Public/automatic
1
0
Fork 0
mirror of https://github.com/vladmandic/automatic synced 2026-09-19 17:24:32 +02:00
Code Issues Packages Projects Releases Wiki Activity

warn on quanto with offload

Browse Source 
Signed-off-by: Vladimir Mandic <mandic00@live.com>
This commit is contained in:
Vladimir Mandic
2024-12-10 10:39:13 -05:00
parent 042178fedb
commit 944408e93b
2 changed files with 7 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files
modules/model_quant.py
+5 -2
View File
@@ -58,11 +58,11 @@ def load_torchao(msg='', silent=False):
import torchao
ao = torchao
fn = f'{sys._getframe(2).f_code.co_name}:{sys._getframe(1).f_code.co_name}' # pylint: disable=protected-access
log.debug(f'Quantization: type=quanto version={ao.__version__} fn={fn}') # pylint: disable=protected-access
log.debug(f'Quantization: type=torchao version={ao.__version__} fn={fn}') # pylint: disable=protected-access
return ao
except Exception as e:
if len(msg) > 0:
log.error(f"{msg} failed to import optimum.quanto: {e}")
log.error(f"{msg} failed to import torchao: {e}")
ao = None
if not silent:
raise
@@ -92,6 +92,7 @@ def load_bnb(msg='', silent=False):
def load_quanto(msg='', silent=False):
from modules import shared
global quanto # pylint: disable=global-statement
if quanto is not None:
return quanto
@@ -101,6 +102,8 @@ def load_quanto(msg='', silent=False):
quanto = optimum_quanto
fn = f'{sys._getframe(2).f_code.co_name}:{sys._getframe(1).f_code.co_name}' # pylint: disable=protected-access
log.debug(f'Quantization: type=quanto version={quanto.__version__} fn={fn}') # pylint: disable=protected-access
if shared.opts.diffusers_offload_mode != 'none':
shared.log.error(f'Quantization: type=quanto offload={shared.opts.diffusers_offload_mode} not supported')
return quanto
except Exception as e:
if len(msg) > 0:
modules/shared.py
+2 -2
View File
@@ -549,7 +549,7 @@ options_templates.update(options_section(('backends', "Backend Settings"), {
"diffusers_fuse_projections": OptionInfo(False, "Fused projections"),
"torch_expandable_segments": OptionInfo(False, "Expandable segments"),
"cuda_mem_fraction": OptionInfo(0.0, "Memory limit", gr.Slider, {"minimum": 0, "maximum": 2.0, "step": 0.05}),
"torch_gc_threshold": OptionInfo(80, "GC threshold", gr.Slider, {"minimum": 0, "maximum": 100, "step": 1}),
"torch_gc_threshold": OptionInfo(70, "GC threshold", gr.Slider, {"minimum": 0, "maximum": 100, "step": 1}),
"inference_mode": OptionInfo("no-grad", "Inference mode", gr.Radio, {"choices": ["no-grad", "inference-mode", "none"]}),
"torch_malloc": OptionInfo("native", "Memory allocator", gr.Radio, {"choices": ['native', 'cudaMallocAsync'] }),
@@ -566,7 +566,7 @@ options_templates.update(options_section(('backends', "Backend Settings"), {
"olive_cache_optimized": OptionInfo(True, 'Olive cache optimized models'),
"ipex_sep": OptionInfo("<h2>IPEX</h2>", "", gr.HTML, {"visible": devices.backend == "ipex"}),
"ipex_optimize": OptionInfo([], "IPEX Optimize for Intel GPUs", gr.CheckboxGroup, {"choices": ["Model", "VAE", "Text Encoder", "Upscaler"], "visible": devices.backend == "ipex"}),
"ipex_optimize": OptionInfo([], "IPEX Optimize", gr.CheckboxGroup, {"choices": ["Model", "VAE", "Text Encoder", "Upscaler"], "visible": devices.backend == "ipex"}),
"openvino_sep": OptionInfo("<h2>OpenVINO</h2>", "", gr.HTML, {"visible": cmd_opts.use_openvino}),
"openvino_devices": OptionInfo([], "OpenVINO devices to use", gr.CheckboxGroup, {"choices": get_openvino_device_list() if cmd_opts.use_openvino else [], "visible": cmd_opts.use_openvino}), # pylint: disable=E0606