Oleksandr Bezdieniezhnykh 3984507221 [AZ-180] Fix INT8 conversion: set FP16 flag alongside INT8 for TensorRT 10.x ...

In TensorRT 10.x, INT8 conversion requires FP16 to be set as a fallback for
network layers (e.g. normalization ops in detection models) that have no INT8
kernel implementation. Without FP16, build_serialized_network can return None
on Jetson for YOLO-type models. INT8 flag is still the primary precision;
FP16 is only the layer-level fallback within the same engine.

Made-with: Cursor

2026-04-02 07:32:16 +03:00

..

test_ai_config_from_dict.py

[AZ-173] [AZ-174] Stream-based detection API and DB-driven AI config

2026-03-31 06:30:22 +03:00

test_az174_db_driven_config.py

[AZ-178] Fix Critical/High security findings: auth, CVEs, non-root containers, per-job SSE

2026-04-02 06:32:12 +03:00

test_az175_api_calls.py

[AZ-175] Media table integration with XxHash64 content hashing and status lifecycle

2026-03-31 06:36:56 +03:00

test_az177_video_single_write.py

[AZ-180] Fix pre-existing test failures: patch JWT_SECRET in auth-dependent tests

2026-04-02 07:23:10 +03:00

test_az178_realvideo_streaming.py

[AZ-178] Add real-video streaming test, update e2e tests, mark task done

2026-04-01 05:02:25 +03:00

test_az178_streaming_video.py

[AZ-180] Fix pre-existing test failures: patch JWT_SECRET in auth-dependent tests

2026-04-02 07:23:10 +03:00

test_az180_jetson_int8.py

[AZ-180] Fix INT8 conversion: set FP16 flag alongside INT8 for TensorRT 10.x

2026-04-02 07:32:16 +03:00

test_media_hash.py

[AZ-175] Media table integration with XxHash64 content hashing and status lifecycle

2026-03-31 06:36:56 +03:00