diff options
Diffstat (limited to 'bitsandbytes/optim/optimizer.py')
| -rw-r--r-- | bitsandbytes/optim/optimizer.py | 31 |
1 files changed, 28 insertions, 3 deletions
diff --git a/bitsandbytes/optim/optimizer.py b/bitsandbytes/optim/optimizer.py index cfbd72e..5a5bb1e 100644 --- a/bitsandbytes/optim/optimizer.py +++ b/bitsandbytes/optim/optimizer.py @@ -26,6 +26,7 @@ class GlobalOptimManager(object): self.index2config = {} self.optimizer = None self.uses_config_override = False + self.module_weight_config_triple = [] @classmethod def get_instance(cls): @@ -77,12 +78,16 @@ class GlobalOptimManager(object): if id(p) in self.pid2config:self.pid2config[id(p)].update(key_value_dict) else: self.pid2config[id(p)] = key_value_dict + def register_module_override(self, module, param_name, config): + self.module_weight_config_triple.append((module, param_name, config)) + + class Optimizer8bit(torch.optim.Optimizer): def __init__(self, params, defaults, optim_bits=32): super(Optimizer8bit, self).__init__(params, defaults) - self.checked_if_on_gpu = False + self.initialized = False self.name2qmap = {} self.mng = GlobalOptimManager.get_instance() @@ -172,7 +177,6 @@ class Optimizer8bit(torch.optim.Optimizer): self.__setstate__({'state': state, 'param_groups': param_groups}) def to_gpu(self): - self.checked_if_on_gpu = True for gindex, group in enumerate(self.param_groups): for pindex, p in enumerate(group['params']): if p in self.state: @@ -181,6 +185,23 @@ class Optimizer8bit(torch.optim.Optimizer): if isinstance(v, torch.Tensor): self.state[p][k] = v.to(p.device) + def check_overrides(self): + for module, attr, config in self.mng.module_weight_config_triple: + pmodule = getattr(module, attr) + assert pmodule is not None + assert isinstance(pmodule, torch.Tensor) or isinstance(pmodule, torch.Parameter) + found = False + for gindex, group in enumerate(self.param_groups): + if found: break + for pindex, p in enumerate(group['params']): + if found: break + if id(p) == id(pmodule): + # found the matching parameter + # init override + self.mng.pid2config[id(p)] = config + self.mng.index2config[(gindex, pindex)] = self.mng.pid2config[id(p)] + found = True + @torch.no_grad() def step(self, closure=None): """Performs a single optimization step. @@ -196,7 +217,11 @@ class Optimizer8bit(torch.optim.Optimizer): overflows = [] - if not self.checked_if_on_gpu: self.to_gpu() # needed for fairseq pure fp16 training + if not self.initialized: + self.check_overrides() + self.to_gpu() # needed for fairseq pure fp16 training + self.initialized = True + for gindex, group in enumerate(self.param_groups): for pindex, p in enumerate(group['params']): if p.grad is None: |