fix: editing the code

Signed-off-by: omobayode.fagbohungbe <omobayode.fagbohungbe@ibm.com>

Author: bayo-ibm

fms_mo/dq.py

@@ -145,7 +145,7 @@ def run_dq(model_args, data_args, opt_args, fms_mo_args):
     ]
     qcfg["large_model"] = any(
         name in model_args.model_name_or_path for name in known_large_models
-    ) or (gpu_mem_util_per > 0.1)
+    ) or (gpu_mem_util_per > 0.7)
     dev = "cpu" if qcfg["large_model"] else "cuda"
     if model_args.device_map is None:
         model.to(dev)
@@ -175,7 +175,6 @@ def run_dq(model_args, data_args, opt_args, fms_mo_args):
     qcfg["model"] = model_args.model_name_or_path
     qcfg["smoothq"] = True
     qcfg["plotsvg"] = False
-    
 
     calibration_dataset = load_from_disk(data_args.training_data_path)
     calibration_dataset = calibration_dataset.with_format("torch")
@@ -186,8 +185,7 @@ def run_dq(model_args, data_args, opt_args, fms_mo_args):
         collate_fn=default_data_collator,
         batch_size=1,
     )
-    #print(fms_mo_args)
-    #ii
+
     # For loading or creating smoothquant scale. Sometimes we may include scales in ckpt as well.
     if not fms_mo_args.inference:
         scale_file = Path(f"./act_scales/{qcfg['model'].replace('/', '-')}.pt")

fms_mo/prep.py

@@ -902,7 +902,7 @@ def qmodel_prep(
             model, device_ids=DPorDDPdevices
         )
 
-    qconfig_save(qcfg, fname="qcfg2.json")
+    qconfig_save(qcfg, fname="qcfg.json")
     qcfg["tb_writer"] = tb_writer
 
     logger.info(f"--- Quantized model --- \n{model}\n")

fms_mo/training_args.py

@@ -176,7 +176,6 @@ class FMSMOArguments(TypeChecker):
     inference: bool = field(default=False)
 
 
-
 @dataclass
 class GPTQArguments(TypeChecker):
     """Dataclass for GPTQ related arguments that will be used by auto-gptq."""