From ea214ea75a61b803efbe9c9ceefec5c90ebabd68 Mon Sep 17 00:00:00 2001 From: iimironov Date: Thu, 24 Jun 2021 12:48:51 +0300 Subject: [PATCH 01/10] Add retinanet convert doc --- .../tf_specific/Convert_RetinaNet_From_Tensorflow.md | 10 ++++++++++ 1 file changed, 10 insertions(+) create mode 100644 docs/MO_DG/prepare_model/convert_model/tf_specific/Convert_RetinaNet_From_Tensorflow.md diff --git a/docs/MO_DG/prepare_model/convert_model/tf_specific/Convert_RetinaNet_From_Tensorflow.md b/docs/MO_DG/prepare_model/convert_model/tf_specific/Convert_RetinaNet_From_Tensorflow.md new file mode 100644 index 00000000000000..44111ea2baa049 --- /dev/null +++ b/docs/MO_DG/prepare_model/convert_model/tf_specific/Convert_RetinaNet_From_Tensorflow.md @@ -0,0 +1,10 @@ +# Converting RetinaNet Model from TensorFlow* to the Intermediate Representation {#openvino_docs_MO_DG_prepare_model_convert_model_tf_specific_Convert_RetinaNet_From_Tensorflow} + +This tutorial explains how to convert RetinaNet model to Intermediate Representation (IR). + +[Public RetinaNet model](https://github.com/fizyr/keras-retinanet) does not contain pre-trained TensorFlow* weights. To convert this model to the IR you can use [ Reproduce Keras* to TensorFlow* Conversion tutorial](https://docs.openvinotoolkit.org/latest/omz_models_model_retinanet_tf.html) + +After you convert model to TensorFlow* you can run the Model-Optimizer command below: +```sh +python mo.py --input input_1 --input_model retinanet_resnet50_coco_best_v2.1.0.pb --output_dir C:\projects\pycharm\models\IRs --data_type FP32 --input_shape [1,1333,1333,3] --output filtered_detections/map/TensorArrayStack/TensorArrayGatherV3,filtered_detections/map/TensorArrayStack_1/TensorArrayGatherV3,iltered_detections/map/TensorArrayStack_2/TensorArrayGatherV3 --transformations_config ./extensions/front/tf/retinanet.json +``` From 7d8bc799d647879e9c128c68a114fe7b7993368d Mon Sep 17 00:00:00 2001 From: iimironov Date: Tue, 29 Jun 2021 12:05:23 +0300 Subject: [PATCH 02/10] Fix doc --- .../tf_specific/Convert_RetinaNet_From_Tensorflow.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/docs/MO_DG/prepare_model/convert_model/tf_specific/Convert_RetinaNet_From_Tensorflow.md b/docs/MO_DG/prepare_model/convert_model/tf_specific/Convert_RetinaNet_From_Tensorflow.md index 44111ea2baa049..87fda7312d97be 100644 --- a/docs/MO_DG/prepare_model/convert_model/tf_specific/Convert_RetinaNet_From_Tensorflow.md +++ b/docs/MO_DG/prepare_model/convert_model/tf_specific/Convert_RetinaNet_From_Tensorflow.md @@ -2,9 +2,9 @@ This tutorial explains how to convert RetinaNet model to Intermediate Representation (IR). -[Public RetinaNet model](https://github.com/fizyr/keras-retinanet) does not contain pre-trained TensorFlow* weights. To convert this model to the IR you can use [ Reproduce Keras* to TensorFlow* Conversion tutorial](https://docs.openvinotoolkit.org/latest/omz_models_model_retinanet_tf.html) +[Public RetinaNet model](https://github.com/fizyr/keras-retinanet) does not contain pretrained TensorFlow* weights. To convert this model to the IR you can use [Reproduce Keras* to TensorFlow* Conversion tutorial](https://docs.openvinotoolkit.org/latest/omz_models_model_retinanet_tf.html). After you convert model to TensorFlow* you can run the Model-Optimizer command below: ```sh -python mo.py --input input_1 --input_model retinanet_resnet50_coco_best_v2.1.0.pb --output_dir C:\projects\pycharm\models\IRs --data_type FP32 --input_shape [1,1333,1333,3] --output filtered_detections/map/TensorArrayStack/TensorArrayGatherV3,filtered_detections/map/TensorArrayStack_1/TensorArrayGatherV3,iltered_detections/map/TensorArrayStack_2/TensorArrayGatherV3 --transformations_config ./extensions/front/tf/retinanet.json +python mo.py --input "input_1[1 1333 1333 3]" --input_model retinanet_resnet50_coco_best_v2.1.0.pb --data_type FP32 --output filtered_detections/map/TensorArrayStack/TensorArrayGatherV3,filtered_detections/map/TensorArrayStack_1/TensorArrayGatherV3,iltered_detections/map/TensorArrayStack_2/TensorArrayGatherV3 --transformations_config ./extensions/front/tf/retinanet.json ``` From ffc6adbc94d6fbd4d78c912edba2f8f6efa021b6 Mon Sep 17 00:00:00 2001 From: iimironov Date: Wed, 30 Jun 2021 12:47:43 +0300 Subject: [PATCH 03/10] Update doc --- .../tf_specific/Convert_RetinaNet_From_Tensorflow.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/MO_DG/prepare_model/convert_model/tf_specific/Convert_RetinaNet_From_Tensorflow.md b/docs/MO_DG/prepare_model/convert_model/tf_specific/Convert_RetinaNet_From_Tensorflow.md index 87fda7312d97be..56ec9a803e2c04 100644 --- a/docs/MO_DG/prepare_model/convert_model/tf_specific/Convert_RetinaNet_From_Tensorflow.md +++ b/docs/MO_DG/prepare_model/convert_model/tf_specific/Convert_RetinaNet_From_Tensorflow.md @@ -6,5 +6,5 @@ This tutorial explains how to convert RetinaNet model to Intermediate Representa After you convert model to TensorFlow* you can run the Model-Optimizer command below: ```sh -python mo.py --input "input_1[1 1333 1333 3]" --input_model retinanet_resnet50_coco_best_v2.1.0.pb --data_type FP32 --output filtered_detections/map/TensorArrayStack/TensorArrayGatherV3,filtered_detections/map/TensorArrayStack_1/TensorArrayGatherV3,iltered_detections/map/TensorArrayStack_2/TensorArrayGatherV3 --transformations_config ./extensions/front/tf/retinanet.json +python mo.py --input "input_1[1 1333 1333 3]" --input_model retinanet_resnet50_coco_best_v2.1.0.pb --data_type FP32 --transformations_config ./extensions/front/tf/retinanet.json ``` From 747e09fda745d7749d2dcfba7a852d6ed3bbc0ab Mon Sep 17 00:00:00 2001 From: iimironov Date: Thu, 1 Jul 2021 11:37:58 +0300 Subject: [PATCH 04/10] Fix doc mistakes --- .../tf_specific/Convert_RetinaNet_From_Tensorflow.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/docs/MO_DG/prepare_model/convert_model/tf_specific/Convert_RetinaNet_From_Tensorflow.md b/docs/MO_DG/prepare_model/convert_model/tf_specific/Convert_RetinaNet_From_Tensorflow.md index 56ec9a803e2c04..3c3d6488d6ce0d 100644 --- a/docs/MO_DG/prepare_model/convert_model/tf_specific/Convert_RetinaNet_From_Tensorflow.md +++ b/docs/MO_DG/prepare_model/convert_model/tf_specific/Convert_RetinaNet_From_Tensorflow.md @@ -2,9 +2,9 @@ This tutorial explains how to convert RetinaNet model to Intermediate Representation (IR). -[Public RetinaNet model](https://github.com/fizyr/keras-retinanet) does not contain pretrained TensorFlow* weights. To convert this model to the IR you can use [Reproduce Keras* to TensorFlow* Conversion tutorial](https://docs.openvinotoolkit.org/latest/omz_models_model_retinanet_tf.html). +[Public RetinaNet model](https://github.com/fizyr/keras-retinanet) does not contain pretrained TensorFlow* weights. To convert this model to the TensorFlow* format you can use [Reproduce Keras* to TensorFlow* Conversion tutorial](https://docs.openvinotoolkit.org/latest/omz_models_model_retinanet_tf.html). -After you convert model to TensorFlow* you can run the Model-Optimizer command below: +After you convert model to TensorFlow* format you can run the Model-Optimizer command below: ```sh python mo.py --input "input_1[1 1333 1333 3]" --input_model retinanet_resnet50_coco_best_v2.1.0.pb --data_type FP32 --transformations_config ./extensions/front/tf/retinanet.json ``` From 866e33ea6066b7715539a11e471677469609430b Mon Sep 17 00:00:00 2001 From: iimironov Date: Mon, 5 Jul 2021 14:14:48 +0300 Subject: [PATCH 05/10] Update doc --- .../tf_specific/Convert_RetinaNet_From_Tensorflow.md | 2 ++ 1 file changed, 2 insertions(+) diff --git a/docs/MO_DG/prepare_model/convert_model/tf_specific/Convert_RetinaNet_From_Tensorflow.md b/docs/MO_DG/prepare_model/convert_model/tf_specific/Convert_RetinaNet_From_Tensorflow.md index 3c3d6488d6ce0d..94b48427de5c79 100644 --- a/docs/MO_DG/prepare_model/convert_model/tf_specific/Convert_RetinaNet_From_Tensorflow.md +++ b/docs/MO_DG/prepare_model/convert_model/tf_specific/Convert_RetinaNet_From_Tensorflow.md @@ -8,3 +8,5 @@ After you convert model to TensorFlow* format you can run the Model-Optimizer co ```sh python mo.py --input "input_1[1 1333 1333 3]" --input_model retinanet_resnet50_coco_best_v2.1.0.pb --data_type FP32 --transformations_config ./extensions/front/tf/retinanet.json ``` + +Where `transformations_config` command line parameter specifies the configuration json file containing hints to the Model Optimizer on how to convert the model. The json file contains some parameters which must be changed if you train the model yourself. Also it contain how to matching end pints for replacing subgraph nodes. After converting model to IR output nodes will be replaced to DetectionOutput layer. From 790ba9a0ad1eac11af6f916f457f804d5de9fad0 Mon Sep 17 00:00:00 2001 From: iimironov Date: Tue, 6 Jul 2021 14:37:14 +0300 Subject: [PATCH 06/10] Update doc according to review --- .../tf_specific/Convert_RetinaNet_From_Tensorflow.md | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/docs/MO_DG/prepare_model/convert_model/tf_specific/Convert_RetinaNet_From_Tensorflow.md b/docs/MO_DG/prepare_model/convert_model/tf_specific/Convert_RetinaNet_From_Tensorflow.md index 94b48427de5c79..2e35e3378c1841 100644 --- a/docs/MO_DG/prepare_model/convert_model/tf_specific/Convert_RetinaNet_From_Tensorflow.md +++ b/docs/MO_DG/prepare_model/convert_model/tf_specific/Convert_RetinaNet_From_Tensorflow.md @@ -1,12 +1,12 @@ # Converting RetinaNet Model from TensorFlow* to the Intermediate Representation {#openvino_docs_MO_DG_prepare_model_convert_model_tf_specific_Convert_RetinaNet_From_Tensorflow} -This tutorial explains how to convert RetinaNet model to Intermediate Representation (IR). +This tutorial explains how to convert RetinaNet model to the Intermediate Representation (IR). -[Public RetinaNet model](https://github.com/fizyr/keras-retinanet) does not contain pretrained TensorFlow* weights. To convert this model to the TensorFlow* format you can use [Reproduce Keras* to TensorFlow* Conversion tutorial](https://docs.openvinotoolkit.org/latest/omz_models_model_retinanet_tf.html). +[Public RetinaNet model](https://github.com/fizyr/keras-retinanet) does not contain pretrained TensorFlow\* weights. To convert this model to the TensorFlow\* format, you can use [Reproduce Keras* to TensorFlow* Conversion tutorial](https://docs.openvinotoolkit.org/latest/omz_models_model_retinanet_tf.html). -After you convert model to TensorFlow* format you can run the Model-Optimizer command below: +After you convert the model to TensorFlow* format, run the Model Optimizer command below: ```sh python mo.py --input "input_1[1 1333 1333 3]" --input_model retinanet_resnet50_coco_best_v2.1.0.pb --data_type FP32 --transformations_config ./extensions/front/tf/retinanet.json ``` -Where `transformations_config` command line parameter specifies the configuration json file containing hints to the Model Optimizer on how to convert the model. The json file contains some parameters which must be changed if you train the model yourself. Also it contain how to matching end pints for replacing subgraph nodes. After converting model to IR output nodes will be replaced to DetectionOutput layer. +Where `transformations_config` command-line parameter specifies the configuration json file containing model conversion hints for the Model Optimizer. The json file contains some parameters that need to be changed if you train the model yourself. It also contains information on how to match endpoints to replace the subgraph nodes. After the model is converted to IR, the output nodes will be replaced with DetectionOutput layer. From 3825caa93651b575c7d41048598fa464c0c7f65e Mon Sep 17 00:00:00 2001 From: iimironov Date: Thu, 15 Jul 2021 13:39:14 +0300 Subject: [PATCH 07/10] Split some text to several lines --- .../tf_specific/Convert_RetinaNet_From_Tensorflow.md | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/docs/MO_DG/prepare_model/convert_model/tf_specific/Convert_RetinaNet_From_Tensorflow.md b/docs/MO_DG/prepare_model/convert_model/tf_specific/Convert_RetinaNet_From_Tensorflow.md index 2e35e3378c1841..f02d50499fd857 100644 --- a/docs/MO_DG/prepare_model/convert_model/tf_specific/Convert_RetinaNet_From_Tensorflow.md +++ b/docs/MO_DG/prepare_model/convert_model/tf_specific/Convert_RetinaNet_From_Tensorflow.md @@ -2,11 +2,14 @@ This tutorial explains how to convert RetinaNet model to the Intermediate Representation (IR). -[Public RetinaNet model](https://github.com/fizyr/keras-retinanet) does not contain pretrained TensorFlow\* weights. To convert this model to the TensorFlow\* format, you can use [Reproduce Keras* to TensorFlow* Conversion tutorial](https://docs.openvinotoolkit.org/latest/omz_models_model_retinanet_tf.html). +[Public RetinaNet model](https://github.com/fizyr/keras-retinanet) does not contain pretrained TensorFlow\* weights. +To convert this model to the TensorFlow\* format, you can use [Reproduce Keras* to TensorFlow* Conversion tutorial](https://docs.openvinotoolkit.org/latest/omz_models_model_retinanet_tf.html). After you convert the model to TensorFlow* format, run the Model Optimizer command below: ```sh python mo.py --input "input_1[1 1333 1333 3]" --input_model retinanet_resnet50_coco_best_v2.1.0.pb --data_type FP32 --transformations_config ./extensions/front/tf/retinanet.json ``` -Where `transformations_config` command-line parameter specifies the configuration json file containing model conversion hints for the Model Optimizer. The json file contains some parameters that need to be changed if you train the model yourself. It also contains information on how to match endpoints to replace the subgraph nodes. After the model is converted to IR, the output nodes will be replaced with DetectionOutput layer. +Where `transformations_config` command-line parameter specifies the configuration json file containing model conversion hints for the Model Optimizer. +The json file contains some parameters that need to be changed if you train the model yourself. It also contains information on how to match endpoints +to replace the subgraph nodes. After the model is converted to IR, the output nodes will be replaced with DetectionOutput layer. From 9a6eb070a747079242ea114c67c6bec620081963 Mon Sep 17 00:00:00 2001 From: iimironov Date: Thu, 15 Jul 2021 18:53:59 +0300 Subject: [PATCH 08/10] Update ie_docs --- docs/doxygen/ie_docs.xml | 1 + 1 file changed, 1 insertion(+) diff --git a/docs/doxygen/ie_docs.xml b/docs/doxygen/ie_docs.xml index 19a87a1e11e97c..30811bbc80b2d7 100644 --- a/docs/doxygen/ie_docs.xml +++ b/docs/doxygen/ie_docs.xml @@ -34,6 +34,7 @@ limitations under the License. + From 7259bd395b61e3f28de45a0ee466868eeb2ec4e5 Mon Sep 17 00:00:00 2001 From: iimironov Date: Fri, 16 Jul 2021 10:51:35 +0300 Subject: [PATCH 09/10] Update title for ie_docs --- docs/doxygen/ie_docs.xml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/doxygen/ie_docs.xml b/docs/doxygen/ie_docs.xml index 30811bbc80b2d7..d170261cc0f991 100644 --- a/docs/doxygen/ie_docs.xml +++ b/docs/doxygen/ie_docs.xml @@ -34,7 +34,7 @@ limitations under the License. - + From babdf8d2061010f452c7d2caa44d6ef3eb42cfe5 Mon Sep 17 00:00:00 2001 From: iimironov Date: Tue, 20 Jul 2021 16:01:31 +0300 Subject: [PATCH 10/10] Change tabs to space --- docs/doxygen/ie_docs.xml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/doxygen/ie_docs.xml b/docs/doxygen/ie_docs.xml index d170261cc0f991..7ed795ea5c3720 100644 --- a/docs/doxygen/ie_docs.xml +++ b/docs/doxygen/ie_docs.xml @@ -34,7 +34,7 @@ limitations under the License. - +