Merge branch 'develop' into gz/v2.1_xai_tiling

openvinotoolkit · Apr 11, 2024 · a0d35cc · a0d35cc
2 parents 4d553bc + e8e5673
commit a0d35cc
Show file tree

Hide file tree

Showing 28 changed files with 3,326 additions and 136 deletions.
diff --git a/.github/workflows/code_scan.yaml b/.github/workflows/code_scan.yaml
@@ -28,7 +28,7 @@ jobs:
       - name: Freeze dependencies
         run: pip-compile --extra=docs,base,mmlab,anomaly -o requirements.txt pyproject.toml
       - name: Trivy Scanning
-        uses: aquasecurity/trivy-action@062f2592684a31eb3aa050cc61e7ca1451cecd3d # 0.18.0
+        uses: aquasecurity/trivy-action@d710430a6722f083d3b36b8339ff66b32f22ee55 # 0.19.0
         with:
           trivy-config: ".ci/trivy.yaml"
           scan-type: "fs"

diff --git a/MANIFEST.in b/MANIFEST.in
@@ -4,6 +4,7 @@ recursive-include src/otx *.yaml
 recursive-include src/otx *.json
 recursive-include src/otx requirements.txt
 recursive-include src/otx README.md
+recursive-include src/otx LICENSE
 recursive-exclude src/otx *.c
 graft tests
 global-exclude *.py[cod]
diff --git a/docs/source/guide/tutorials/base/export.rst b/docs/source/guide/tutorials/base/export.rst
@@ -66,7 +66,20 @@ using the command below:
                --export_format EXPORTABLE_CODE
                --work-dir outputs/deploy
 
-After that, you can use the resulting ``openvino.zip`` archive in other applications.
+After that, you can use the resulting ``exportable_code.zip`` archive in other applications.
+
+4. It is also possible to pass already exported/optimized OpenVINO IR model
+to create archive with demo and pack in the IR model.
+
+.. code-block:: shell
+
+    (otx) ...$ otx export -c CONFIG
+               --checkpoint {OPENVINO_IR.XML}
+               --work-dir outputs/deploy
+
+.. note::
+
+    You can also obtain ``exportable_code.zip`` right away during model optimization passing ``export_demo_package=True`` parameter to CLI or API call.
 
 *************
 Demonstration
@@ -75,7 +88,7 @@ Demonstration
 Using the exported demo, we're able to run the model in the demonstration mode outside of this repository, using only the ported ``.zip`` archive with minimum required packages.
 The demo allows us to apply our model on the custom data or the online footage from a web camera and see how it will work in a real-life scenario. It is not required to install OTX or PyTorch.
 
-1. Unzip the ``openvino.zip``
+1. Unzip the ``exportable_code.zip``
 archive.
 
 .. code-block::

diff --git a/docs/source/guide/tutorials/base/how_to_train/classification.rst b/docs/source/guide/tutorials/base/how_to_train/classification.rst
@@ -85,10 +85,10 @@ please keep the exact same name for the train/val/test folder, to identify the d
       ├── tulips
     val
       ├── daisy
-      ├── ... 
+      ├── ...
     test
       ├── daisy
-      ├── ... 
+      ├── ...
 
 *********
 Training
@@ -108,26 +108,26 @@ The list of supported recipes for classification is available with the command l
     .. code-block:: shell
 
       (otx) ...$ otx find --task MULTI_CLASS_CLS
-      ┏━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓                                  
-      ┃ Task            ┃ Model Name               ┃ Recipe Path                                                                    ┃                                  
-      ┡━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩                                  
-      │ MULTI_CLASS_CLS │ openvino_model           │ src/otx/recipe/classification/multi_class_cls/openvino_model.yaml              │                                  
-      │ MULTI_CLASS_CLS │ tv_efficientnet_b0       │ src/otx/recipe/classification/multi_class_cls/tv_efficientnet_b0.yaml          │                                  
-      │ MULTI_CLASS_CLS │ tv_resnet_50             │ src/otx/recipe/classification/multi_class_cls/tv_resnet_50.yaml                │                                  
-      │ MULTI_CLASS_CLS │ efficientnet_v2_light    │ src/otx/recipe/classification/multi_class_cls/efficientnet_v2_light.yaml       │                                  
-      │ MULTI_CLASS_CLS │ tv_efficientnet_b3       │ src/otx/recipe/classification/multi_class_cls/tv_efficientnet_b3.yaml          │                                  
-      │ MULTI_CLASS_CLS │ efficientnet_b0_light    │ src/otx/recipe/classification/multi_class_cls/efficientnet_b0_light.yaml       │                                  
-      │ MULTI_CLASS_CLS │ tv_efficientnet_v2_l     │ src/otx/recipe/classification/multi_class_cls/tv_efficientnet_v2_l.yaml        │                                  
-      │ MULTI_CLASS_CLS │ tv_efficientnet_b1       │ src/otx/recipe/classification/multi_class_cls/tv_efficientnet_b1.yaml          │                                  
-      │ MULTI_CLASS_CLS │ tv_mobilenet_v3_small    │ src/otx/recipe/classification/multi_class_cls/tv_mobilenet_v3_small.yaml       │                                  
-      │ MULTI_CLASS_CLS │ otx_mobilenet_v3_large   │ src/otx/recipe/classification/multi_class_cls/otx_mobilenet_v3_large.yaml      │                                  
-      │ MULTI_CLASS_CLS │ otx_deit_tiny            │ src/otx/recipe/classification/multi_class_cls/otx_deit_tiny.yaml               │                                  
-      │ MULTI_CLASS_CLS │ tv_efficientnet_b4       │ src/otx/recipe/classification/multi_class_cls/tv_efficientnet_b4.yaml          │                                  
-      │ MULTI_CLASS_CLS │ otx_efficientnet_v2      │ src/otx/recipe/classification/multi_class_cls/otx_efficientnet_v2.yaml         │                                  
-      │ MULTI_CLASS_CLS │ mobilenet_v3_large_light │ src/otx/recipe/classification/multi_class_cls/mobilenet_v3_large_light.yaml    │                                  
-      │ MULTI_CLASS_CLS │ otx_efficientnet_b0      │ src/otx/recipe/classification/multi_class_cls/otx_efficientnet_b0.yaml         │                                  
-      │ MULTI_CLASS_CLS │ otx_dino_v2              │ src/otx/recipe/classification/multi_class_cls/otx_dino_v2.yaml                 │                                  
-      │ MULTI_CLASS_CLS │ otx_dino_v2_linear_probe │ src/otx/recipe/classification/multi_class_cls/otx_dino_v2_linear_probe.yaml    │                                  
+      ┏━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓
+      ┃ Task            ┃ Model Name               ┃ Recipe Path                                                                    ┃
+      ┡━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩
+      │ MULTI_CLASS_CLS │ openvino_model           │ src/otx/recipe/classification/multi_class_cls/openvino_model.yaml              │
+      │ MULTI_CLASS_CLS │ tv_efficientnet_b0       │ src/otx/recipe/classification/multi_class_cls/tv_efficientnet_b0.yaml          │
+      │ MULTI_CLASS_CLS │ tv_resnet_50             │ src/otx/recipe/classification/multi_class_cls/tv_resnet_50.yaml                │
+      │ MULTI_CLASS_CLS │ efficientnet_v2_light    │ src/otx/recipe/classification/multi_class_cls/efficientnet_v2_light.yaml       │
+      │ MULTI_CLASS_CLS │ tv_efficientnet_b3       │ src/otx/recipe/classification/multi_class_cls/tv_efficientnet_b3.yaml          │
+      │ MULTI_CLASS_CLS │ efficientnet_b0_light    │ src/otx/recipe/classification/multi_class_cls/efficientnet_b0_light.yaml       │
+      │ MULTI_CLASS_CLS │ tv_efficientnet_v2_l     │ src/otx/recipe/classification/multi_class_cls/tv_efficientnet_v2_l.yaml        │
+      │ MULTI_CLASS_CLS │ tv_efficientnet_b1       │ src/otx/recipe/classification/multi_class_cls/tv_efficientnet_b1.yaml          │
+      │ MULTI_CLASS_CLS │ tv_mobilenet_v3_small    │ src/otx/recipe/classification/multi_class_cls/tv_mobilenet_v3_small.yaml       │
+      │ MULTI_CLASS_CLS │ otx_mobilenet_v3_large   │ src/otx/recipe/classification/multi_class_cls/otx_mobilenet_v3_large.yaml      │
+      │ MULTI_CLASS_CLS │ otx_deit_tiny            │ src/otx/recipe/classification/multi_class_cls/otx_deit_tiny.yaml               │
+      │ MULTI_CLASS_CLS │ tv_efficientnet_b4       │ src/otx/recipe/classification/multi_class_cls/tv_efficientnet_b4.yaml          │
+      │ MULTI_CLASS_CLS │ otx_efficientnet_v2      │ src/otx/recipe/classification/multi_class_cls/otx_efficientnet_v2.yaml         │
+      │ MULTI_CLASS_CLS │ mobilenet_v3_large_light │ src/otx/recipe/classification/multi_class_cls/mobilenet_v3_large_light.yaml    │
+      │ MULTI_CLASS_CLS │ otx_efficientnet_b0      │ src/otx/recipe/classification/multi_class_cls/otx_efficientnet_b0.yaml         │
+      │ MULTI_CLASS_CLS │ otx_dino_v2              │ src/otx/recipe/classification/multi_class_cls/otx_dino_v2.yaml                 │
+      │ MULTI_CLASS_CLS │ otx_dino_v2_linear_probe │ src/otx/recipe/classification/multi_class_cls/otx_dino_v2_linear_probe.yaml    │
       └─────────────────┴──────────────────────────┴────────────────────────────────────────────────────────────────────────────────┘
 
   .. tab-item:: API
@@ -136,7 +136,7 @@ The list of supported recipes for classification is available with the command l
 
       from otx.engine.utils.api import list_models
 
-      model_lists = list_models(task="MULTI_CLASS_CLS", pattern="*efficient") 
+      model_lists = list_models(task="MULTI_CLASS_CLS", pattern="*efficient")
       print(model_lists)
       '''
       [
@@ -486,7 +486,7 @@ with OpenVINO™ PTQ.
 
         .. code-block:: shell
 
-            (otx) ...$ otx optimize  --work_dir otx-workspace \ 
+            (otx) ...$ otx optimize  --work_dir otx-workspace \
                                      --checkpoint otx-workspace/20240312_052847/exported_model.xml
 
             ...
@@ -505,6 +505,10 @@ with OpenVINO™ PTQ.
 The optimization time highly relies on the hardware characteristics, for example on Intel(R) Core(TM) i9-10980XE it took about 9 seconds.
 Please note, that PTQ will take some time without logging to optimize the model.
 
+.. note::
+
+    You can also pass ``export_demo_package=True`` parameter to obtain ``exportable_code.zip`` archive with packed optimized model and demo package. Please refer to :doc:`export tutorial <../export>`.
+
 3. Finally, we can also evaluate the optimized model by passing
 it to the ``otx test`` function.
 
@@ -514,7 +518,7 @@ it to the ``otx test`` function.
 
         .. code-block:: shell
 
-            (otx) ...$ otx test --work_dir otx-workspace \ 
+            (otx) ...$ otx test --work_dir otx-workspace \
                                 --checkpoint otx-workspace/20240312_055042/optimized_model.xml \
                                 --engine.device cpu
 

diff --git a/docs/source/guide/tutorials/base/how_to_train/detection.rst b/docs/source/guide/tutorials/base/how_to_train/detection.rst
@@ -129,12 +129,12 @@ The list of supported recipes for object detection is available with the command
         .. code-block:: shell
 
             (otx) ...$ otx find --task DETECTION --pattern atss
-            ┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓                              
-            ┃ Task      ┃ Model Name            ┃ Recipe Path                                                    ┃                              
-            ┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩                              
-            │ DETECTION │ atss_mobilenetv2_tile │ src/otx/recipe/detection/atss_mobilenetv2_tile.yaml            │                              
-            │ DETECTION │ atss_resnext101       │ src/otx/recipe/detection/atss_resnext101.yaml                  │                              
-            │ DETECTION │ atss_mobilenetv2      │ src/otx/recipe/detection/atss_mobilenetv2.yaml                 │                              
+            ┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┓
+            ┃ Task      ┃ Model Name            ┃ Recipe Path                                                    ┃
+            ┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┩
+            │ DETECTION │ atss_mobilenetv2_tile │ src/otx/recipe/detection/atss_mobilenetv2_tile.yaml            │
+            │ DETECTION │ atss_resnext101       │ src/otx/recipe/detection/atss_resnext101.yaml                  │
+            │ DETECTION │ atss_mobilenetv2      │ src/otx/recipe/detection/atss_mobilenetv2.yaml                 │
             └───────────┴───────────────────────┴────────────────────────────────────────────────────────────────┘
 
     .. tab-item:: API
@@ -520,7 +520,7 @@ with OpenVINO™ PTQ.
 
         .. code-block:: shell
 
-            (otx) ...$ otx optimize  --work_dir otx-workspace \ 
+            (otx) ...$ otx optimize  --work_dir otx-workspace \
                                      --checkpoint otx-workspace/20240312_052847/exported_model.xml
 
             ...
@@ -539,6 +539,10 @@ with OpenVINO™ PTQ.
 The optimization time highly relies on the hardware characteristics, for example on Intel(R) Core(TM) i9-11900 it took about 25 seconds.
 Please note, that PTQ will take some time without logging to optimize the model.
 
+.. note::
+
+    You can also pass ``export_demo_package=True`` parameter to obtain ``exportable_code.zip`` archive with packed optimized model and demo package. Please refer to :doc:`export tutorial <../export>`.
+
 3. Finally, we can also evaluate the optimized model by passing
 it to the ``otx test`` function.
 
@@ -548,7 +552,7 @@ it to the ``otx test`` function.
 
         .. code-block:: shell
 
-            (otx) ...$ otx test --work_dir otx-workspace \ 
+            (otx) ...$ otx test --work_dir otx-workspace \
                                 --checkpoint otx-workspace/20240312_055042/optimized_model.xml \
                                 --engine.device cpu