Add op statistics dump for woq (#1876)

Signed-off-by: Kaihui-intel <kaihui.tang@intel.com>

Author: Kaihui-intel

neural_compressor/torch/algorithms/weight_only/utility.py

@@ -12,8 +12,6 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
-import math
-
 import torch
 
 from neural_compressor.torch.utils import accelerator, device_synchronize, logger

neural_compressor/torch/quantization/algorithm_entry.py

@@ -45,7 +45,14 @@
     StaticQuantConfig,
     TEQConfig,
 )
-from neural_compressor.torch.utils import get_quantizer, is_ipex_imported, logger, postprocess_model, register_algo
+from neural_compressor.torch.utils import (
+    dump_model_op_stats,
+    get_quantizer,
+    is_ipex_imported,
+    logger,
+    postprocess_model,
+    register_algo,
+)
 from neural_compressor.torch.utils.constants import PT2E_DYNAMIC_QUANT, PT2E_STATIC_QUANT
 
 
@@ -89,6 +96,7 @@ def rtn_entry(
     model.qconfig = configs_mapping
     model.save = MethodType(save, model)
     postprocess_model(model, mode, quantizer)
+    dump_model_op_stats(mode, configs_mapping)
     return model
 
 
@@ -141,6 +149,7 @@ def gptq_entry(
     model.qconfig = configs_mapping
     model.save = MethodType(save, model)
     postprocess_model(model, mode, quantizer)
+    dump_model_op_stats(mode, configs_mapping)
 
     return model
 
@@ -361,6 +370,7 @@ def awq_quantize_entry(
     model.qconfig = configs_mapping
     model.save = MethodType(save, model)
     postprocess_model(model, mode, quantizer)
+    dump_model_op_stats(mode, configs_mapping)
     return model
 
 
@@ -415,6 +425,7 @@ def teq_quantize_entry(
     model.qconfig = configs_mapping
     model.save = MethodType(save, model)
     postprocess_model(model, mode, quantizer)
+    dump_model_op_stats(mode, configs_mapping)
 
     return model
 
@@ -491,6 +502,7 @@ def autoround_quantize_entry(
     model.qconfig = configs_mapping
     model.save = MethodType(save, model)
     postprocess_model(model, mode, quantizer)
+    dump_model_op_stats(mode, configs_mapping)
     return model
 
 
@@ -511,6 +523,7 @@ def hqq_entry(
     quantizer = get_quantizer(model, quantizer_cls=HQQuantizer, quant_config=configs_mapping)
     model = quantizer.execute(model, mode=mode)
     postprocess_model(model, mode, quantizer)
+    dump_model_op_stats(mode, configs_mapping)
 
     return model
 

neural_compressor/torch/utils/utility.py

@@ -16,6 +16,7 @@
 from typing import Callable, Dict, List, Tuple, Union
 
 import torch
+from prettytable import PrettyTable
 from typing_extensions import TypeAlias
 
 from neural_compressor.common.utils import LazyImport, Mode, logger
@@ -163,3 +164,101 @@ def postprocess_model(model, mode, quantizer):
     elif mode == Mode.CONVERT or mode == Mode.QUANTIZE:
         if getattr(model, "quantizer", False):
             del model.quantizer
+
+
+class Statistics:  # pragma: no cover
+    """The statistics printer."""
+
+    def __init__(self, data, header, field_names, output_handle=logger.info):
+        """Init a Statistics object.
+
+        Args:
+            data: The statistics data
+            header: The table header
+            field_names: The field names
+            output_handle: The output logging method
+        """
+        self.field_names = field_names
+        self.header = header
+        self.data = data
+        self.output_handle = output_handle
+        self.tb = PrettyTable(min_table_width=40)
+
+    def print_stat(self):
+        """Print the statistics."""
+        valid_field_names = []
+        for index, value in enumerate(self.field_names):
+            if index < 2:
+                valid_field_names.append(value)
+                continue
+
+            if any(i[index] for i in self.data):
+                valid_field_names.append(value)
+        self.tb.field_names = valid_field_names
+        for i in self.data:
+            tmp_data = []
+            for index, value in enumerate(i):
+                if self.field_names[index] in valid_field_names:
+                    tmp_data.append(value)
+            if any(tmp_data[1:]):
+                self.tb.add_row(tmp_data)
+        lines = self.tb.get_string().split("\n")
+        self.output_handle("|" + self.header.center(len(lines[0]) - 2, "*") + "|")
+        for i in lines:
+            self.output_handle(i)
+
+
+def dump_model_op_stats(mode, tune_cfg):
+    """This is a function to dump quantizable ops of model to user.
+
+    Args:
+        model (object): input model
+        tune_cfg (dict): quantization config
+    Returns:
+        None
+    """
+    if mode == Mode.PREPARE:
+        return
+    res = {}
+    # collect all dtype info and build empty results with existing op_type
+    dtype_set = set()
+    for op, config in tune_cfg.items():
+        op_type = op[1]
+        config = config.to_dict()
+        # import pdb; pdb.set_trace()
+        if not config["dtype"] == "fp32":
+            num_bits = config["bits"]
+            group_size = config["group_size"]
+            dtype_str = "A32W{}G{}".format(num_bits, group_size)
+            dtype_set.add(dtype_str)
+    dtype_set.add("FP32")
+    dtype_list = list(dtype_set)
+    dtype_list.sort()
+
+    for op, config in tune_cfg.items():
+        config = config.to_dict()
+        op_type = op[1]
+        if op_type not in res.keys():
+            res[op_type] = {dtype: 0 for dtype in dtype_list}
+
+    # fill in results with op_type and dtype
+    for op, config in tune_cfg.items():
+        config = config.to_dict()
+        if config["dtype"] == "fp32":
+            res[op_type]["FP32"] += 1
+        else:
+            num_bits = config["bits"]
+            group_size = config["group_size"]
+            dtype_str = "A32W{}G{}".format(num_bits, group_size)
+            res[op_type][dtype_str] += 1
+
+    # update stats format for dump.
+    field_names = ["Op Type", "Total"]
+    field_names.extend(dtype_list)
+    output_data = []
+    for op_type in res.keys():
+        field_results = [op_type, sum(res[op_type].values())]
+        field_results.extend([res[op_type][dtype] for dtype in dtype_list])
+        output_data.append(field_results)
+
+    Statistics(output_data, header="Mixed Precision Statistics", field_names=field_names).print_stat()

requirements_pt.txt

@@ -1,5 +1,6 @@
 numpy < 2.0
 peft==0.10.0
+prettytable
 psutil
 py-cpuinfo
 pydantic