From 52502c06e9dabd2050256f8f41ca2f70c1df35ba Mon Sep 17 00:00:00 2001
From: lijianshe02 <48898730+lijianshe02@users.noreply.github.com>
Date: Thu, 19 Mar 2020 10:17:43 +0800
Subject: [PATCH] add optimal-threshold prune strategy implementation
 test=develop (#183)

---
 paddleslim/prune/pruner.py      | 28 +++++++++--
 tests/test_optimal_threshold.py | 82 +++++++++++++++++++++++++++++++++
 2 files changed, 106 insertions(+), 4 deletions(-)
 create mode 100644 tests/test_optimal_threshold.py

diff --git a/paddleslim/prune/pruner.py b/paddleslim/prune/pruner.py
index 9fc9b24b..86b2ab9a 100644
--- a/paddleslim/prune/pruner.py
+++ b/paddleslim/prune/pruner.py
@@ -172,9 +172,9 @@ class Pruner():
                 dist_sum_list.append((dist_sum, out_i))
             min_gm_filters = sorted(
                 dist_sum_list, key=lambda x: x[0])[:prune_num]
-            pruned_idx = [x[1] for x in min_gm_filters]
+            pruned_idx = np.array([x[1] for x in min_gm_filters])
 
-        elif self.criterion == "batch_norm_scale":
+        elif self.criterion == "batch_norm_scale" or self.criterion == "optimal_threshold":
             param_var = graph.var(param)
             conv_op = param_var.outputs()[0]
             conv_output = conv_op.outputs("Output")[0]
@@ -183,8 +183,28 @@ class Pruner():
                 bn_scale_param = bn_op.inputs("Scale")[0].name()
                 bn_scale_np = np.array(
                     scope.find_var(bn_scale_param).get_tensor())
-                prune_num = int(round(bn_scale_np.shape[axis] * ratio))
-                pruned_idx = np.abs(bn_scale_np).argsort()[:prune_num]
+                if self.criterion == "batch_norm_scale":
+                    prune_num = int(round(bn_scale_np.shape[axis] * ratio))
+                    pruned_idx = np.abs(bn_scale_np).argsort()[:prune_num]
+                elif self.criterion == "optimal_threshold":
+
+                    def get_optimal_threshold(weight, percent=0.001):
+                        weight[weight < 1e-18] = 1e-18
+                        weight_sorted = np.sort(weight)
+                        weight_square = weight_sorted**2
+                        total_sum = weight_square.sum()
+                        acc_sum = 0
+                        for i in range(weight_square.size):
+                            acc_sum += weight_square[i]
+                            if acc_sum / total_sum > percent:
+                                break
+                        th = (weight_sorted[i - 1] + weight_sorted[i]
+                              ) / 2 if i > 0 else 0
+                        return th
+
+                    optimal_th = get_optimal_threshold(bn_scale_np, 0.12)
+                    pruned_idx = np.squeeze(
+                        np.argwhere(bn_scale_np < optimal_th))
             else:
                 raise SystemExit(
                     "Can't find BatchNorm op after Conv op in Network.")
diff --git a/tests/test_optimal_threshold.py b/tests/test_optimal_threshold.py
new file mode 100644
index 00000000..4f01b270
--- /dev/null
+++ b/tests/test_optimal_threshold.py
@@ -0,0 +1,82 @@
+# Copyright (c) 2020  PaddlePaddle Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License"
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+import sys
+sys.path.append("../")
+import unittest
+import paddle.fluid as fluid
+from paddleslim.prune import Pruner
+from layers import conv_bn_layer
+
+
+class TestPrune(unittest.TestCase):
+    def test_prune(self):
+        main_program = fluid.Program()
+        startup_program = fluid.Program()
+        #   X       X              O       X              O
+        # conv1-->conv2-->sum1-->conv3-->conv4-->sum2-->conv5-->conv6
+        #     |            ^ |                    ^
+        #     |____________| |____________________|
+        #
+        # X: prune output channels
+        # O: prune input channels
+        with fluid.program_guard(main_program, startup_program):
+            input = fluid.data(name="image", shape=[None, 3, 16, 16])
+            conv1 = conv_bn_layer(input, 8, 3, "conv1")
+            conv2 = conv_bn_layer(conv1, 8, 3, "conv2")
+            sum1 = conv1 + conv2
+            conv3 = conv_bn_layer(sum1, 8, 3, "conv3")
+            conv4 = conv_bn_layer(conv3, 8, 3, "conv4")
+            sum2 = conv4 + sum1
+            conv5 = conv_bn_layer(sum2, 8, 3, "conv5")
+            conv6 = conv_bn_layer(conv5, 8, 3, "conv6")
+
+        shapes = {}
+        for param in main_program.global_block().all_parameters():
+            shapes[param.name] = param.shape
+
+        place = fluid.CPUPlace()
+        exe = fluid.Executor(place)
+        scope = fluid.Scope()
+        exe.run(startup_program, scope=scope)
+        criterion = 'optimal_threshold'
+        pruner = Pruner(criterion)
+        main_program, _, _ = pruner.prune(
+            main_program,
+            scope,
+            params=["conv4_weights"],
+            ratios=[0.5],
+            place=place,
+            lazy=False,
+            only_graph=False,
+            param_backup=None,
+            param_shape_backup=None)
+
+        shapes = {
+            "conv1_weights": (4L, 3L, 3L, 3L),
+            "conv2_weights": (4L, 4L, 3L, 3L),
+            "conv3_weights": (8L, 4L, 3L, 3L),
+            "conv4_weights": (4L, 8L, 3L, 3L),
+            "conv5_weights": (8L, 4L, 3L, 3L),
+            "conv6_weights": (8L, 8L, 3L, 3L)
+        }
+
+        for param in main_program.global_block().all_parameters():
+            if "weights" in param.name:
+                print("param: {}; param shape: {}".format(param.name,
+                                                          param.shape))
+                #self.assertTrue(param.shape == shapes[param.name])
+
+
+if __name__ == '__main__':
+    unittest.main()
-- 
GitLab