You cannot select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
petals/tests
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.
1 year ago
..
scripts Fix arguments in remove_old_models.py (#153) 1 year ago
conftest.py Fix logging: do not duplicate lines, enable colors in Colab (#156) 1 year ago
test.id Add automated tests (#23) 2 years ago
test_aux_functions.py Speed up loading blocks using init with meta weights (#285) 1 year ago
test_block_exact_match.py Refactor RemoteSequenceManager (#309) 1 year ago
test_chained_calls.py Refactor RemoteSequenceManager (#309) 1 year ago
test_full_model.py Speed up loading blocks using init with meta weights (#285) 1 year ago
test_priority_pool.py Fix issues related to `petals` as a module (#159) 1 year ago
test_remote_sequential.py Refactor RemoteSequenceManager (#309) 1 year ago
test_sequence_manager.py Replace .make_sequence(..., mode="random") with mode="max_throughput" (#313) 1 year ago
test_server_stats.py Refactor RemoteSequenceManager (#309) 1 year ago
test_tensor_parallel.py Speed up loading blocks using init with meta weights (#285) 1 year ago
test_utils.py Implement RemoteSequential slicing and extra repr, add tests (#30) 2 years ago