training_extensions

training_extensions copied to clipboard

Post processing of image classification model

2

Hi, I have trained _MobileNet-V3-large-1x_ classifier for 2 classes using openvino training extensions. While inferring the trained onnx model, I get the result as something similar to this vector [[-1.5802931e-10...

raushani2v

RuntimeError: result type ComplexFloat can't be cast to the desired output type Float

3

**Describe the bug** tested on branch `releases/1.5.0` failures on regression test for semantic segmentation * train type: class incr * template * SegNext-B, SegNext-t, SegNext-s  **Steps to Reproduce** 1....

yunchu

Typo on model_template_id of EfficientNet-B0

3

**Describe the bug** The model_template_id of EfficientNet-B0 is wrong https://github.com/openvinotoolkit/training_extensions/blob/ead81401bb9024b6370716965885228870871230/src/otx/algorithms/classification/configs/efficientnet_b0_cls_incr/template.yaml#L2C19-L2C63 The name should be `Custom_Image_Classification_EfficientNet-B0`. **Steps to Reproduce** 1. 3. 4. 5. **Environment:** - OS: - Framework version: - Python...

wonjuleee

Failed to import dataset 'imagenet'

9

I'm running the [tutorial](https://openvinotoolkit.github.io/training_extensions/latest/guide/tutorials/base/how_to_train/detection.html): ```otx build MobileNetV2-ATSS --train-data-roots data/wgisd``` But I get the error: ``` raise DatasetImportError(f"Failed to import dataset '{format}' at '{path}'.") from e datumaro.components.errors.DatasetImportError: Failed to import dataset...

leo-smi

following TCs were reported as failed due to too low accuracy * tests/e2e/cli/detection/test_detection.py::TestToolsOTXDetection::test_otx_eval_openvino[True-Custom_Object_Detection_Gen3_SSD] * AssertionError: trained_performance[k]=0.5714285714285714, exported_performance[k]=0.44444444444444436 * tests/e2e/cli/detection/test_detection.py::TestToolsOTXDetection::test_otx_eval_openvino[True-Object_Detection_YOLOX_L] * AssertionError: trained_performance[k]=0.6666666666666665, exported_performance[k]=0.3999999999999999 * tests/e2e/cli/detection/test_detection.py::TestToolsOTXDetection::test_otx_eval_openvino[False-Custom_Object_Detection_Gen3_SSD] * AssertionError: trained_performance[k]=0.5714285714285714, exported_performance[k]=0.44444444444444436 *...

yunchu

ViT inference does not work by FeatureVectorHook.

1

**Describe the bug** `otx train` throws `AttributeError: 'list' object has no attribute 'size'` for inference time by FeatureVectorHook. **Steps to Reproduce** ``` otx build --task classification --backbone mmcls.VisionTransformer cd otx-workspace-CLASSIFICATION...

JihwanEom

[suggestion] Could not find module 'geos_c.dll'

3

**Describe the bug** ```Could not find module 'geos_c.dll'``` **Steps to Reproduce** 1. running the step [```pip install -e .[full]```](https://openvinotoolkit.github.io/training_extensions/latest/guide/get_started/installation.html) leads to the problem **Environment:** - OS: Windows 10 - Python...

leo-smi

DINO running on small resolution results in crash on topk due to smaller # of proposals than K

1

**Describe the bug** ```bash File "/mnt/sdb/workarea/otx/.tox/unittest-all-py310/lib/python3.10/site-packages/mmdet/models/detectors/single_stage.py", line 102, in simple_test results_list = self.bbox_head.simple_test( File "/mnt/sdb/workarea/otx/.tox/unittest-all-py310/lib/python3.10/site-packages/mmdet/models/dense_heads/base_dense_head.py", line 360, in simple_test return self.simple_test_bboxes(feats, img_metas, rescale=rescale) File "/mnt/sdb/workarea/otx/src/otx/algorithms/detection/adapters/mmdet/models/heads/custom_dino_head.py", line 702, in simple_test_bboxes hidden_states,...

goodsong81

**Describe the bug** While verifying the torch version upgrade from 1.13.1 to 2.0.1, there was integration test error(s) regarding NNCF optimize. [Error log from CI run] * Draft PR: https://github.com/openvinotoolkit/training_extensions/pull/2448...

goodsong81

**Describe the bug** Same with title. **Steps to Reproduce** 1. at E2E CI test

sungmanc