refactor(mge/quantization): add `narrow_range` to control quant dtype's lower bound

GitOrigin-RevId: 92389341be

python_module/megengine/module/qat/module.py

@@ -37,15 +37,14 @@ class QATModule(Module):
         Set quantization related configs with ``qconfig``, including
         observer and fake_quant for weight and activation.
         """
-        self.weight_observer = qconfig.weight_observer()
-        self.act_observer = qconfig.act_observer()
 
-        if qconfig.fake_quant is None:
-            self.weight_fake_quant = None
-            self.act_fake_quant = None
-        else:
-            self.weight_fake_quant = qconfig.fake_quant(self.weight_observer.dtype)
-            self.act_fake_quant = qconfig.fake_quant(self.act_observer.dtype)
+        def safe_call(func):
+            return func() if func is not None else None
+
+        self.weight_observer = safe_call(qconfig.weight_observer)
+        self.act_observer = safe_call(qconfig.act_observer)
+        self.weight_fake_quant = safe_call(qconfig.weight_fake_quant)
+        self.act_fake_quant = safe_call(qconfig.act_fake_quant)
 
     def _apply_fakequant_with_observer(
         self, target: Tensor, fake_quant: FakeQuantize, observer: Observer

python_module/megengine/quantization/fake_quant.py

@@ -19,7 +19,7 @@ from .observer import ObserverMode, Round
 
 
 class _FakeQuantize(Module):
-    def __init__(self, dtype: str, enable: bool = True):
+    def __init__(self, dtype: str, narrow_range: bool = False, enable: bool = True):
         super().__init__()
         if not dtype in _metadata_dict.keys():
             raise ValueError(
@@ -28,7 +28,10 @@ class _FakeQuantize(Module):
                 )
             )
         self.dtype = dtype
-        self.qmin = _metadata_dict[dtype].qmin
+        self.narrow_range = narrow_range
+        self.qmin = (
+            -_metadata_dict[dtype].qmax if narrow_range else _metadata_dict[dtype].qmin
+        )
         self.qmax = _metadata_dict[dtype].qmax
         self.enabled = enable
 
@@ -90,12 +93,12 @@ class TQT_Function(Function):
 
 class TQT(_FakeQuantize):
     """
-    TQT: https://arxiv.org/abs/1903.08066 Trained Quantization Thresholds 
+    TQT: https://arxiv.org/abs/1903.08066 Trained Quantization Thresholds
     for Accurate and Efficient Fixed-Point Inference of Deep Neural Networks
     """
 
-    def __init__(self, dtype: str, enable: bool = True):
-        super().__init__(dtype, enable)
+    def __init__(self, dtype: str, narrow_range: bool = False, enable: bool = True):
+        super().__init__(dtype, narrow_range, enable)
         self.scale = Parameter(0.0, dtype=np.float32)
 
     def fake_quant_forward(self, inp, q_dict):
@@ -116,6 +119,11 @@ class TQT(_FakeQuantize):
 class FakeQuantize(_FakeQuantize):
     r"""
     A module to do quant and dequant according to observer's scale and zero_point.
+
+    :param dtype: A string indicating the target quantization type of input.
+    :param narrow_range: Whether the absolute value of ``qmin`` is the same as ``qmax``,
+        instead of 1 greater. Usually True for weight and False for activation.
+    :param enable: Whether do ``normal_forward`` or ``fake_quant_forward``.
     """
 
     def fake_quant_forward(self, inp, q_dict):

python_module/megengine/quantization/observer.py

@@ -31,9 +31,11 @@ class Observer(Module):
     A base class for Observer Module.
 
     :param dtype: a string indicating to collect scale and zero_point of which dtype
+    :param narrow_range: Whether the absolute value of ``qmin`` is the same as ``qmax``,
+        instead of 1 greater. Usually True for weight and False for activation.
     """
 
-    def __init__(self, dtype="qint8"):
+    def __init__(self, dtype: str, narrow_range: bool = False):
         super().__init__()
         if dtype not in _metadata_dict.keys():
             raise ValueError(
@@ -42,7 +44,10 @@ class Observer(Module):
                 )
             )
         self.dtype = dtype
-        self.qmin = _metadata_dict[dtype].qmin
+        self.narrow_range = narrow_range
+        self.qmin = (
+            -_metadata_dict[dtype].qmax if narrow_range else _metadata_dict[dtype].qmin
+        )
         self.qmax = _metadata_dict[dtype].qmax
         self.enabled = True
 
@@ -96,8 +101,14 @@ def create_observer_dict(mode):
 
 
 class MinMaxObserver(Observer):
-    def __init__(self, mode=ObserverMode.SYMMERTIC, eps=0.00001, dtype="qint8"):
-        super().__init__(dtype)
+    def __init__(
+        self,
+        mode=ObserverMode.SYMMERTIC,
+        eps=0.00001,
+        dtype="qint8",
+        narrow_range: bool = False,
+    ):
+        super().__init__(dtype, narrow_range)
         self.mode = mode
         self.min_val = Buffer(np.finfo(np.float32).max, dtype=np.float32)
         self.max_val = Buffer(np.finfo(np.float32).min, dtype=np.float32)
@@ -153,9 +164,14 @@ class MinMaxObserver(Observer):
 
 class ExponentialMovingAverageObserver(MinMaxObserver):
     def __init__(
-        self, momentum=0.9, mode=ObserverMode.SYMMERTIC, eps=0.00001, dtype="qint8"
+        self,
+        momentum=0.9,
+        mode=ObserverMode.SYMMERTIC,
+        eps=0.00001,
+        dtype="qint8",
+        narrow_range: bool = False,
     ):
-        super().__init__(mode, eps, dtype)
+        super().__init__(mode, eps, dtype, narrow_range)
         self.momentum = Buffer(momentum)
         self.runtime_momentum = Buffer(0.0)
 
@@ -188,11 +204,12 @@ class HistogramObserver(MinMaxObserver):
         self,
         bins=2048,
         upsample_rate=128,
-        dtype="qint8",
         mode=ObserverMode.SYMMERTIC,
         eps=0.00001,
+        dtype="qint8",
+        narrow_range: bool = False,
     ):
-        super().__init__(mode, eps, dtype)
+        super().__init__(mode, eps, dtype, narrow_range)
         self.bins = bins
         self.upsample_rate = upsample_rate
         self.dst_nbins = _metadata_dict[dtype].qmax - _metadata_dict[dtype].qmin + 1

python_module/megengine/quantization/qconfig.py

@@ -5,6 +5,8 @@
 # Unless required by applicable law or agreed to in writing,
 # software distributed under the License is distributed on an
 # "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+from functools import partial
+
 from ..module import Module
 from .fake_quant import TQT, FakeQuantize
 from .observer import (
@@ -22,9 +24,9 @@ class QConfig:
     :param weight_observer: interface to instantiate an :class:`~.Observer` indicating
         how to collect scales and zero_point of wegiht.
     :param act_observer: similar to ``weight_observer`` but toward activation.
-    :param fake_quant: interface to instantiate a :class:`~.FakeQuantize` indicating
-        how to do fake_quant calculation. can be invoked multi times to get different
-        instance for each target tensor, for better control on enable and disable.
+    :param weight_fake_quant: interface to instantiate a :class:`~.FakeQuantize` indicating
+        how to do fake_quant calculation.
+    :param act_observer: similar to ``weight_fake_quant`` but toward activation.
 
     Examples:
 
@@ -32,14 +34,24 @@ class QConfig:
 
         # Default EMA QConfig for QAT.
         ema_fakequant_qconfig = QConfig(
-            weight_observer=MinMaxObserver,
-            act_observer=ExponentialMovingAverageObserver,
-            fake_quant=FakeQuantize,
+            weight_observer=partial(MinMaxObserver, dtype="qint8", narrow_range=True),
+            act_observer=partial(ExponentialMovingAverageObserver, dtype="qint8", narrow_range=False),
+            weight_fake_quant=partial(FakeQuantize, dtype="qint8", narrow_range=True),
+            act_fake_quant=partial(FakeQuantize, dtype="qint8", narrow_range=False),
         )
+
+    Each parameter is a ``class`` rather than an instance. And we recommand using ``functools.partial``
+    to add initialization parameters of the ``class``, so that don't need to provide parameters in
+    :meth:`~.QATModule.set_qconfig`.
+
+    Usually we set ``narrow_range`` of weight related paramters to ``True`` and of activation related
+    parameters to ``False``. For the result of multiplication and addition as ``a * b + c * d``, if
+    four variables are all -128 of dtype ``qint8``, then the result will be ``2^15`` and cause overflow.
+    Weights are commonly calculated in this way, so needed to narrow the range.
     """
 
     def __init__(
-        self, act_observer, weight_observer, fake_quant,
+        self, weight_observer, act_observer, weight_fake_quant, act_fake_quant
     ):
         if isinstance(act_observer, Module) or isinstance(weight_observer, Module):
             raise ValueError(
@@ -47,30 +59,42 @@ class QConfig:
                 " class generator using `partial(Observer, ...)` instead. Use"
                 " partial(MyObserver, x=1) to override arguments to constructor if needed"
             )
-        self.act_observer = act_observer
         self.weight_observer = weight_observer
-        self.fake_quant = fake_quant
+        self.act_observer = act_observer
+        self.weight_fake_quant = weight_fake_quant
+        self.act_fake_quant = act_fake_quant
 
 
 tqt_quant_qconfig = QConfig(
-    weight_observer=ExponentialMovingAverageObserver,
-    act_observer=ExponentialMovingAverageObserver,
-    fake_quant=TQT,
+    weight_observer=partial(
+        ExponentialMovingAverageObserver, dtype="qint8", narrow_range=True
+    ),
+    act_observer=partial(
+        ExponentialMovingAverageObserver, dtype="qint8", narrow_range=False
+    ),
+    weight_fake_quant=partial(TQT, dtype="qint8", narrow_range=True),
+    act_fake_quant=partial(TQT, dtype="qint8", narrow_range=False),
 )
 
-# Default QAT QConfigs
 min_max_fakequant_qconfig = QConfig(
-    weight_observer=MinMaxObserver,
-    act_observer=MinMaxObserver,
-    fake_quant=FakeQuantize,
+    weight_observer=partial(MinMaxObserver, dtype="qint8", narrow_range=True),
+    act_observer=partial(MinMaxObserver, dtype="qint8", narrow_range=False),
+    weight_fake_quant=partial(FakeQuantize, dtype="qint8", narrow_range=True),
+    act_fake_quant=partial(FakeQuantize, dtype="qint8", narrow_range=False),
 )
 
 ema_fakequant_qconfig = QConfig(
-    weight_observer=MinMaxObserver,
-    act_observer=ExponentialMovingAverageObserver,
-    fake_quant=FakeQuantize,
+    weight_observer=partial(MinMaxObserver, dtype="qint8", narrow_range=True),
+    act_observer=partial(
+        ExponentialMovingAverageObserver, dtype="qint8", narrow_range=False
+    ),
+    weight_fake_quant=partial(FakeQuantize, dtype="qint8", narrow_range=True),
+    act_fake_quant=partial(FakeQuantize, dtype="qint8", narrow_range=False),
 )
 
 calibration_qconfig = QConfig(
-    weight_observer=MinMaxObserver, act_observer=HistogramObserver, fake_quant=None,
+    weight_observer=partial(MinMaxObserver, dtype="qint8", narrow_range=True),
+    act_observer=partial(HistogramObserver, dtype="qint8", narrow_range=False),
+    weight_fake_quant=None,
+    act_fake_quant=None,
 )