vall-e/vall_e/utils/__init__.py

from .utils import (
    dispatch_attribute,
    flatten_dict,
    gather_attribute,
    load_state_dict_non_strict,
    setup_logging,
    to_device,
    tree_map,
    do_gc,
    set_seed,
    passes_policy
)
Rewrite init 2023-08-02 21:53:35 +00:00			`from .utils import (`
			`dispatch_attribute,`
			`flatten_dict,`
			`gather_attribute,`
			`load_state_dict_non_strict,`
			`setup_logging,`
			`to_device,`
			`tree_map,`
			`do_gc,`
backport fix from tortoise_tts with local trainer + loading state when training lora 2024-06-25 18:41:29 +00:00			`set_seed,`
naive model offloading support (handles automatically splitting parts of the model to requested device per memory constraints, either inferred or requested in the yaml, input tensors are automatically migrated to the right device, it SEEMS to work for training under the test trainer when split between GPU and CPU) (this was specifically only because that Flux imagegen model released so I can test it there) 2024-08-02 01:12:06 +00:00			`passes_policy`
Rewrite init 2023-08-02 21:53:35 +00:00			`)`