Yolo-v5 is not supported #830
Comments
|
It could also mean that the model-files IR-format is newer or older... You can find the IR-format version in the first files within the XML-model-file. |
|
Can you try the model in the latest DL Streamer / DL Streamer Pipeline Server containers: docker pull intel/dlstreamer-pipeline-server:latest native yolo-v5 support in DL Streamer is only recent - and wouldn't be supported with previous versions. Also - on the DL Streamer github issues there are some questions / tips on ensuring the model proc meets the input size of your model (and conversion guidelines). |
|
@nnshah1 This is our current docker file of the pod, where we define Vaserving and openvisualcloud version, now tell me how to proceed with it, tried with latest ovc version as well given in latest smart-city-sample there also support is not available. # smtc_analytics_common_xeon_gst FROM centos:7 as build ARG VA_SERVING_REPO=https://raw.githubusercontent.com/intel/video-analytics-serving RUN mkdir -p /home/vaserving/common/utils && touch /home/vaserving/init.py /home/vaserving/common/init.py /home/vaserving/common/utils/init.py && for x in common/utils/logging.py common/settings.py arguments.py ffmpeg_pipeline.py gstreamer_pipeline.py app_destination.py app_source.py gstreamer_app_destination.py gstreamer_app_source.py model_manager.py pipeline.py pipeline_manager.py pipeline.py schema.py vaserving.py; do curl -sSf -o /home/vaserving/$x -L ${VA_SERVING_REPO}/${VA_SERVING_TAG}/vaserving/$x; done FROM openvisualcloud/xeone3-ubuntu1804-analytics-gst:21.3 RUN apt-get update -qq && apt-get install -y gpg-agent wget software-properties-common && \ COPY --from=build /home/ /home/ ENV FRAMEWORK=gstreamer ARG USER=docker |
|
I have done the IR conversion using OpenVino 21.1. The model is working well with OpenVino But if I run with VA_SERVING_TAG="v0.4.1-beta and openvisualcloud/xeone3-ubuntu1804-analytics-gst:21.3- the Dockerfile that @divdaisymuffin has mentioned above, it is not detecting any thing here is the output: logs of yolov5 (not detecting) here you can see the message has no objects If I run the working model yolov4-tiny then the logs are: here you can see the are detected object in yolov4-tiny logs. yolo-v5 proc-file: |
|
Support for Yolo-V5 was added to DL-Streamer just recently to the "2022.2.0-release". But users reported issues due to e.g. a default model-proc-file provided as part of the release. You might want to check these GitHub issues "dlstreamer/dlstreamer#320" and "dlstreamer/dlstreamer#322". |
Hi @nnshah1 @xwu2git,
We have trained one model in yolo-v5 and converted it to IR INT8, and we found a model-proc file at https://github.com/dlstreamer/dlstreamer/blob/master/samples/gstreamer/model_proc/public/yolo-v5.json,
{ "json_schema_version": "2.2.0", "input_preproc": [], "output_postproc": [ { "converter": "yolo_v5", "iou_threshold": 0.4, "bbox_number_on_cell": 3, "cells_number": 20, "classes": 3, "anchors": [ 10.0, 13.0, 16.0, 30.0, 33.0, 23.0, 30.0, 61.0, 62.0, 45.0, 59.0, 119.0, 116.0, 90.0, 156.0, 198.0, 373.0, 326.0 ], "masks": [ 6, 7, 8, 3, 4, 5, 0, 1, 2 ], "labels":["head", "staff", "customer"] } ] }but when we tried it in Openvisualcloud, the analytics pod starts throwing error
{"levelname": "INFO", "asctime": "2022-11-07 13:04:59,671", "message": "========================", "module": "vaserving"} {"levelname": "INFO", "asctime": "2022-11-07 13:04:59,671", "message": "Options for vaserving.py", "module": "vaserving"} {"levelname": "INFO", "asctime": "2022-11-07 13:04:59,671", "message": "========================", "module": "vaserving"} {"levelname": "INFO", "asctime": "2022-11-07 13:04:59,671", "message": "port == 8080", "module": "vaserving"} {"levelname": "INFO", "asctime": "2022-11-07 13:04:59,671", "message": "framework == gstreamer", "module": "vaserving"} {"levelname": "INFO", "asctime": "2022-11-07 13:04:59,671", "message": "pipeline_dir == /home/pipelines", "module": "vaserving"} {"levelname": "INFO", "asctime": "2022-11-07 13:04:59,671", "message": "model_dir == /home/models", "module": "vaserving"} {"levelname": "INFO", "asctime": "2022-11-07 13:04:59,671", "message": "network_preference == {}", "module": "vaserving"} {"levelname": "INFO", "asctime": "2022-11-07 13:04:59,671", "message": "max_running_pipelines == 1", "module": "vaserving"} {"levelname": "INFO", "asctime": "2022-11-07 13:04:59,671", "message": "log_level == INFO", "module": "vaserving"} {"levelname": "INFO", "asctime": "2022-11-07 13:04:59,671", "message": "config_path == /home/vaserving/..", "module": "vaserving"} {"levelname": "INFO", "asctime": "2022-11-07 13:04:59,671", "message": "ignore_init_errors == False", "module": "vaserving"} {"levelname": "INFO", "asctime": "2022-11-07 13:04:59,671", "message": "========================", "module": "vaserving"} {"levelname": "INFO", "asctime": "2022-11-07 13:04:59,671", "message": "==============", "module": "model_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:04:59,671", "message": "Loading Models", "module": "model_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:04:59,671", "message": "==============", "module": "model_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:04:59,671", "message": "Loading Models from Path /home/models", "module": "model_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:04:59,672", "message": "Loading Model: head_yolov5_tiny_608to416_FP32 version: 1 type: IntelDLDT from {'FP32': '/home/models/head_yolov5_tiny_608to416_FP32/1/FP32/dbest.xml', 'model-proc': '/home/models/head_yolov5_tiny_608to416_FP32/1/head_yolov4_tiny_608to416_INT8.json'}", "module": "model_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:04:59,672", "message": "Loading Model: head_yolov4_tiny_608to416_INT8 version: 1 type: IntelDLDT from {'INT8': '/home/models/head_yolov4_tiny_608to416_INT8/1/INT8/head_yolov4_tiny_608to416_INT8.xml', 'model-proc': '/home/models/head_yolov4_tiny_608to416_INT8/1/head_yolov4_tiny_608to416_INT8.json'}", "module": "model_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:04:59,673", "message": "Loading Model: head_yolov4_tiny_608to416_default_anchors_mask_012_FP32 version: 1 type: IntelDLDT from {'FP32': '/home/models/head_yolov4_tiny_608to416_default_anchors_mask_012_FP32/1/FP32/head_yolov4_tiny_608to416_default_anchors_mask_012_FP32.xml', 'model-proc': '/home/models/head_yolov4_tiny_608to416_default_anchors_mask_012_FP32/1/head_yolov4_tiny_608to416_default_anchors_mask_012_FP32.json'}", "module": "model_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:04:59,673", "message": "Loading Model: head_yolov4_tiny_608to416_default_anchors_mask_012_FP32copy version: 1 type: IntelDLDT from {'FP32': '/home/models/head_yolov4_tiny_608to416_default_anchors_mask_012_FP32copy/1/FP32/head_yolov4_tiny_608to416_INT8.xml', 'model-proc': '/home/models/head_yolov4_tiny_608to416_default_anchors_mask_012_FP32copy/1/head_yolov4_tiny_608to416_default_anchors_mask_012_FP32.json'}", "module": "model_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:04:59,673", "message": "========================", "module": "model_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:04:59,673", "message": "Completed Loading Models", "module": "model_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:04:59,673", "message": "========================", "module": "model_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:04:59,673", "message": "=================", "module": "pipeline_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:04:59,673", "message": "Loading Pipelines", "module": "pipeline_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:04:59,673", "message": "=================", "module": "pipeline_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:05:00,115", "message": "Loading Pipelines from Config Path /home/pipelines", "module": "pipeline_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:05:00,140", "message": "Loading Pipeline: algorithmName version: 3 type: GStreamer from /home/pipelines/algorithmName/3/pipeline.json", "module": "pipeline_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:05:00,142", "message": "Loading Pipeline: algorithmName version: 2 type: GStreamer from /home/pipelines/algorithmName/2/pipeline.json", "module": "pipeline_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:05:00,147", "message": "Loading Pipeline: algorithmName version: 1 type: GStreamer from /home/pipelines/algorithmName/1/pipeline.json", "module": "pipeline_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:05:00,148", "message": "===========================", "module": "pipeline_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:05:00,148", "message": "Completed Loading Pipelines", "module": "pipeline_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:05:00,148", "message": "===========================", "module": "pipeline_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:05:00,148", "message": "Creating Instance of Pipeline algorithmName/1", "module": "pipeline_manager"} 0:00:01.752592619 1 0x115c4a0 ERROR libav :0:: Could not find ref with POC 9 terminate called after throwing an instance of 'std::_Nested_exception<std::invalid_argument>' what(): Parser for 2.2.0 version not foundParser for 2.2.0 version not found
Then we used json_schema version 2.0.0 and model-proc looks like this
{ "json_schema_version": "2.0.0", "input_preproc": [], "output_postproc": [ { "converter": "yolo_v5", "iou_threshold": 0.4, "bbox_number_on_cell": 3, "cells_number": 20, "classes": 3, "anchors": [ 10.0, 13.0, 16.0, 30.0, 33.0, 23.0, 30.0, 61.0, 62.0, 45.0, 59.0, 119.0, 116.0, 90.0, 156.0, 198.0, 373.0, 326.0 ], "masks": [ 6, 7, 8, 3, 4, 5, 0, 1, 2 ], "labels":["head", "staff", "customer"] } ] }Then its able to process model -proc but throughs this error
Successfully tagged sd:latest root@aivid15:/home/aivid15/Desktop/develop/DLEngine# docker run sd {"levelname": "INFO", "asctime": "2022-11-07 13:07:01,058", "message": "========================", "module": "vaserving"} {"levelname": "INFO", "asctime": "2022-11-07 13:07:01,059", "message": "Options for vaserving.py", "module": "vaserving"} {"levelname": "INFO", "asctime": "2022-11-07 13:07:01,059", "message": "========================", "module": "vaserving"} {"levelname": "INFO", "asctime": "2022-11-07 13:07:01,059", "message": "port == 8080", "module": "vaserving"} {"levelname": "INFO", "asctime": "2022-11-07 13:07:01,059", "message": "framework == gstreamer", "module": "vaserving"} {"levelname": "INFO", "asctime": "2022-11-07 13:07:01,059", "message": "pipeline_dir == /home/pipelines", "module": "vaserving"} {"levelname": "INFO", "asctime": "2022-11-07 13:07:01,059", "message": "model_dir == /home/models", "module": "vaserving"} {"levelname": "INFO", "asctime": "2022-11-07 13:07:01,059", "message": "network_preference == {}", "module": "vaserving"} {"levelname": "INFO", "asctime": "2022-11-07 13:07:01,059", "message": "max_running_pipelines == 1", "module": "vaserving"} {"levelname": "INFO", "asctime": "2022-11-07 13:07:01,059", "message": "log_level == INFO", "module": "vaserving"} {"levelname": "INFO", "asctime": "2022-11-07 13:07:01,059", "message": "config_path == /home/vaserving/..", "module": "vaserving"} {"levelname": "INFO", "asctime": "2022-11-07 13:07:01,059", "message": "ignore_init_errors == False", "module": "vaserving"} {"levelname": "INFO", "asctime": "2022-11-07 13:07:01,059", "message": "========================", "module": "vaserving"} {"levelname": "INFO", "asctime": "2022-11-07 13:07:01,059", "message": "==============", "module": "model_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:07:01,059", "message": "Loading Models", "module": "model_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:07:01,059", "message": "==============", "module": "model_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:07:01,059", "message": "Loading Models from Path /home/models", "module": "model_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:07:01,060", "message": "Loading Model: head_yolov5_tiny_608to416_FP32 version: 1 type: IntelDLDT from {'FP32': '/home/models/head_yolov5_tiny_608to416_FP32/1/FP32/dbest.xml', 'model-proc': '/home/models/head_yolov5_tiny_608to416_FP32/1/head_yolov4_tiny_608to416_INT8.json'}", "module": "model_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:07:01,060", "message": "Loading Model: head_yolov4_tiny_608to416_INT8 version: 1 type: IntelDLDT from {'INT8': '/home/models/head_yolov4_tiny_608to416_INT8/1/INT8/head_yolov4_tiny_608to416_INT8.xml', 'model-proc': '/home/models/head_yolov4_tiny_608to416_INT8/1/head_yolov4_tiny_608to416_INT8.json'}", "module": "model_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:07:01,060", "message": "Loading Model: head_yolov4_tiny_608to416_default_anchors_mask_012_FP32 version: 1 type: IntelDLDT from {'FP32': '/home/models/head_yolov4_tiny_608to416_default_anchors_mask_012_FP32/1/FP32/head_yolov4_tiny_608to416_default_anchors_mask_012_FP32.xml', 'model-proc': '/home/models/head_yolov4_tiny_608to416_default_anchors_mask_012_FP32/1/head_yolov4_tiny_608to416_default_anchors_mask_012_FP32.json'}", "module": "model_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:07:01,060", "message": "Loading Model: head_yolov4_tiny_608to416_default_anchors_mask_012_FP32copy version: 1 type: IntelDLDT from {'FP32': '/home/models/head_yolov4_tiny_608to416_default_anchors_mask_012_FP32copy/1/FP32/head_yolov4_tiny_608to416_INT8.xml', 'model-proc': '/home/models/head_yolov4_tiny_608to416_default_anchors_mask_012_FP32copy/1/head_yolov4_tiny_608to416_default_anchors_mask_012_FP32.json'}", "module": "model_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:07:01,061", "message": "========================", "module": "model_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:07:01,061", "message": "Completed Loading Models", "module": "model_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:07:01,061", "message": "========================", "module": "model_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:07:01,061", "message": "=================", "module": "pipeline_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:07:01,061", "message": "Loading Pipelines", "module": "pipeline_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:07:01,061", "message": "=================", "module": "pipeline_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:07:01,484", "message": "Loading Pipelines from Config Path /home/pipelines", "module": "pipeline_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:07:01,510", "message": "Loading Pipeline: algorithmName version: 3 type: GStreamer from /home/pipelines/algorithmName/3/pipeline.json", "module": "pipeline_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:07:01,511", "message": "Loading Pipeline: algorithmName version: 2 type: GStreamer from /home/pipelines/algorithmName/2/pipeline.json", "module": "pipeline_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:07:01,515", "message": "Loading Pipeline: algorithmName version: 1 type: GStreamer from /home/pipelines/algorithmName/1/pipeline.json", "module": "pipeline_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:07:01,515", "message": "===========================", "module": "pipeline_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:07:01,515", "message": "Completed Loading Pipelines", "module": "pipeline_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:07:01,516", "message": "===========================", "module": "pipeline_manager"} {"levelname": "INFO", "asctime": "2022-11-07 13:07:01,516", "message": "Creating Instance of Pipeline algorithmName/1", "module": "pipeline_manager"} 0:00:04.149104535 1 0x26dd8a0 ERROR libav :0:: Could not find ref with POC 71 {"levelname": "ERROR", "asctime": "2022-11-07 13:07:05,339", "message": "Error on Pipeline__5 1: gst-library-error-quark: base_inference plugin intitialization failed (3): /opt/build/gst-video-analytics/gst/inference_elements/base/inference_singleton.cpp(137): acquire_inference_instance (): /GstPipeline:pipeline3/GstGvaDetect:detection:\nFailed to load model '/home/models/head_yolov5_tiny_608to416_FP32/1/FP32/dbest.xml'\n\tUnknown model format! Cannot find reader for model format: xml and read the model: /home/models/head_yolov5_tiny_608to416_FP32/1/FP32/dbest.xml. Please check that reader library exists in your PATH.\n", "module": "gstreamer_pipeline"}Which says Unknown model format.
so is OVC does not provide yolo-v5 support? Please help us out as soon as possible,
Thanks
The text was updated successfully, but these errors were encountered: