Skip to content

Commit 2f67686

Browse files
yunchuGalyaZalesskayajaegukhyun
authored
Mergeback 1.5.0 to develop (#2642)
* Update publish workflow for tag checking (#2632) * Update e2e tests for XAI Detection (#2634) * Disable QAT for newly added models (#2636) * Update release note and readme (#2637) * remove package upload step on internal publish wf * update release note and, changelog, and readme * update version string to 1.6.0dev --------- Co-authored-by: Galina Zalesskaya <galina.zalesskaya@intel.com> Co-authored-by: Jaeguk Hyun <jaeguk.hyun@intel.com>
1 parent 104404b commit 2f67686

File tree

13 files changed

+64
-40
lines changed

13 files changed

+64
-40
lines changed

‎.github/workflows/publish_internal.yml

+7-12
Original file line numberDiff line numberDiff line change
@@ -2,8 +2,6 @@ name: Build and upload to internal PyPI
22

33
on:
44
workflow_dispatch: # run on request (no need for PR)
5-
release:
6-
types: [published]
75

86
jobs:
97
build_wheels:
@@ -61,23 +59,20 @@ jobs:
6159
uses: actions-ecosystem/action-regex-match@v2
6260
with:
6361
text: ${{ github.ref }}
64-
regex: '^refs/tags/[0-9]+\.[0-9]+\.[0-9]+(\.[0-9]+rc[0-9]+|rc[0-9]+)?$'
65-
- name: Upload package distributions to github
66-
if: ${{ steps.check-tag.outputs.match != '' }}
67-
uses: svenstaro/upload-release-action@v2
68-
with:
69-
repo_token: ${{ secrets.GITHUB_TOKEN }}
70-
file: dist/*
71-
tag: ${{ github.ref }}
72-
overwrite: true
73-
file_glob: true
62+
regex: '^refs/heads/releases/[0-9]+\.[0-9]+\.[0-9]+(\.[0-9]+rc[0-9]+|rc[0-9]+)?$'
7463
- name: Check dist contents
7564
run: twine check dist/*
7665
- name: Publish package dist to internal PyPI
66+
if: ${{ steps.check-tag.outputs.match != '' }}
7767
run: |
7868
export no_proxy=${{ secrets.PYPI_HOST }}
7969
export REPOSITORY_URL=http://${{ secrets.PYPI_HOST }}:${{ secrets.PYPI_PORT }}
8070
twine upload --verbose --repository-url $REPOSITORY_URL dist/* -u ${{ secrets.PYPI_USER }} -p ${{ secrets.PYPI_PASSWORD }}
71+
- name: Publish package distributions to TestPyPI
72+
if: ${{ steps.check-tag.outputs.match == '' }}
73+
run: |
74+
export REPOSITORY_URL=https://test.pypi.org/legacy/
75+
twine upload --verbose --repository-url $REPOSITORY_URL dist/* -u __token__ -p ${{ secrets.TESTPYPI_API_TOKEN }}
8176
- name: Clean up dist
8277
if: ${{ always() }}
8378
run: |

‎CHANGELOG.md

+8-6
Original file line numberDiff line numberDiff line change
@@ -2,18 +2,20 @@
22

33
All notable changes to this project will be documented in this file.
44

5-
## \[v1.5.0 - unreleased\]
5+
## \[unreleased\]
6+
7+
## \[v1.5.0\]
68

79
### New features
810

9-
- Enable configurable confidence threshold for otx eval and export(<https://github.com/openvinotoolkit/training_extensions/pull/2388>)
11+
- Enable configurable confidence threshold for otx eval and export (<https://github.com/openvinotoolkit/training_extensions/pull/2388>)
1012
- Add YOLOX variants as new object detector models (<https://github.com/openvinotoolkit/training_extensions/pull/2402>)
11-
- Enable FeatureVectorHook to support action tasks(<https://github.com/openvinotoolkit/training_extensions/pull/2408>)
13+
- Enable FeatureVectorHook to support action tasks (<https://github.com/openvinotoolkit/training_extensions/pull/2408>)
1214
- Add ONNX metadata to detection, instance segmantation, and segmentation models (<https://github.com/openvinotoolkit/training_extensions/pull/2418>)
13-
- Add a new feature to configure input size(<https://github.com/openvinotoolkit/training_extensions/pull/2420>)
15+
- Add a new feature to configure input size (<https://github.com/openvinotoolkit/training_extensions/pull/2420>)
1416
- Introduce the OTXSampler and AdaptiveRepeatDataHook to achieve faster training at the small data regime (<https://github.com/openvinotoolkit/training_extensions/pull/2428>)
15-
- Add a new object detector Lite-DINO(<https://github.com/openvinotoolkit/training_extensions/pull/2457>)
16-
- Add Semi-SL Mean Teacher algorithm for Instance Segmentation task(<https://github.com/openvinotoolkit/training_extensions/pull/2444>)
17+
- Add a new object detector Lite-DINO (<https://github.com/openvinotoolkit/training_extensions/pull/2457>)
18+
- Add Semi-SL Mean Teacher algorithm for Instance Segmentation task (<https://github.com/openvinotoolkit/training_extensions/pull/2444>)
1719
- Official supports for YOLOX-X, YOLOX-L, YOLOX-S, ResNeXt101-ATSS (<https://github.com/openvinotoolkit/training_extensions/pull/2485>)
1820
- Add new argument to track resource usage in train command (<https://github.com/openvinotoolkit/training_extensions/pull/2500>)
1921
- Add Self-SL for semantic segmentation of SegNext families (<https://github.com/openvinotoolkit/training_extensions/pull/2215>)

‎README.md

+17-13
Original file line numberDiff line numberDiff line change
@@ -42,7 +42,7 @@ If you are an experienced user, you can configure your own model based on [torch
4242

4343
Furthermore, OpenVINO™ Training Extensions provides automatic configuration for ease of use.
4444
The framework will analyze your dataset and identify the most suitable model and figure out the best input size setting and other hyper-parameters.
45-
The development team is continuously extending this [Auto-configuration](https://openvinotoolkit.github.io/training_extensions/latest/guide/explanation/additional_features/auto_configuration.html) functionalities to make training as simple as possible so that single CLI command can obtain accurate, efficient and robust models ready to be integrated into your project.
45+
The development team is continuously extending this [Auto-configuration](https://openvinotoolkit.github.io/training_extensions/stable/guide/explanation/additional_features/auto_configuration.html) functionalities to make training as simple as possible so that single CLI command can obtain accurate, efficient and robust models ready to be integrated into your project.
4646

4747
### Key Features
4848

@@ -63,11 +63,11 @@ OpenVINO™ Training Extensions supports the [following learning methods](https:
6363

6464
OpenVINO™ Training Extensions provides the following usability features:
6565

66-
- [Auto-configuration](https://openvinotoolkit.github.io/training_extensions/latest/guide/explanation/additional_features/auto_configuration.html). OpenVINO™ Training Extensions analyzes provided dataset and selects the proper task and model with appropriate input size to provide the best accuracy/speed trade-off. It will also make a random auto-split of your dataset if there is no validation set provided.
66+
- [Auto-configuration](https://openvinotoolkit.github.io/training_extensions/stable/guide/explanation/additional_features/auto_configuration.html). OpenVINO™ Training Extensions analyzes provided dataset and selects the proper task and model with appropriate input size to provide the best accuracy/speed trade-off. It will also make a random auto-split of your dataset if there is no validation set provided.
6767
- [Datumaro](https://openvinotoolkit.github.io/datumaro/stable/index.html) data frontend: OpenVINO™ Training Extensions supports the most common academic field dataset formats for each task. We are constantly working to extend supported formats to give more freedom of datasets format choice.
6868
- **Distributed training** to accelerate the training process when you have multiple GPUs
6969
- **Mixed-precision training** to save GPUs memory and use larger batch sizes
70-
- Integrated, efficient [hyper-parameter optimization module (HPO)](https://openvinotoolkit.github.io/training_extensions/latest/guide/explanation/additional_features/hpo.html). Through dataset proxy and built-in hyper-parameter optimizer, you can get much faster hyper-parameter optimization compared to other off-the-shelf tools. The hyperparameter optimization is dynamically scheduled based on your resource budget.
70+
- Integrated, efficient [hyper-parameter optimization module (HPO)](https://openvinotoolkit.github.io/training_extensions/stable/guide/explanation/additional_features/hpo.html). Through dataset proxy and built-in hyper-parameter optimizer, you can get much faster hyper-parameter optimization compared to other off-the-shelf tools. The hyperparameter optimization is dynamically scheduled based on your resource budget.
7171

7272
---
7373

@@ -97,16 +97,20 @@ You can find more details with examples in the [CLI command intro](https://openv
9797

9898
## Updates
9999

100-
### v1.4.0 (3Q23)
101-
102-
- Support encrypted dataset training (<https://github.com/openvinotoolkit/training_extensions/pull/2209>)
103-
- Add custom max iou assigner to prevent CPU OOM when large annotations are used (<https://github.com/openvinotoolkit/training_extensions/pull/2228>)
104-
- Auto train type detection for Semi-SL, Self-SL and Incremental: "--train-type" now is optional (<https://github.com/openvinotoolkit/training_extensions/pull/2195>)
105-
- Add per-class XAI saliency maps for Mask R-CNN model (<https://github.com/openvinotoolkit/training_extensions/pull/2227>)
106-
- Add new object detector Deformable DETR (<https://github.com/openvinotoolkit/training_extensions/pull/2249>)
107-
- Add new object detector DINO (<https://github.com/openvinotoolkit/training_extensions/pull/2266>)
108-
- Add new visual prompting task (<https://github.com/openvinotoolkit/training_extensions/pull/2203>, <https://github.com/openvinotoolkit/training_extensions/pull/2274>, <https://github.com/openvinotoolkit/training_extensions/pull/2311>, <https://github.com/openvinotoolkit/training_extensions/pull/2354>, <https://github.com/openvinotoolkit/training_extensions/pull/2318>)
109-
- Add new object detector ResNeXt101-ATSS (<https://github.com/openvinotoolkit/training_extensions/pull/2309>)
100+
### v1.5.0 (4Q23)
101+
102+
- Enable configurable confidence threshold for otx eval and export (<https://github.com/openvinotoolkit/training_extensions/pull/2388>)
103+
- Add YOLOX variants as new object detector models (<https://github.com/openvinotoolkit/training_extensions/pull/2402>)
104+
- Enable FeatureVectorHook to support action tasks (<https://github.com/openvinotoolkit/training_extensions/pull/2408>)
105+
- Add ONNX metadata to detection, instance segmantation, and segmentation models (<https://github.com/openvinotoolkit/training_extensions/pull/2418>)
106+
- Add a new feature to configure input size (<https://github.com/openvinotoolkit/training_extensions/pull/2420>)
107+
- Introduce the OTXSampler and AdaptiveRepeatDataHook to achieve faster training at the small data regime (<https://github.com/openvinotoolkit/training_extensions/pull/2428>)
108+
- Add a new object detector Lite-DINO (<https://github.com/openvinotoolkit/training_extensions/pull/2457>)
109+
- Add Semi-SL Mean Teacher algorithm for Instance Segmentation task (<https://github.com/openvinotoolkit/training_extensions/pull/2444>)
110+
- Official supports for YOLOX-X, YOLOX-L, YOLOX-S, ResNeXt101-ATSS (<https://github.com/openvinotoolkit/training_extensions/pull/2485>)
111+
- Add new argument to track resource usage in train command (<https://github.com/openvinotoolkit/training_extensions/pull/2500>)
112+
- Add Self-SL for semantic segmentation of SegNext families (<https://github.com/openvinotoolkit/training_extensions/pull/2215>)
113+
- Adapt input size automatically based on dataset statistics (<https://github.com/openvinotoolkit/training_extensions/pull/2499>)
110114

111115
### Release History
112116

‎docs/source/guide/release_notes/index.rst

+29
Original file line numberDiff line numberDiff line change
@@ -4,6 +4,35 @@ Releases
44
.. toctree::
55
:maxdepth: 1
66

7+
v1.5.0 (4Q23)
8+
-------------
9+
10+
- Enable configurable confidence threshold for otx eval and export
11+
- Add YOLOX variants as new object detector models
12+
- Enable FeatureVectorHook to support action tasks
13+
- Add ONNX metadata to detection, instance segmantation, and segmentation models
14+
- Add a new feature to configure input size
15+
- Introduce the OTXSampler and AdaptiveRepeatDataHook to achieve faster training at the small data regime
16+
- Add a new object detector Lite-DINO
17+
- Add Semi-SL Mean Teacher algorithm for Instance Segmentation task
18+
- Official supports for YOLOX-X, YOLOX-L, YOLOX-S, ResNeXt101-ATSS
19+
- Add new argument to track resource usage in train command
20+
- Add Self-SL for semantic segmentation of SegNext families
21+
- Adapt input size automatically based on dataset statistics
22+
- Refine input data in-memory caching
23+
- Adapt timeout value of initialization for distributed training
24+
- Optimize data loading by merging load & resize operations w/ caching support for cls/det/iseg/sseg
25+
- Support torch==2.0.1
26+
- Set "Auto" as default input size mode
27+
28+
29+
v1.4.4 (4Q23)
30+
-------------
31+
32+
- Update ModelAPI configuration
33+
- Add Anomaly modelAPI changes
34+
- Update Image numpy access
35+
736
v1.4.3 (4Q23)
837
-------------
938

‎requirements/base.txt

+1-1
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,5 @@
11
# # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # #
2-
# Base Algo Requirements. #
2+
# Base Algo Requirements. #
33
natsort==8.1.*
44
prettytable==3.9.*
55
protobuf==3.20.*

‎src/otx/__init__.py

+1-1
Original file line numberDiff line numberDiff line change
@@ -3,5 +3,5 @@
33
# Copyright (C) 2021-2023 Intel Corporation
44
# SPDX-License-Identifier: Apache-2.0
55

6-
__version__ = "1.5.0"
6+
__version__ = "1.6.0dev"
77
# NOTE: Sync w/ src/otx/api/usecases/exportable_code/demo/requirements.txt on release

‎src/otx/algorithms/detection/configs/detection/cspdarknet_yolox_l/template.yaml

-1
Original file line numberDiff line numberDiff line change
@@ -14,7 +14,6 @@ framework: OTXDetection v2.9.1
1414
entrypoints:
1515
base: otx.algorithms.detection.adapters.mmdet.task.MMDetectionTask
1616
openvino: otx.algorithms.detection.adapters.openvino.task.OpenVINODetectionTask
17-
nncf: otx.algorithms.detection.adapters.mmdet.nncf.task.DetectionNNCFTask
1817

1918
# Capabilities.
2019
capabilities:

‎src/otx/algorithms/detection/configs/detection/cspdarknet_yolox_s/template.yaml

-1
Original file line numberDiff line numberDiff line change
@@ -14,7 +14,6 @@ framework: OTXDetection v2.9.1
1414
entrypoints:
1515
base: otx.algorithms.detection.adapters.mmdet.task.MMDetectionTask
1616
openvino: otx.algorithms.detection.adapters.openvino.task.OpenVINODetectionTask
17-
nncf: otx.algorithms.detection.adapters.mmdet.nncf.task.DetectionNNCFTask
1817

1918
# Capabilities.
2019
capabilities:

‎src/otx/algorithms/detection/configs/detection/cspdarknet_yolox_x/template.yaml

-1
Original file line numberDiff line numberDiff line change
@@ -14,7 +14,6 @@ framework: OTXDetection v2.9.1
1414
entrypoints:
1515
base: otx.algorithms.detection.adapters.mmdet.task.MMDetectionTask
1616
openvino: otx.algorithms.detection.adapters.openvino.task.OpenVINODetectionTask
17-
nncf: otx.algorithms.detection.adapters.mmdet.nncf.task.DetectionNNCFTask
1817

1918
# Capabilities.
2019
capabilities:

‎src/otx/algorithms/detection/configs/detection/resnext101_atss/template.yaml

-1
Original file line numberDiff line numberDiff line change
@@ -14,7 +14,6 @@ framework: OTXDetection v2.9.1
1414
entrypoints:
1515
base: otx.algorithms.detection.adapters.mmdet.task.MMDetectionTask
1616
openvino: otx.algorithms.detection.adapters.openvino.task.OpenVINODetectionTask
17-
nncf: otx.algorithms.detection.adapters.mmdet.nncf.task.DetectionNNCFTask
1817

1918
# Capabilities.
2019
capabilities:

‎src/otx/algorithms/detection/configs/instance_segmentation/convnext_maskrcnn/template_experimental.yaml

-1
Original file line numberDiff line numberDiff line change
@@ -14,7 +14,6 @@ framework: OTXDetection v2.9.1
1414
entrypoints:
1515
base: otx.algorithms.detection.adapters.mmdet.task.MMDetectionTask
1616
openvino: otx.algorithms.detection.adapters.openvino.task.OpenVINODetectionTask
17-
nncf: otx.algorithms.detection.adapters.mmdet.nncf.task.DetectionNNCFTask
1817

1918
# Capabilities.
2019
capabilities:

‎src/otx/algorithms/detection/configs/instance_segmentation/maskrcnn_swin_t/template.yaml

-1
Original file line numberDiff line numberDiff line change
@@ -14,7 +14,6 @@ framework: OTXDetection v2.9.1
1414
entrypoints:
1515
base: otx.algorithms.detection.adapters.mmdet.task.MMDetectionTask
1616
openvino: otx.algorithms.detection.adapters.openvino.task.OpenVINODetectionTask
17-
nncf: otx.algorithms.detection.adapters.mmdet.nncf.task.DetectionNNCFTask
1817

1918
# Capabilities.
2019
capabilities:

‎tests/e2e/cli/detection/test_api_xai_sanity_detection.py

+1-1
Original file line numberDiff line numberDiff line change
@@ -32,7 +32,7 @@
3232

3333
class TestOVDetXAIAPI(DetectionTaskAPIBase):
3434
ref_raw_saliency_shapes = {
35-
"MobileNetV2-ATSS": (4, 4), # Need to be adapted to configurable or adaptive input size
35+
"MobileNetV2-ATSS": (16, 16), # Need to be adapted to configurable or adaptive input size
3636
}
3737

3838
@e2e_pytest_api

0 commit comments

Comments
 (0)