Skip to content

Commit 1cd8345

Browse files
hemalshahNVhemalshahNV
authored
Merge pull request #18 from NVIDIA-ISAAC-ROS/release-dp-1.0.1
Isaac ROS 0.10.1 (DP)
2 parents b55bb4d + 1f9bac0 commit 1cd8345

File tree

4 files changed

+41
-41
lines changed

4 files changed

+41
-41
lines changed

‎README.md‎

Lines changed: 31 additions & 31 deletions
Original file line numberDiff line numberDiff line change
@@ -171,7 +171,7 @@ To simplify development, we strongly recommend leveraging the Isaac ROS Dev Dock
171171
nanosec: 0
172172
frame_id: ''
173173
tensors:
174-
- name: output
174+
- name: output_tensor
175175
shape:
176176
rank: 4
177177
dims:
@@ -240,7 +240,7 @@ To simplify development, we strongly recommend leveraging the Isaac ROS Dev Dock
240240
nanosec: 0
241241
frame_id: ''
242242
tensors:
243-
- name: output
243+
- name: output_tensor
244244
shape:
245245
rank: 4
246246
dims:
@@ -321,18 +321,18 @@ ros2 launch isaac_ros_triton isaac_ros_triton.launch.py model_name:=<model_name>
321321

322322
#### ROS Parameters
323323

324-
| ROS Parameter | Type | Default | Description |
325-
| ------------------------- | ------------- | ------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ |
326-
| `model_repository_paths` | `string list` | `['']` | The absolute paths to your model repositories in your local file system (the structure should follow Triton requirements) <br/> E.g. `['/tmp/models']` |
327-
| `model_name` | `string` | `""` | The name of your model. Under `model_repository_paths`, there should be a directory with this name, and it should align with the model name in the model configuration under this directory <br/> E.g. `mobilenetv2-1.0_triton_onnx` |
328-
| `max_batch_size` | `uint16_t` | `8` | The maximum batch size allowed for the model. It should align with the model configuration |
329-
| `num_concurrent_requests` | `uint16_t` | `10` | The number of requests the Triton server can take at a time. This should be set according to the tensor publisher frequency |
330-
| `input_tensor_names` | `string list` | `['']` | A list of tensor names to be bound to specified input bindings names. Bindings occur in sequential order, so the first name here will be mapped to the first name in input_binding_names <br/> E.g. `['input']` |
331-
| `input_binding_names` | `string list` | `['']` | A list of input tensor binding names specified by model <br/> E.g. `['data']` |
332-
| `input_tensor_formats` | `string list` | `['']` | A list of input tensor nitros formats. This should be given in sequential order <br/> E.g. `['nitros_tensor_list_nchw_rgb_f32']` |
333-
| `output_tensor_names` | `string list` | `['']` | A list of tensor names to be bound to specified output binding names <br/> E.g. `['output']` |
334-
| `output_binding_names` | `string list` | `['']` | A list of tensor names to be bound to specified output binding names <br/> E.g. `['mobilenetv20_output_flatten0_reshape0']` |
335-
| `output_tensor_formats` | `string list` | `['']` | A list of input tensor nitros formats. This should be given in sequential order <br/> E.g. `[nitros_tensor_list_nchw_rgb_f32]` |
324+
| ROS Parameter | Type | Default | Description |
325+
| ------------------------- | ------------- | ------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
326+
| `model_repository_paths` | `string list` | `['']` | The absolute paths to your model repositories in your local file system (the structure should follow Triton requirements) <br/> E.g. `['/tmp/models']` |
327+
| `model_name` | `string` | `""` | The name of your model. Under `model_repository_paths`, there should be a directory with this name, and it should align with the model name in the model configuration under this directory <br/> E.g. `peoplesemsegnet_shuffleseg` |
328+
| `max_batch_size` | `uint16_t` | `8` | The maximum batch size allowed for the model. It should align with the model configuration |
329+
| `num_concurrent_requests` | `uint16_t` | `10` | The number of requests the Triton server can take at a time. This should be set according to the tensor publisher frequency |
330+
| `input_tensor_names` | `string list` | `['input_tensor']` | A list of tensor names to be bound to specified input bindings names. Bindings occur in sequential order, so the first name here will be mapped to the first name in input_binding_names |
331+
| `input_binding_names` | `string list` | `['']` | A list of input tensor binding names specified by model <br/> E.g. `['input_2:0']` |
332+
| `input_tensor_formats` | `string list` | `['']` | A list of input tensor nitros formats. This should be given in sequential order <br/> E.g. `['nitros_tensor_list_nchw_rgb_f32']` |
333+
| `output_tensor_names` | `string list` | `['output_tensor']` | A list of tensor names to be bound to specified output binding names |
334+
| `output_binding_names` | `string list` | `['']` | A list of tensor names to be bound to specified output binding names <br/> E.g. `['argmax_1']` |
335+
| `output_tensor_formats` | `string list` | `['']` | A list of input tensor nitros formats. This should be given in sequential order <br/> E.g. `[nitros_tensor_list_nchw_rgb_f32]` |
336336

337337

338338
#### ROS Topics Subscribed
@@ -353,23 +353,23 @@ ros2 launch isaac_ros_tensor_rt isaac_ros_tensor_rt.launch.py model_file_path:=<
353353

354354
#### ROS Parameters
355355

356-
| ROS Parameter | Type | Default | Description |
357-
| ------------------------- | ------------- | ---------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
358-
| `model_file_path` | `string` | `model.onnx` | The absolute path to your model file in the local file system (the model file must be .onnx) <br/> E.g. `model.onnx` |
359-
| `engine_file_path` | `string` | `/tmp/trt_engine.plan` | The absolute path to either where you want your TensorRT engine plan to be generated (from your model file) or where your pre-generated engine plan file is located <br/> E.g. `model.plan` |
360-
| `force_engine_update` | `bool` | `true` | If set to true, the node will always try to generate a TensorRT engine plan from your model file and needs to be set to false to use the pre-generated TensorRT engine plan |
361-
| `input_tensor_names` | `string list` | `['']` | A list of tensor names to be bound to specified input bindings names. Bindings occur in sequential order, so the first name here will be mapped to the first name in input_binding_names <br/> E.g. `['input']` |
362-
| `input_binding_names` | `string list` | `['']` | A list of input tensor binding names specified by model <br/> E.g. `['data']` |
363-
| `input_tensor_formats` | `string list` | `['']` | A list of input tensor nitros formats. This should be given in sequential order <br/> E.g. `['nitros_tensor_list_nchw_rgb_f32']` |
364-
| `output_tensor_names` | `string list` | `['']` | A list of tensor names to be bound to specified output binding names <br/> E.g. `['output']` |
365-
| `output_binding_names` | `string list` | `['']` | A list of tensor names to be bound to specified output binding names <br/> E.g. `['mobilenetv20_output_flatten0_reshape0']` |
366-
| `output_tensor_formats` | `string list` | `['']` | A list of input tensor nitros formats. This should be given in sequential order <br/> E.g. `[nitros_tensor_list_nchw_rgb_f32]` |
367-
| `verbose` | `bool` | `true` | If set to true, the node will enable verbose logging to console from the internal TensorRT execution |
368-
| `max_workspace_size` | `int64_t` | `67108864l` | The size of the working space in bytes |
369-
| `max_batch_size` | `int32_t` | `1` | The maximum possible batch size in case the first dimension is dynamic and used as the batch size |
370-
| `dla_core` | `int64_t` | `-1` | The DLA Core to use. Fallback to GPU is always enabled. The default setting is GPU only |
371-
| `enable_fp16` | `bool` | `true` | Enables building a TensorRT engine plan file which uses FP16 precision for inference. If this setting is false, the plan file will use FP32 precision |
372-
| `relaxed_dimension_check` | `bool` | `true` | Ignores dimensions of 1 for the input-tensor dimension check |
356+
| ROS Parameter | Type | Default | Description |
357+
| ------------------------- | ------------- | ---------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
358+
| `model_file_path` | `string` | `model.onnx` | The absolute path to your model file in the local file system (the model file must be .onnx) <br/> E.g. `model.onnx` |
359+
| `engine_file_path` | `string` | `/tmp/trt_engine.plan` | The absolute path to either where you want your TensorRT engine plan to be generated (from your model file) or where your pre-generated engine plan file is located <br/> E.g. `model.plan` |
360+
| `force_engine_update` | `bool` | `true` | If set to true, the node will always try to generate a TensorRT engine plan from your model file and needs to be set to false to use the pre-generated TensorRT engine plan |
361+
| `input_tensor_names` | `string list` | `['input_tensor']` | A list of tensor names to be bound to specified input bindings names. Bindings occur in sequential order, so the first name here will be mapped to the first name in input_binding_names |
362+
| `input_binding_names` | `string list` | `['']` | A list of input tensor binding names specified by model <br/> E.g. `['input_2:0']` |
363+
| `input_tensor_formats` | `string list` | `['']` | A list of input tensor nitros formats. This should be given in sequential order <br/> E.g. `['nitros_tensor_list_nchw_rgb_f32']` |
364+
| `output_tensor_names` | `string list` | `['output_tensor']` | A list of tensor names to be bound to specified output binding names |
365+
| `output_binding_names` | `string list` | `['']` | A list of tensor names to be bound to specified output binding names <br/> E.g. `['argmax_1']` |
366+
| `output_tensor_formats` | `string list` | `['']` | A list of input tensor nitros formats. This should be given in sequential order <br/> E.g. `[nitros_tensor_list_nchw_rgb_f32]` |
367+
| `verbose` | `bool` | `true` | If set to true, the node will enable verbose logging to console from the internal TensorRT execution |
368+
| `max_workspace_size` | `int64_t` | `67108864l` | The size of the working space in bytes |
369+
| `max_batch_size` | `int32_t` | `1` | The maximum possible batch size in case the first dimension is dynamic and used as the batch size |
370+
| `dla_core` | `int64_t` | `-1` | The DLA Core to use. Fallback to GPU is always enabled. The default setting is GPU only |
371+
| `enable_fp16` | `bool` | `true` | Enables building a TensorRT engine plan file which uses FP16 precision for inference. If this setting is false, the plan file will use FP32 precision |
372+
| `relaxed_dimension_check` | `bool` | `true` | Ignores dimensions of 1 for the input-tensor dimension check |
373373
#### ROS Topics Subscribed
374374
| ROS Topic | Type | Description |
375375
| ------------ | ----------------------------------------------------------------------------------------------------------------------------------------------------------------- | ----------------------- |

‎isaac_ros_dnn_inference_test/src/test_tensor_publisher_node.cpp‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -43,7 +43,7 @@ TestTensorPublisherNode::TestTensorPublisherNode(
4343
const rclcpp::NodeOptions & options)
4444
: Node("test_tensor_publisher", options),
4545
dimensions_(declare_parameter<std::vector<int64_t>>("dimensions", {1, 3, 224, 224})),
46-
tensor_name_(declare_parameter<std::string>("tensor_name", "input")),
46+
tensor_name_(declare_parameter<std::string>("tensor_name", "input_tensor")),
4747
data_type_(declare_parameter<int>("data_type", 9)),
4848
length_(declare_parameter<int>("length", calculate_length(dimensions_))),
4949
rank_(declare_parameter<int>("rank", 4)),

‎isaac_ros_tensor_rt/launch/isaac_ros_tensor_rt.launch.py‎

Lines changed: 5 additions & 5 deletions
Original file line numberDiff line numberDiff line change
@@ -27,20 +27,20 @@ def generate_launch_description():
2727
description='The absolute file path to the TensorRT engine file'),
2828
DeclareLaunchArgument(
2929
'input_tensor_names',
30-
default_value='["input"]',
30+
default_value='["input_tensor"]',
3131
description='A list of tensor names to bound to the specified input binding names'),
3232
DeclareLaunchArgument(
3333
'input_binding_names',
34-
default_value='["data"]',
34+
default_value='[""]',
3535
description='A list of input tensor binding names (specified by model)'),
3636
DeclareLaunchArgument(
3737
'output_tensor_names',
38-
default_value='["output"]',
38+
default_value='["output_tensor"]',
3939
description='A list of tensor names to bound to the specified output binding names'),
4040
DeclareLaunchArgument(
4141
'output_binding_names',
42-
default_value='["mobilenetv20_output_flatten0_reshape0"]',
43-
description='A list of output tensor binding names (specified by model)'),
42+
default_value='[""]',
43+
description='A list of output tensor binding names (specified by model)'),
4444
DeclareLaunchArgument(
4545
'verbose',
4646
default_value='False',

0 commit comments

Comments
 (0)