training_extensions

training_extensions copied to clipboard

Error when fine tuning the horizontal text detection

1

Hi, I encounter this error when I am running the fine-tuning step in [https://github.com/openvinotoolkit/training_extensions/tree/misc/models/object_detection/model_templates/horizontal-text-detection](url) Step 4.b. Evaluate Traceback (most recent call last): File "/home/tham/training_extensions/external/mmdetection/tools/test.py", line 260, in No CUDA runtime...

aljie28

NonMaxSuppression parameters are not deployed correctly

4

I trained an object detection model based on **YOLOX** with [otx](https://github.com/openvinotoolkit/training_extensions) and exported / optimized it as openvino model. The network includes a `NonMaxSuppression` operator to post-process the detected objects...

cansik

Error in get_dataset_adapter() in OTX API. "AttributeError: VISUAL_PROMPTING." and "TypeError: get_dataset_adapter() missing 1 required positional argument: 'train_type'".

1

**Describe the bug** While running the OTX API following [this doc](https://openvinotoolkit.github.io/training_extensions/latest/guide/tutorials/advanced/api_tutorial.html#model-template-and-dataset-loading) , in executing below command, ```python dataset_adapter = get_dataset_adapter(task_type = model_template.task_type, # set a path to the root folder...

VishwaBhushanBH

**Describe the bug** When training tilling ins-seg model with "coco decremental dataset" which is used for regression test dataset, an error is raised due to index out of range error...

eunwoosh

Anomaly task sometimes show large performance drop before and after export, optimize, deploy

2

**Describe the bug** In regression task, some category of mvtec dataset shows large performance drop before and after export, optimize, deploy **Steps to Reproduce** 1. Run regression test 2. 3....

jaegukhyun

**Describe the bug** Action detector trained on toy dataset shows 0.0 accuracy when it is exported. Fortunately, action detector trained on public dataset don't show this issue. **Steps to Reproduce**...

jaegukhyun

Got failures from the regression tests for the detetion task with tiling all failures are coming from Custom_Object_Detection_YOLOX template. ``` ================================================ short test summary info ================================================= FAILED tests/regression/detection/test_tiling_detection.py::TestRegressionTilingDetection::test_otx_train[Custom_Object_Detection_YOLOX] - AssertionError:...

yunchu

**Describe the bug** https://github.com/openvinotoolkit/training_extensions/blob/develop/otx/algorithms/classification/adapters/mmcls/configurer.py#L632-L647 Current loss patching logic is dynamically and forcefully changing the loss type to either `CrossEntropy` or `IBLoss`. This code prevents the loss type from being updated...

JihwanEom

Decreased performance with SupCon Loss in Incremental Learning Research Framework

4

Hi OTX team, Our results show that using Supervised Contrastive Loss shows decreased performance. I discussed this with Intel Labs, who noted that this might be caused by two things....

Daankrol

**Describe the bug** GPU memory increases although input image size decreases in instance segmentation. When I trained both instance segmentation model with 700x700 input size using wgisd-5 dataset, GPU memory...

eunwoosh