Commit Graph

  • 1def0a6901 Update github repo link main clearml 2025-01-13 18:40:02 +0200
  • 9f51a9334f Fix torch import clearml 2024-12-16 18:51:58 +0200
  • aff27c62b8 Fix gRPC errors print stack traces and full verbose details. Add support for controlling error printouts using CLEARML_SERVING_AIO_RPC_IGNORE_ERRORS and CLEARML_SERVING_AIO_RPC_VERBOSE_ERRORS (pass a whitespace-separated list of error codes or error names) clearml 2024-12-12 23:57:21 +0200
  • b67eba1093
    Merge 61be8733c8 into 724c99c605 eajechiloae 2024-08-15 16:46:35 +0300
  • 61be8733c8 remove dbg print Eugen Ajechiloae 2024-08-15 16:45:36 +0300
  • 64909cdd65 remove dbg prints Eugen Ajechiloae 2024-08-15 16:38:27 +0300
  • df098f7749 dont call gc collect on remove Eugen Ajechiloae 2024-08-15 16:25:59 +0300
  • fd8b7604a9 call gc on remove as well Eugen Ajechiloae 2024-08-15 16:18:41 +0300
  • d0131d0b58 set status in exit Eugen Ajechiloae 2024-08-15 16:00:40 +0300
  • 4fa5b5aa36 force exit using os._exit Eugen Ajechiloae 2024-08-15 15:55:34 +0300
  • 1f4e1599e3 let gc handle unload Eugen Ajechiloae 2024-08-15 15:23:29 +0300
  • 5d5188de40 add debugging comment Eugen Ajechiloae 2024-08-14 21:20:13 +0300
  • 79b987ac0c more dbg prints Eugen Ajechiloae 2024-08-14 21:03:25 +0300
  • 67325df4d5 change dbg print Eugen Ajechiloae 2024-08-14 20:50:57 +0300
  • f95939cffe more debug prints Eugen Ajechiloae 2024-08-14 20:46:45 +0300
  • 7d801a11da add degug print Eugen Ajechiloae 2024-08-14 19:47:58 +0300
  • 6a8e616256 not tested - exit on cuda oom Eugen Ajechiloae 2024-08-14 19:43:44 +0300
  • 6ef1f67ad0 initial commit for oom issues - allow for model unload and restart serving on crash Eugen Ajechiloae 2024-08-14 17:01:29 +0300
  • e92bd66670
    Merge 91304a1c50 into 724c99c605 IlyaMescheryakov1402 2024-07-25 13:16:24 +0300
  • 91304a1c50 set CLEARML_SERVING_NUM_PROCESS IlyaMescheryakov1402 2024-07-25 13:14:54 +0300
  • 4d0368da27 fix serving_num_processes IlyaMescheryakov1402 2024-07-25 12:52:27 +0300
  • 724c99c605
    Add clearml_serving_inference restart on CUDA OOM (#75) IlyaMescheryakov1402 2024-07-07 15:54:08 +0300
  • 666ce26ab2
    Add exit-on-error option for tritonserver (#76) stephanbertl 2024-07-07 14:51:23 +0200
  • a0bfdf2c92 add OOM handler for MIG profiles Meshcheryakov Ilya 2024-05-30 00:58:17 +0300
  • 375431b75c added exit-on-error option for tritonserver. Fixes allegroai/clearml-serving#60 = 2024-05-03 15:32:45 +0200
  • 6859920848 initial commit Meshcheryakov Ilya 2024-04-16 00:54:35 +0300
  • 8390a06aff
    Merge 20bd0a5c34 into 7ba356efc9 Emilio Botero 2024-03-18 12:50:09 -0700
  • 7ba356efc9 Update README allegroai 2024-03-11 16:54:08 +0200
  • 047a120100 Fix broken requirements allegroai 2024-03-01 13:13:26 +0200
  • 71c104c9df Add exception prints to serving session Task and inference Task, for better debugging capabilities Add report instance ID when reporting back to the main serving session task allegroai 2024-03-01 13:12:56 +0200
  • c658780d97 Update README allegroai 2024-03-01 00:02:56 +0200
  • 8df521b949 Fix python < 3.10 support Fix custom_async engine Suppress warning 1.3.1 allegroai 2024-02-27 09:45:32 +0200
  • bca162810b Add async pipeline example (--engine custom_async) allegroai 2024-02-27 09:44:53 +0200
  • 9488ead080 Add async pipeline version allegroai 2024-02-27 09:44:20 +0200
  • 8f6feef938 Fix broken link allegroai 2024-02-27 09:43:47 +0200
  • 3611a040f7 Fix requirements allegroai 2024-02-27 09:43:18 +0200
  • dc6fd46a46 Update requirements allegroai 2024-02-26 11:34:21 +0200
  • 0f4122247d Fix ping serving session task to make sure everyone knows we are alive allegroai 2024-02-26 11:31:58 +0200
  • f2ba37c8d4 Fix version enabled endpoints on Triton engine were not called allegroai 2024-02-26 11:27:12 +0200
  • 4ac13d5287 Fix requirements issue allegroai 2024-01-11 15:21:13 +0200
  • 4335ebd340 Improve preprocess template docstring allegroai 2024-01-06 17:58:20 +0200
  • b42a0b0cfc Fix requirements allegroai 2024-01-06 17:57:50 +0200
  • 368a03dc70 Fix internal ValueError exception should return 422 (not 404 as before) allegroai 2024-01-06 17:55:49 +0200
  • 02fef01657 Add python 3.11 to support matrix allegroai 2024-01-06 17:54:56 +0200
  • 91dc126ab3
    Merge branch 'allegroai:main' into main pollfly 2023-11-28 13:39:39 +0200
  • 6d81e7b754
    Update serving requirements.txt Ilia Bakalets 2023-11-28 15:35:36 +0700
  • 6c4bece663
    Fix Pillow vulnerability "libwebp: OOB write in BuildHuffmanTable" Jake Henning 2023-10-04 13:23:42 +0300
  • c20bbd66b9
    Fix Pillow vulnerability "libwebp: OOB write in BuildHuffmanTable" Jake Henning 2023-10-04 13:22:46 +0300
  • 05cbfade2a Update requirements allegroai 2023-09-23 18:03:24 +0300
  • cc5823cfc6 Update README allegroai 2023-09-23 17:48:25 +0300
  • 82ade1e24a Fix check triton config.pbtxt for missing values or colliding specifications (#62) allegroai 2023-09-23 17:42:57 +0300
  • 96b335e3c2 Fix ignore auto detected platform when passing config.pbtxt with platform entry allegroai 2023-09-23 17:40:54 +0300
  • 083635c803 Add str type to Triton type conversion allegroai 2023-09-23 17:36:28 +0300
  • 58d826e427 Fail-safe Kafka pulling allegroai 2023-09-23 17:36:01 +0300
  • e4c07c756a Add traceback for failing to load preprocess class (#57) allegroai 2023-09-23 17:35:21 +0300
  • 2e83c14848 Fix Slack link revital 2023-06-04 11:18:20 +0300
  • 4a737b95c6
    Update README.md Jake Henning 2023-05-17 19:42:28 +0300
  • 115770547c
    Adds missing await (#55) Amir Mousavi 2023-05-08 17:46:52 +0800
  • 7186863997 Adds missing await Amir Mousavi 2023-05-08 12:30:45 +0800
  • 20bd0a5c34 Merge remote-tracking branch 'origin/update-fork' emilio 2023-04-20 09:55:32 -0400
  • 2d3ac1fe63 fix model name allegroai 2023-04-13 01:02:27 +0300
  • eaa2b8a9e8
    Update README.md 1.3.0 Allegro AI 2023-04-13 00:36:06 +0300
  • fe04382fdc version bump allegroai 2023-04-12 23:39:05 +0300
  • aca8b4aa03 Upgrade to python 3.11 allegroai 2023-04-12 23:38:56 +0300
  • d9599ba942 docstring allegroai 2023-04-12 23:35:00 +0300
  • 78a03cc166 Register models on serving session allegroai 2023-04-12 23:34:49 +0300
  • 3bddccbaef Add new preprocess template allegroai 2023-04-12 23:31:32 +0300
  • 31a4ebb965 Add CLEARML_GRPC_* environement variable support to configure grpc channel options (notice CLEARML_GRPC_var is converted into grpc.var when setting grpc channel, casing does not change) #49 allegroai 2023-04-12 23:30:59 +0300
  • b16c51e631 typo allegroai 2023-04-12 23:29:15 +0300
  • 095f67f924 Adds proposal skeleton Amir Mousavi 2023-04-12 14:36:25 +0800
  • 41df4eb287
    Removes unnessary flatten (#50) Amir Mousavi 2023-04-11 14:15:54 +0800
  • 214010b788 Python version as build argument emilio 2023-04-10 10:18:01 -0400
  • 736c529186 Revert "Use python 3.8 in serving container" emilio 2023-04-10 10:09:01 -0400
  • f831783645 Use python 3.8 in serving container emilio 2023-04-10 09:45:59 -0400
  • c844a13b07 Removes unnessary flatten Amir Mousavi 2023-04-04 20:28:04 -0700
  • 6fb0006776
    Add environment vars for advanced usage (#45) Sebastian Meyer 2023-03-27 12:57:05 +0200
  • c56cf6a5bd changed np.bool deprecated in numpy 1.20.0, changed to bool anon-it 2023-03-24 23:13:54 +0000
  • 2a6c7c18fe
    added environment vars for advanced usage Sebastian Meyer 2023-03-23 16:15:50 +0100
  • a04d1bda03
    Remove never-used but now deprecated np.int (#42) Victor Sonck 2023-02-07 07:05:01 +0100
  • 24c6cfe683 Remove never-used but now deprecated np.int Victor Sonck 2023-01-26 15:36:57 +0100
  • e09e636214 version bump 1.2.0 allegroai 2022-10-08 14:41:53 +0300
  • 0c5d9820df Optimize containers allegroai 2022-10-08 02:22:32 +0300
  • 219fa308df Add huggingface transformers example allegroai 2022-10-08 02:12:31 +0300
  • 363276ba79 Add raw bytes serving example allegroai 2022-10-08 02:12:18 +0300
  • 395a547c04 Optimize async processing for increased speed allegroai 2022-10-08 02:11:57 +0300
  • f4aaf095a3 Add triton logging control allegroai 2022-10-08 02:10:46 +0300
  • 4d4091281a add more model frameworks support allegroai 2022-10-08 02:10:34 +0300
  • 1569f08d1d
    Merge pull request #37 from codechem/feature/bytes-payload Allegro AI 2022-10-08 00:01:01 +0300
  • 29c9732e8e [DEV] feature/bytes-payload | Add examples Aleksandar Ivanovski 2022-10-07 09:19:17 +0200
  • d89d1370d8 [DEV] feature/bytes-payload | Add typing Aleksandar Ivanovski 2022-10-06 16:01:31 +0200
  • 2aa91a3d43 [DEV] feature/bytes-payload | Handle keys when req is bytes Aleksandar Ivanovski 2022-10-06 15:13:36 +0200
  • 09ed480bc2 [DEV] feature/bytes-payload | Add bytes as payload Aleksandar Ivanovski 2022-10-06 13:31:54 +0200
  • 4b52103636
    Update README.md Allegro AI 2022-09-03 00:46:52 +0300
  • 748ab67be0 version bump 1.1.0 allegroai 2022-09-03 00:02:33 +0300
  • f4eed33f10 Add Triton support for variable length requests, adds support for HuggingFace Transformers Add triton_grpc_compression=False (default) for grpc connection compression control allegroai 2022-09-02 23:41:49 +0300
  • c6c40c9a36 Add support for Preprocess class inside a module (i.e. __init__.py with subfolders) allegroai 2022-09-02 21:50:41 +0300
  • 0588f58405 Fix examples requirements.txt allegroai 2022-09-02 21:48:06 +0300
  • 6005e238ca Add custom model serving example allegroai 2022-07-15 22:27:55 +0300
  • 4f31c4c178 Add python 3.10 to support table allegroai 2022-07-15 22:27:35 +0300
  • 7fa3f2e94a Fix --aux-config not working, issue #30 allegroai 2022-07-15 22:27:10 +0300