Alexander Borzunov 6137b1b4b0 Replace .make_sequence(..., mode="random") with mode="max_throughput" (#313) ... We need to sample the next server using its throughput as the weight to actually achieve max throughput for fine-tuning. As an example, imagine a situation where we have 3 servers with throughputs [1000, 500, 1] hosting the same blocks, then compare the uniform and weighted sampling strategies.		2023-05-09 22:38:20 +04:00
..
scripts	Fix arguments in remove_old_models.py (#153)	2022-12-13 19:01:12 +03:00
conftest.py	Fix logging: do not duplicate lines, enable colors in Colab (#156)	2022-12-15 09:12:18 +04:00
test_aux_functions.py	Speed up loading blocks using init with meta weights (#285)	2023-03-13 00:49:04 +03:00
test_block_exact_match.py	Refactor RemoteSequenceManager (#309)	2023-05-07 13:41:13 +04:00
test_chained_calls.py	Refactor RemoteSequenceManager (#309)	2023-05-07 13:41:13 +04:00
test_full_model.py	Speed up loading blocks using init with meta weights (#285)	2023-03-13 00:49:04 +03:00
test_priority_pool.py	Fix issues related to petals as a module (#159)	2022-12-16 09:09:06 +04:00
test_remote_sequential.py	Refactor RemoteSequenceManager (#309)	2023-05-07 13:41:13 +04:00
test_sequence_manager.py	Replace .make_sequence(..., mode="random") with mode="max_throughput" (#313)	2023-05-09 22:38:20 +04:00
test_server_stats.py	Refactor RemoteSequenceManager (#309)	2023-05-07 13:41:13 +04:00
test_tensor_parallel.py	Speed up loading blocks using init with meta weights (#285)	2023-03-13 00:49:04 +03:00
test_utils.py	Implement RemoteSequential slicing and extra repr, add tests (#30)	2022-07-19 04:28:04 +03:00
test.id	Add automated tests (#23)	2022-07-16 01:59:23 +03:00