sparsepixels 0.2.0__tar.gz → 0.2.1__tar.gz

{sparsepixels-0.2.0/sparsepixels.egg-info → sparsepixels-0.2.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sparsepixels
-Version: 0.2.0
+Version: 0.2.1
 Summary: Efficient convolution for sparse data on FPGAs
 Home-page: https://github.com/hftsoi/sparse-pixels
 Author: Ho Fung Tsoi
@@ -53,7 +53,6 @@ import keras
 from keras.layers import Flatten, Activation, ReLU
 from hgq.layers import QConv2D, QDense
 from hgq.config import QuantizerConfigScope, LayerConfigScope
-from hgq.quantizer.config import QuantizerConfig
 from sparsepixels.layers import InputReduce, QConv2DSparse, AveragePooling2DSparse
 ```
 
@@ -71,8 +70,7 @@ with (
     x, keep_mask = InputReduce(n_max_pixels=20, threshold=0.1, name='input_reduce')(x_in)
 
     # Sparse convolution
-    x = QConv2DSparse(filters=3, kernel_size=3, name='conv1', padding='same', strides=1,
-                      bq_conf=QuantizerConfig('default', 'bias'))([x, keep_mask])
+    x = QConv2DSparse(filters=3, kernel_size=3, name='conv1', padding='same', strides=1)([x, keep_mask])
     x = ReLU(name='relu1')(x)
 
     # Sparse pooling

{sparsepixels-0.2.0 → sparsepixels-0.2.1}/README.md

@@ -35,7 +35,6 @@ import keras
 from keras.layers import Flatten, Activation, ReLU
 from hgq.layers import QConv2D, QDense
 from hgq.config import QuantizerConfigScope, LayerConfigScope
-from hgq.quantizer.config import QuantizerConfig
 from sparsepixels.layers import InputReduce, QConv2DSparse, AveragePooling2DSparse
 ```
 
@@ -53,8 +52,7 @@ with (
     x, keep_mask = InputReduce(n_max_pixels=20, threshold=0.1, name='input_reduce')(x_in)
 
     # Sparse convolution
-    x = QConv2DSparse(filters=3, kernel_size=3, name='conv1', padding='same', strides=1,
-                      bq_conf=QuantizerConfig('default', 'bias'))([x, keep_mask])
+    x = QConv2DSparse(filters=3, kernel_size=3, name='conv1', padding='same', strides=1)([x, keep_mask])
     x = ReLU(name='relu1')(x)
 
     # Sparse pooling

sparsepixels-0.2.1/pyproject.toml

@@ -0,0 +1,24 @@
+[build-system]
+requires = ["setuptools"]
+build-backend = "setuptools.build_meta"
+
+[tool.ruff]
+target-version = "py310"
+line-length = 120
+
+[tool.ruff.lint]
+select = [
+    "E",
+    "F",
+    "I",
+    "W",
+    "UP",
+]
+ignore = [
+    "E402",
+    "F403",
+    "F405",
+]
+
+[tool.ruff.lint.isort]
+known-first-party = ["sparsepixels"]

{sparsepixels-0.2.0 → sparsepixels-0.2.1}/setup.cfg

@@ -1,6 +1,6 @@
 [metadata]
 name = sparsepixels
-version = 0.2.0
+version = 0.2.1
 description = Efficient convolution for sparse data on FPGAs
 author = Ho Fung Tsoi
 author_email = ho.fung.tsoi@cern.ch

{sparsepixels-0.2.0 → sparsepixels-0.2.1}/sparsepixels/layers.py

@@ -1,9 +1,9 @@
 import keras
-from keras import ops
-from keras.layers import AveragePooling2D, MaxPooling2D
 from hgq.layers import QConv2D
 from hgq.quantizer import Quantizer
 from hgq.quantizer.config import QuantizerConfig
+from keras import ops
+from keras.layers import AveragePooling2D, MaxPooling2D
 
 
 class InputReduce(keras.layers.Layer):
@@ -37,10 +37,12 @@ class InputReduce(keras.layers.Layer):
 
     def get_config(self):
         config = super().get_config()
-        config.update({
-            "n_max_pixels": self.n_max_pixels,
-            "threshold": self.threshold,
-        })
+        config.update(
+            {
+                "n_max_pixels": self.n_max_pixels,
+                "threshold": self.threshold,
+            }
+        )
         return config
 
 
@@ -60,7 +62,8 @@ class RemoveDilatedPixels(keras.layers.Layer):
 class QConv2DSparse(keras.layers.Layer):
     def __init__(self, *conv_args, **conv_kwargs):
         super().__init__(name=conv_kwargs.get("name", None))
-        self._bq_conf = conv_kwargs.pop("bq_conf", None)
+        self._use_bias = conv_kwargs.pop("use_bias", True)
+        self._bq_conf = conv_kwargs.pop("bq_conf", None) or QuantizerConfig("default", "bias")
 
         conv_kwargs["use_bias"] = False
         conv_kwargs.setdefault("enable_iq", False)
@@ -68,17 +71,15 @@ class QConv2DSparse(keras.layers.Layer):
         self.masker = RemoveDilatedPixels()
 
     def build(self, input_shape):
-        self.sparse_bias = self.add_weight(
-            name="sparse_bias",
-            shape=(self.conv.filters,),
-            initializer="zeros",
-            trainable=True,
-        )
-        if self._bq_conf is not None:
+        if self._use_bias:
+            self.sparse_bias = self.add_weight(
+                name="sparse_bias",
+                shape=(self.conv.filters,),
+                initializer="zeros",
+                trainable=True,
+            )
             self._bq = Quantizer(self._bq_conf, name=f"{self.name}_bq")
             self._bq.build((self.conv.filters,))
-        else:
-            self._bq = None
         super().build(input_shape)
 
     def call(self, inputs, **kwargs):
@@ -86,13 +87,11 @@ class QConv2DSparse(keras.layers.Layer):
         x = self.masker((x, keep_mask))
         y = self.conv(x, **kwargs)
 
-        b = self.sparse_bias
-        if self._bq is not None:
-            b = self._bq(b)
-        b = ops.reshape(b, (1, 1, 1, -1))
-
-        non_zero = ops.cast(y != 0, y.dtype)
-        y = y + b * non_zero
+        if self._use_bias:
+            b = self._bq(self.sparse_bias)
+            b = ops.reshape(b, (1, 1, 1, -1))
+            non_zero = ops.cast(y != 0, y.dtype)
+            y = y + b * non_zero
 
         y = self.masker((y, keep_mask))
         return y
@@ -100,14 +99,16 @@ class QConv2DSparse(keras.layers.Layer):
     def get_config(self):
         cfg = super().get_config()
         cfg["conv_config"] = self.conv.get_config()
+        cfg["use_bias"] = self._use_bias
         cfg["bq_conf"] = self._bq_conf
         return cfg
 
     @classmethod
     def from_config(cls, config):
         conv_cfg = config.pop("conv_config")
+        use_bias = config.pop("use_bias", True)
         bq_conf = config.pop("bq_conf", None)
-        return cls(**conv_cfg, bq_conf=bq_conf)
+        return cls(**conv_cfg, use_bias=use_bias, bq_conf=bq_conf)
 
 
 class AveragePooling2DSparse(keras.layers.Layer):

{sparsepixels-0.2.0 → sparsepixels-0.2.1/sparsepixels.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sparsepixels
-Version: 0.2.0
+Version: 0.2.1
 Summary: Efficient convolution for sparse data on FPGAs
 Home-page: https://github.com/hftsoi/sparse-pixels
 Author: Ho Fung Tsoi
@@ -53,7 +53,6 @@ import keras
 from keras.layers import Flatten, Activation, ReLU
 from hgq.layers import QConv2D, QDense
 from hgq.config import QuantizerConfigScope, LayerConfigScope
-from hgq.quantizer.config import QuantizerConfig
 from sparsepixels.layers import InputReduce, QConv2DSparse, AveragePooling2DSparse
 ```
 
@@ -71,8 +70,7 @@ with (
     x, keep_mask = InputReduce(n_max_pixels=20, threshold=0.1, name='input_reduce')(x_in)
 
     # Sparse convolution
-    x = QConv2DSparse(filters=3, kernel_size=3, name='conv1', padding='same', strides=1,
-                      bq_conf=QuantizerConfig('default', 'bias'))([x, keep_mask])
+    x = QConv2DSparse(filters=3, kernel_size=3, name='conv1', padding='same', strides=1)([x, keep_mask])
     x = ReLU(name='relu1')(x)
 
     # Sparse pooling

{sparsepixels-0.2.0 → sparsepixels-0.2.1}/tests/test_model.py

@@ -2,7 +2,6 @@ import keras
 from keras.layers import Flatten, Activation, AveragePooling2D, ReLU
 from hgq.layers import QConv2D, QDense
 from hgq.config import QuantizerConfigScope, LayerConfigScope
-from hgq.quantizer.config import QuantizerConfig
 from sparsepixels.layers import InputReduce, QConv2DSparse, AveragePooling2DSparse
 
 
@@ -19,13 +18,11 @@ def build_cnn(is_sparse, n_max_pixels=None):
             x = x_in
 
         if is_sparse:
-            x = QConv2DSparse(filters=1, kernel_size=7, name='conv1', padding='same', strides=1,
-                              bq_conf=QuantizerConfig('default', 'bias'))([x, keep_mask])
+            x = QConv2DSparse(filters=1, kernel_size=7, name='conv1', padding='same', strides=1)([x, keep_mask])
             x = ReLU(name='relu1')(x)
             x, keep_mask = AveragePooling2DSparse(4, name='pool1')([x, keep_mask])
 
-            x = QConv2DSparse(filters=3, kernel_size=5, name='conv2', padding='same', strides=1,
-                              bq_conf=QuantizerConfig('default', 'bias'))([x, keep_mask])
+            x = QConv2DSparse(filters=3, kernel_size=5, name='conv2', padding='same', strides=1)([x, keep_mask])
             x = ReLU(name='relu2')(x)
             x, keep_mask = AveragePooling2DSparse(2, name='pool2')([x, keep_mask])
         else:

sparsepixels-0.2.0/pyproject.toml

@@ -1,3 +0,0 @@
-[build-system]
-requires = ["setuptools"]
-build-backend = "setuptools.build_meta"