1
0
mirror of https://github.com/immich-app/immich.git synced 2024-12-23 02:06:15 +02:00
Commit Graph

25 Commits

Author SHA1 Message Date
Mert
7e18e69c1c
fix(ml): only use openvino if a gpu is available (#7450)
use `device_type`
2024-02-27 00:45:14 -05:00
Sourav Agrawal
2a75f884d9
Fix Smart Search when using OpenVINO (#7389)
* Fix external_path loading in OpenVINO EP

* Fix ruff lint

* Wrap block in try finally

* remove static input shape code

* add unit test

* remove unused imports

* remove repeat line

* linting

* formatting

---------

Co-authored-by: mertalev <101130780+mertalev@users.noreply.github.com>
2024-02-24 18:22:27 -05:00
renovate[bot]
2b8ec30767
chore(deps): update machine-learning (major) (#6934)
* chore(deps): update machine-learning

* formatting

---------

Co-authored-by: renovate[bot] <29139614+renovate[bot]@users.noreply.github.com>
Co-authored-by: mertalev <101130780+mertalev@users.noreply.github.com>
2024-02-12 22:29:15 -05:00
Mert
0c4df216d7
feat(ml): improve test coverage (#7041)
* update e2e

* tokenizer tests

* more tests, remove unnecessary code

* fix e2e setting

* add tests for loading model

* update workflow

* fixed test
2024-02-11 17:58:56 -05:00
Mert
c9b7f4e690
chore(ml): make execution provider log info-level (#7024)
* change debug log to info

* update test
2024-02-11 06:12:11 +00:00
Mert
79d3342c3d
fix(ml): openvino not working with dynamic axes (#6871)
* convert to static

* add comment about gross code

* formatting

* fixed test

* fix typing

* cleanup

* formatting

* Revert "formatting"

This reverts commit 073965c47e.

* Revert "cleanup"

This reverts commit bb56bd3297.

* formatting
2024-02-03 05:35:26 +00:00
Mert
a84b6f5fb1
feat(ml): conditionally download .armnn models (#6650) 2024-01-28 10:31:59 -05:00
Mert
95cfe22866
feat(ml)!: cuda and openvino acceleration (#5619)
* cuda and openvino ep, refactor, update dockerfile

* updated workflow

* typing fixes

* added tests

* updated ml test gh action

* updated README

* updated docker-compose

* added compute to hwaccel.yml

* updated gh matrix

updated gh matrix

updated gh matrix

updated gh matrix

updated gh matrix

give up

* remove cuda/arm64 build

* add hwaccel image tags to docker-compose

* remove unnecessary quotes

* add suffix to git tag

* fixed kwargs in base model

* armnn ld_library_path

* update pyproject.toml

* add armnn workflow

* formatting

* consolidate hwaccel files, update docker compose

* update hw transcoding docs

* add ml hwaccel docs

* update dev and prod docker-compose

* added armnn prerequisite docs

* support 3.10

* updated docker-compose comments

* formatting

* test coverage

* don't set arena extend strategy for openvino

* working openvino

* formatting

* fix dockerfile

* added type annotation

* add wsl configuration for openvino

* updated lock file

* copy python3

* comment out extends section

* fix platforms

* simplify workflow suffix tagging

* simplify aio transcoding doc

* update docs and workflow for `hwaccel.yml` change

* revert docs
2024-01-21 18:22:39 -05:00
Fynn Petersen-Frey
753292956e
feat(ml): ARMNN acceleration (#5667)
* feat(ml): ARMNN acceleration for CLIP

* wrap ANN as ONNX-Session

* strict typing

* normalize ARMNN CLIP embedding

* mutex to handle concurrent execution

* make inputs contiguous

* fine-grained locking; concurrent network execution

---------

Co-authored-by: mertalev <101130780+mertalev@users.noreply.github.com>
2024-01-11 18:26:46 +01:00
Mert
092a23fd7f
feat(server,ml): remove image tagging (#5903)
* remove image tagging

* updated lock

* fixed tests, improved logging

* be nice

* fixed tests
2023-12-20 20:47:56 -05:00
Mert
935f471ccb
chore(ml): use strict mypy (#5001)
* improved typing

* improved export typing

* strict mypy & check export folder

* formatting

* add formatting checks for export folder

* re-added init call
2023-11-13 10:18:46 -06:00
Mert
328a58ac0d
feat(ml): add face models (#4952)
added models to config dropdown

fixed downloading

updated tests

use hf for face models

formatting
2023-11-11 19:04:49 -06:00
Mert
87a0ba3db3
feat(ml): export clip models to ONNX and host models on Hugging Face (#4700)
* export clip models

* export to hf

refactored export code

* export mclip, general refactoring

cleanup

* updated conda deps

* do transforms with pillow and numpy, add tokenization config to export, general refactoring

* moved conda dockerfile, re-added poetry

* minor fixes

* updated link

* updated tests

* removed `requirements.txt` from workflow

* fixed mimalloc path

* removed torchvision

* cleaner np typing

* review suggestions

* update default model name

* update test
2023-10-31 05:02:04 -05:00
Mert
258b98c262
fix(ml): load models in separate threads (#4034)
* load models in thread

* set clip mode logs to debug level

* updated tests

* made fixtures slightly less ugly

* moved responses to json file

* formatting
2023-09-09 16:02:44 +07:00
Mert
3cf0f5f11b
fix(ml): model downloading improvements (#3988)
* handle ort `NoSuchFile` error, stricter file check

* keep exception order
2023-09-06 08:48:40 +07:00
Mert
41461e0d5d
chore(ml): memory optimisations (#3934) 2023-08-31 18:30:53 -05:00
Mert
54b2779b79
chore(ml): improved logging (#3918)
* fixed `minScore` not being set correctly

* apply to init

* don't send `enabled`

* fix eslint warning

* added logger

* added logging

* refinements

* enable access log for info level

* formatting

* merged strings

---------

Co-authored-by: Alex <alex.tran1502@gmail.com>
2023-08-30 08:22:01 +00:00
Mert
bcc36d14a1
feat(ml)!: customizable ML settings (#3891)
* consolidated endpoints, added live configuration

* added ml settings to server

* added settings dashboard

* updated deps, fixed typos

* simplified modelconfig

updated tests

* Added ml setting accordion for admin page

updated tests

* merge `clipText` and `clipVision`

* added face distance setting

clarified setting

* add clip mode in request, dropdown for face models

* polished ml settings

updated descriptions

* update clip field on error

* removed unused import

* add description for image classification threshold

* pin safetensors for arm wheel

updated poetry lock

* moved dto

* set model type only in ml repository

* revert form-data package install

use fetch instead of axios

* added slotted description with link

updated facial recognition description

clarified effect of disabling tasks

* validation before model load

* removed unnecessary getconfig call

* added migration

* updated api

updated api

updated api

---------

Co-authored-by: Alex Tran <alex.tran1502@gmail.com>
2023-08-29 08:58:00 -05:00
Mert
165b91b068
feat(ml)!: switch image classification and CLIP models to ONNX (#3809) 2023-08-25 06:28:51 +02:00
Mert
fa57853bd2
fix(ml): clear cache if bad zip file (#3600) 2023-08-08 12:33:13 -05:00
Mert
c73832bd9c
refactor(ml): model downloading (#3545)
* download facial recognition models

* download hf models

* simplified logic

* updated `predict` for facial recognition

* ensure download method is called

* fixed repo_id for clip

* fixed download destination

* use st's own `snapshot_download`

* conditional download

* fixed predict method

* check if loaded

* minor fixes

* updated mypy overrides

* added pytest-mock

* updated tests

* updated lock
2023-08-05 21:45:13 -05:00
Mert
df1e8679d9
chore(ml): added testing and github workflow (#2969)
* added testing

* github action for python, made mypy happy

* formatted with black

* minor fixes and styling

* test model cache

* cache test dependencies

* narrowed model cache tests

* moved endpoint tests to their own class

* cleaned up fixtures

* formatting

* removed unused dep
2023-06-27 18:21:33 -05:00
Mert
47982641b2
fix(ml): clear model cache on load error (#2951)
* clear model cache on load error

* updated caught exceptions
2023-06-27 16:01:24 -05:00
Mert
4d3ce0a65e
fixed setting different clip, removed unused stubs (#2987) 2023-06-27 12:21:50 -05:00
Mert
a2f5674bbb
refactor(ml): modularization and styling (#2835)
* basic refactor and styling

* removed batching

* module entrypoint

* removed unused imports

* model superclass,  model cache now in app state

* fixed cache dir and enforced abstract method

---------

Co-authored-by: Alex Tran <alex.tran1502@gmail.com>
2023-06-24 22:18:09 -05:00