Releases: replicate/cog
Releases · replicate/cog
v0.10.0-alpha5
Full Changelog: v0.10.0-alpha3...v0.10.0-alpha4
Scary temporary commit for a hemorrhaging-edge release. This adds concurrency to the config and significantly changes the behavior of cog.Path, does something unsavory to upload very large files, and actually enables concurrency.
- add concurrency to config
- this basically works!
- more descriptive names for predict functions
- maybe pass through prediction id and try to make cancelation do both?
- don't cancel from signal handler if a loop is running. expose worker busy state to runner
- move handle_event_stream to PredictionEventHandler
- make setup and canceling work
- drop some checks around cancelation
- try out eager_predict_state_change
- keep track of multiple runner prediction tasks to make idempotent endpoint return the same result and fix tests somewhat
- fix idempotent tests
- fix remaining errors?
- worker predict_generator shouldn't be eager
- wip: make the stuff that handles events and sends webhooks etc async
- drop Runner._result
- drop comments
- inline client code
- get started
- inline webhooks
- move clients into runner, switch to httpx, move create_event_handler into runner
- add some comments
- more notes
- rip out webhooks and most of files and put them in a new ClientManager that handles most of everything. inline upload_files for that
- move create_event_handler into PredictionEventHandler.init
- fix one test
- break out Path.validate into value_to_path and inline get_filename and File.validate
- split out URLPath into BackwardsCompatibleDataURLTempFilePath and URLThatCanBeConvertedToPath with the download part of URLFile inlined
- let's make DataURLTempFilePath also use convert and move value_to_path back to Path.validate
- use httpx for downloading input urls and follow redirects
- take get_filename back out for tests
- don't upload in http and delete cog/files.py
- drop should_cancel
- prediction->request
- split up predict/inner/prediction_ctx into enter_predict/exit_predict/prediction_ctx/inner_async_predict/predict/good_predict as one way to do it. however, exposing all of those for runner predict enter/coro exit still sucks, but this is still an improvement
- bigish change: inline predict_and_handle_errors
- inline make_error_handler into setup
- move runner.setup into runner.Runner.setup
- add concurrency to config in go
- try explicitly using prediction_ctx enter and exit
- make runner setup more correct and marginally better
- fix a few tests
- notes
- wip ClientManager.convert
- relax setup argument requirement to str
- glom worker into runner
- add logging message
- fix prediction retry and improve logging
- split out handle_event
- use CURL_CA_BUNDLE for file upload
- clean up comments
- dubious upload fix
- small fixes
- attempt to add context logging?
- tweak names
- fix error for predictionOutputType(multi=False)
- improve comments
- fix lints
- add a note about this release
v0.10.0-alpha4
Full Changelog: v0.10.0-alpha3...v0.10.0-alpha4
Scary temporary commit for a hemorrhaging-edge release. This adds concurrency to the config and significantly changes the behavior of cog.Path, does something unsavory to upload very large files, and actually enables concurrency.
- add concurrency to config
- this basically works!
- more descriptive names for predict functions
- maybe pass through prediction id and try to make cancelation do both?
- don't cancel from signal handler if a loop is running. expose worker busy state to runner
- move handle_event_stream to PredictionEventHandler
- make setup and canceling work
- drop some checks around cancelation
- try out eager_predict_state_change
- keep track of multiple runner prediction tasks to make idempotent endpoint return the same result and fix tests somewhat
- fix idempotent tests
- fix remaining errors?
- worker predict_generator shouldn't be eager
- wip: make the stuff that handles events and sends webhooks etc async
- drop Runner._result
- drop comments
- inline client code
- get started
- inline webhooks
- move clients into runner, switch to httpx, move create_event_handler into runner
- add some comments
- more notes
- rip out webhooks and most of files and put them in a new ClientManager that handles most of everything. inline upload_files for that
- move create_event_handler into PredictionEventHandler.init
- fix one test
- break out Path.validate into value_to_path and inline get_filename and File.validate
- split out URLPath into BackwardsCompatibleDataURLTempFilePath and URLThatCanBeConvertedToPath with the download part of URLFile inlined
- let's make DataURLTempFilePath also use convert and move value_to_path back to Path.validate
- use httpx for downloading input urls and follow redirects
- take get_filename back out for tests
- don't upload in http and delete cog/files.py
- drop should_cancel
- prediction->request
- split up predict/inner/prediction_ctx into enter_predict/exit_predict/prediction_ctx/inner_async_predict/predict/good_predict as one way to do it. however, exposing all of those for runner predict enter/coro exit still sucks, but this is still an improvement
- bigish change: inline predict_and_handle_errors
- inline make_error_handler into setup
- move runner.setup into runner.Runner.setup
- add concurrency to config in go
- try explicitly using prediction_ctx enter and exit
- make runner setup more correct and marginally better
- fix a few tests
- notes
- wip ClientManager.convert
- relax setup argument requirement to str
- glom worker into runner
- add logging message
- fix prediction retry and improve logging
- split out handle_event
- use CURL_CA_BUNDLE for file upload
- clean up comments
- dubious upload fix
- small fixes
- attempt to add context logging?
- tweak names
- fix error for predictionOutputType(multi=False)
- improve comments
- fix lints
- add a note about this release
v0.9.4
v0.10.0-alpha3
v0.9.3
v0.10.0-alpha2
v0.10.0-alpha1
Full Changelog: v0.9.0+async.1...v0.10.0-alpha1
v0.9.2
What's Changed
- tweak schema file console message by @technillogue in #1371
- Add ENV NVIDIA_DRIVER_CAPABILITIES=all to permit Video SDK transcoding, 3D rendering, and X11 OpenGL Display by @nelsonjchen in #1288
- Makefile: don't hardcode bash path and use the one from PATH instead by @yorickvP in #1387
- cog predict: specific hint if image name contains '=' by @yorickvP in #1388
- Fix types, run typechecker in CI by @yorickvP in #1393
- create /var/run if it doesn't exist by @technillogue in #1402
- update compatibility matrices: support cuda 12.2 by @technillogue in #1396
- Fix FastAPI state hack, freeze pyright version by @yorickvP in #1415
- Stricter image cleanup by @nickstenning in #1412
- fix(python_test): added "slow" hypothesis profile in python worker test by @dkhokhlov in #1416
- fix(test): 'docker' 'rmi' failed in python integration tests by @dkhokhlov in #1417
- fix: linting in python code by @dkhokhlov in #1419
- run linters in CI by @technillogue in #1418
- fix: make target for python dev env setup by @dkhokhlov in #1421
- Don't return a PredictionResponse from PredictionRunner.setup by @nickstenning in #1433
- make a website for the docs by @zeke in #1435
- Set a reasonable default parallelism on non-Linux platforms by @nickstenning in #1436
- fix: handle early exceptions in model setup by @dkhokhlov in #1428
- check if setup_result is None by @technillogue in #1443
- Add note for naming
Output
class by @iamargentum in #1444 - fix: update the version typing_extensions to use latest version by @mbukeRepo in #1439
- document how to install git-hosted python packages by @zeke in #1383
- add upgrade instructions to README by @zeke in #1445
- build: fix goreleaser configuration by @caarlos0 in #1466
- fix: test fixture does not remove docker image by @dkhokhlov in #1451
- Capitalize OS name in release artifacts by @zeke in #1467
- remove upgrade anchor from readme by @zeke in #1468
- Add training schemas and operations to OpenAPI schema, when available by @mattt in https://githu
New Contributors
- @nelsonjchen made their first contribution in #1288
- @yorickvP made their first contribution in #1387
- @dkhokhlov made their first contribution in #1416
- @iamargentum made their first contribution in #1444
- @mbukeRepo made their first contribution in #1439
- @caarlos0 made their first contribution in #1466
Full Changelog: v0.9.0-beta10...v0.9.0
v0.9.1
This was created in error.
v0.9.0+async.1
What's Changed
- un-revert async runner
- un-revert async predict
Full Changelog: v0.9.0-rc1...v0.9.0+async.1