Skip to content

Remove all unused quantize settings and flags. #2555

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
awan-10 wants to merge 4 commits into
base: master
Choose a base branch
from
Open

Remove all unused quantize settings and flags. #2555

Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
279bf74
Remove all unused quantize settings and flags.
awan-10 Nov 28, 2022
34ec4b1
Merge branch 'master' into remove-unused-quantize-settings
awan-10 Feb 6, 2023
af4bfb7
Merge branch 'master' into remove-unused-quantize-settings
awan-10 Mar 2, 2023
c13fc07
Merge branch 'master' into remove-unused-quantize-settings
awan-10 Mar 7, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
40 changes: 5 additions & 35 deletions deepspeed/inference/engine.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -69,19 +69,13 @@ def __init__(self, model, config):
self.mpu = config.tensor_parallel.mpu

#self._validate_args(self.mpu, config.replace_with_kernel_inject)
self.quantize_merge_count = 1
self.quantization_scales = None


# these are not needed in the config as we are creating them ourselves in the inference engine
self.ep_group = None # config.moe.ep_group
self.expert_mp_group = None # config.moe.ep_mp_group

self.cuda_graph_created = False
self.checkpoint_engine = TorchCheckpointEngine()
quantization_setting = None
self._init_quantization_setting(
quantization_setting
) # todo: update with the new quant config for weight quant
self.model_profile_enabled = False
self._model_times = []

Expand Down Expand Up @@ -241,21 +235,6 @@ def _create_ep_parallel_group(self, moe_experts):
if dist.get_rank() in expert_mp_comm_ranks:
self.expert_mp_group.update({moe_ep_size: _expert_mp_group})

def _init_quantization_setting(self, quantization_setting):
self.quantize_bits = 8
self.mlp_extra_grouping = False
self.quantize_groups = 1
if type(quantization_setting) is tuple:
self.mlp_extra_grouping, \
self.quantize_groups = quantization_setting
elif quantization_setting is not None:
self.quantize_groups = quantization_setting
log_dist(
f"quantize_bits = {self.quantize_bits} "
f"mlp_extra_grouping = {self.mlp_extra_grouping}, "
f"quantize_groups = {self.quantize_groups}",
[0])

# TODO: remove this function and add this functionality to pydantic config checking
def _validate_args(self, mpu, replace_with_kernel_inject):
# TODO: to support SD pipeline we need to avoid this check for now
Expand Down Expand Up @@ -423,14 +402,11 @@ def _load_checkpoint(self, load_dir, load_module_strict=True, tag=None):
else:
mp_rank = 0 if self.mpu is None else self.mpu.get_model_parallel_rank()

load_path, checkpoint, quantize_config = sd_loader.load(self._config.tensor_parallel.tp_size,
load_path, checkpoint, _ = sd_loader.load(self._config.tensor_parallel.tp_size,
mp_rank,
is_pipe_parallel=is_pipe_parallel,
quantize=(self._config.dtype is torch.int8),
quantize_groups=self.quantize_groups,
mlp_extra_grouping=self.mlp_extra_grouping)

self.quantization_scales, self.quantize_merge_count = quantize_config
quantize=(self._config.dtype is torch.int8)
)

moe, _ = has_moe_layers(self.module)
if moe:
Expand Down Expand Up @@ -463,13 +439,7 @@ def _convert_to_dtype(self, config):
if not isinstance(self.module, torch.nn.Module):
return

if False: #config.dtype is torch.int8 and self.quantization_scales is None:
quantizer = WeightQuantization(mlp_extra_grouping=self.mlp_extra_grouping)
model, self.quantization_scales = quantizer.model_quantize(self.module,
self.injection_dict,
self.quantize_bits,
self.quantize_groups)
elif config.dtype == torch.half:
if config.dtype == torch.half:
self.module.half()
elif config.dtype == torch.bfloat16:
self.module.bfloat16()
Expand Down