export process, visit the [Ultralytics documentation page on exporting](../modes/export.md). ## Deploying Exported YOLO11 RKNN Models diff --git a/docs/en/integrations/sony-imx500.md b/docs/en/integrations/sony-imx500.md index ea0d70de..f0dab01d 100644 --- a/docs/en/integrations/sony-imx500.md +++ b/docs/en/integrations/sony-imx500.md @@ -85,20 +85,16 @@ yolov8n_imx_model └── yolov8n_imx500_model.pbtxt ``` -## Arguments +## Export Arguments -When exporting a model to IMX500 format, you can specify various arguments: +| Argument | Type | Default | Description | +| -------- | ---------------- | ------------ | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | +| `format` | `str` | `imx` | Target format for the exported model, defining compatibility with various deployment environments. | +| `imgsz` | `int` or `tuple` | `640` | Desired image size for the model input. Can be an integer for square images or a tuple `(height, width)` for specific dimensions. | +| `int8` | `bool` | `True` | Activates INT8 quantization, further compressing the model and speeding up inference with minimal [accuracy](https://www.ultralytics.com/glossary/accuracy) loss, primarily for edge devices. | +| `data` | `str` | `coco8.yaml` | Path to the [dataset](https://docs.ultralytics.com/datasets) configuration file (default: `coco8.yaml`), essential for quantization. | -| 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. +For more details about the export process, visit the [Ultralytics documentation page on exporting](../modes/export.md). ## Using IMX500 Export in Deployment diff --git a/docs/en/integrations/tensorrt.md b/docs/en/integrations/tensorrt.md index 59dbb280..5464ad62 100644 --- a/docs/en/integrations/tensorrt.md +++ b/docs/en/integrations/tensorrt.md @@ -107,6 +107,21 @@ Before diving into the usage instructions, be sure to check out the range of [YO yolo predict model=yolo11n.engine source='https://ultralytics.com/images/bus.jpg' ``` +### Export Arguments + +| Argument | Type | Default | Description | +| ----------- | ----------------- | ------------ | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | +| `format` | `str` | `engine` | Target format for the exported model, defining compatibility with various deployment environments. | +| `imgsz` | `int` or `tuple` | `640` | Desired image size for the model input. Can be an integer for square images or a tuple `(height, width)` for specific dimensions. | +| `half` | `bool` | `False` | Enables FP16 (half-precision) quantization, reducing model size and potentially speeding up inference on supported hardware. | +| `int8` | `bool` | `False` | Activates INT8 quantization, further compressing the model and speeding up inference with minimal [accuracy](https://www.ultralytics.com/glossary/accuracy) loss, primarily for edge devices. | +| `dynamic` | `bool` | `False` | Allows dynamic input sizes, enhancing flexibility in handling varying image dimensions. | +| `simplify` | `bool` | `True` | Simplifies the model graph with `onnxslim`, potentially improving performance and compatibility. | +| `workspace` | `float` or `None` | `None` | Sets the maximum workspace size in GiB for TensorRT optimizations, balancing memory usage and performance; use `None` for auto-allocation by TensorRT up to device maximum. | +| `nms` | `bool` | `False` | Adds Non-Maximum Suppression (NMS), 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. | +| `data` | `str` | `coco8.yaml` | Path to the [dataset](https://docs.ultralytics.com/datasets) configuration file (default: `coco8.yaml`), essential for quantization. | + For more details about the export process, visit the [Ultralytics documentation page on exporting](../modes/export.md). ### Exporting TensorRT with INT8 Quantization diff --git a/docs/en/integrations/tf-graphdef.md b/docs/en/integrations/tf-graphdef.md index fd6d86a3..7ae66812 100644 --- a/docs/en/integrations/tf-graphdef.md +++ b/docs/en/integrations/tf-graphdef.md @@ -103,7 +103,15 @@ Before diving into the usage instructions, it's important to note that while all yolo predict model='yolo11n.pb' source='https://ultralytics.com/images/bus.jpg' ``` -For more details about supported export options, visit the [Ultralytics documentation page on deployment options](../guides/model-deployment-options.md). +### Export Arguments + +| Argument | Type | Default | Description | +| -------- | ---------------- | ------- | --------------------------------------------------------------------------------------------------------------------------------------- | +| `format` | `str` | `pb` | Target format for the exported model, defining compatibility with various deployment environments. | +| `imgsz` | `int` or `tuple` | `640` | Desired image size for the model input. Can be an integer for square images or a tuple `(height, width)` for specific dimensions. | +| `batch` | `int` | `1` | Specifies export model batch inference size or the max number of images the exported model will process concurrently in `predict` mode. | + +For more details about the export process, visit the [Ultralytics documentation page on exporting](../modes/export.md). ## Deploying Exported YOLO11 TF GraphDef Models diff --git a/docs/en/integrations/tf-savedmodel.md b/docs/en/integrations/tf-savedmodel.md index 682743b1..bc0d093c 100644 --- a/docs/en/integrations/tf-savedmodel.md +++ b/docs/en/integrations/tf-savedmodel.md @@ -97,7 +97,18 @@ Before diving into the usage instructions, it's important to note that while all yolo predict model='./yolo11n_saved_model' source='https://ultralytics.com/images/bus.jpg' ``` -For more details about supported export options, visit the [Ultralytics documentation page on deployment options](../guides/model-deployment-options.md). +### Export Arguments + +| Argument | Type | Default | Description | +| -------- | ---------------- | ------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | +| `format` | `str` | `saved_model` | Target format for the exported model, defining compatibility with various deployment environments. | +| `imgsz` | `int` or `tuple` | `640` | Desired image size for the model input. Can be an integer for square images or a tuple `(height, width)` for specific dimensions. | +| `keras` | `bool` | `False` | Enables export to Keras format, providing compatibility with TensorFlow serving and APIs. | +| `int8` | `bool` | `False` | Activates INT8 quantization, further compressing the model and speeding up inference with minimal [accuracy](https://www.ultralytics.com/glossary/accuracy) loss, primarily for edge devices. | +| `nms` | `bool` | `False` | Adds Non-Maximum Suppression (NMS), 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. | + +For more details about the export process, visit the [Ultralytics documentation page on exporting](../modes/export.md). ## Deploying Exported YOLO11 TF SavedModel Models diff --git a/docs/en/integrations/tfjs.md b/docs/en/integrations/tfjs.md index a8168215..eec6d301 100644 --- a/docs/en/integrations/tfjs.md +++ b/docs/en/integrations/tfjs.md @@ -95,7 +95,18 @@ Before diving into the usage instructions, it's important to note that while all yolo predict model='./yolo11n_web_model' source='https://ultralytics.com/images/bus.jpg' ``` -For more details about supported export options, visit the [Ultralytics documentation page on deployment options](../guides/model-deployment-options.md). +### Export Arguments + +| Argument | Type | Default | Description | +| -------- | ---------------- | ------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | +| `format` | `str` | `tfjs` | Target format for the exported model, defining compatibility with various deployment environments. | +| `imgsz` | `int` or `tuple` | `640` | Desired image size for the model input. Can be an integer for square images or a tuple `(height, width)` for specific dimensions. | +| `half` | `bool` | `False` | Enables FP16 (half-precision) quantization, reducing model size and potentially speeding up inference on supported hardware. | +| `int8` | `bool` | `False` | Activates INT8 quantization, further compressing the model and speeding up inference with minimal [accuracy](https://www.ultralytics.com/glossary/accuracy) loss, primarily for edge devices. | +| `nms` | `bool` | `False` | Adds Non-Maximum Suppression (NMS), 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. | + +For more details about the export process, visit the [Ultralytics documentation page on exporting](../modes/export.md). ## Deploying Exported YOLO11 TensorFlow.js Models diff --git a/docs/en/integrations/tflite.md b/docs/en/integrations/tflite.md index 9f0ebad1..82b96708 100644 --- a/docs/en/integrations/tflite.md +++ b/docs/en/integrations/tflite.md @@ -101,6 +101,18 @@ Before diving into the usage instructions, it's important to note that while all yolo predict model='yolo11n_float32.tflite' source='https://ultralytics.com/images/bus.jpg' ``` +### Export Arguments + +| Argument | Type | Default | Description | +| -------- | ---------------- | ------------ | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | +| `format` | `str` | `tflite` | Target format for the exported model, defining compatibility with various deployment environments. | +| `imgsz` | `int` or `tuple` | `640` | Desired image size for the model input. Can be an integer for square images or a tuple `(height, width)` for specific dimensions. | +| `half` | `bool` | `False` | Enables FP16 (half-precision) quantization, reducing model size and potentially speeding up inference on supported hardware. | +| `int8` | `bool` | `False` | Activates INT8 quantization, further compressing the model and speeding up inference with minimal [accuracy](https://www.ultralytics.com/glossary/accuracy) loss, primarily for edge devices. | +| `nms` | `bool` | `False` | Adds Non-Maximum Suppression (NMS), 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. | +| `data` | `str` | `coco8.yaml` | Path to the [dataset](https://docs.ultralytics.com/datasets) configuration file (default: `coco8.yaml`), essential for quantization. | + For more details about the export process, visit the [Ultralytics documentation page on exporting](../modes/export.md). ## Deploying Exported YOLO11 TFLite Models diff --git a/docs/en/integrations/torchscript.md b/docs/en/integrations/torchscript.md index 1be1516c..3746e6b6 100644 --- a/docs/en/integrations/torchscript.md +++ b/docs/en/integrations/torchscript.md @@ -105,6 +105,16 @@ Before diving into the usage instructions, it's important to note that while all yolo predict model=yolo11n.torchscript source='https://ultralytics.com/images/bus.jpg' ``` +### Export Arguments + +| Argument | Type | Default | Description | +| ---------- | ---------------- | ------------- | --------------------------------------------------------------------------------------------------------------------------------------- | +| `format` | `str` | `torchscript` | Target format for the exported model, defining compatibility with various deployment environments. | +| `imgsz` | `int` or `tuple` | `640` | Desired image size for the model input. Can be an integer for square images or a tuple `(height, width)` for specific dimensions. | +| `optimize` | `bool` | `False` | Applies optimization for mobile devices, potentially reducing model size and improving performance. | +| `nms` | `bool` | `False` | Adds Non-Maximum Suppression (NMS), 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. | + For more details about the export process, visit the [Ultralytics documentation page on exporting](../modes/export.md). ## Deploying Exported YOLO11 TorchScript Models diff --git a/docs/en/macros/export-args.md b/docs/en/macros/export-args.md index 0af441cb..c708b374 100644 --- a/docs/en/macros/export-args.md +++ b/docs/en/macros/export-args.md @@ -13,4 +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. | +| `data` | `str` | `coco8.yaml` | Path to the [dataset](https://docs.ultralytics.com/datasets) configuration file (default: `coco8.yaml`), essential for quantization. |