15 Commits (ab41223b17c17dd1035a42318b03d4b92decd063)

Author	SHA1	Message	Date
justheuristic	a2634001e9	Reduce vocabulary size in test model, fix bug in routing when overlapped (#45) ... This PR reduces this vocabulary size to save memory during conversion, keeping only the first 50k tokens As a result, * tests that load client-side embeddings need significantly less RAM * we can now run CI tests with 4 servers instead of 2 - needed to test routing - see bugs uncovered * some of the servers now use load balancing * CI convert_model now takes 4-5 minutes (was 6-7)	2 years ago
justheuristic	e2711a033b	Add automated tests (#23) ... This PR will run basic tests automatically on each subsequent PR - convert a small model on every PR - run existing tests on every PR - enforce black / isort - require checks on merge - make sure tests are not flappy Co-authored-by: Alexander Borzunov <hxrussia@gmail.com> Co-authored-by: Dmitry Baranchuk <dmitrybaranchuk@gmail.com>	2 years ago
justheuristic	5695897620	fix imports	2 years ago
justheuristic	de556c99be	straighten import order	2 years ago
justheuristic	90d65e58aa	set default DHT prefix	2 years ago
justheuristic	41e5a95e8e	set client branch to main by default; ... remove the concept of base branch (redundant)	2 years ago
justheuristic	899cefe588	set client branch to main by default; ... remove the concept of base branch (redundant)	2 years ago
justheuristic	4ad845bce3	black-isort	2 years ago
Dmitry Baranchuk	be83e6d0cb	refactoring	2 years ago
Dmitry Baranchuk	d969172208	set requires_grad=False, lm_layer -> h @ word_embeddings, rm lm_layer from comverted_model	2 years ago
justheuristic	83cd4412a1	black-isort	2 years ago
justheuristic	6047a2ffe0	push config and tokenizer separately	2 years ago
justheuristic	b6f3bbfd97	black	2 years ago
justheuristic	84de19fb1a	better status logs	2 years ago
justheuristic	1555d98f66	push converted model to hub	2 years ago