卷积层通道剪裁 ================ Pruner ---------- .. py:class:: paddleslim.prune.Pruner(criterion="l1_norm") `源代码 `_ 对卷积网络的通道进行一次剪裁。剪裁一个卷积层的通道,是指剪裁该卷积层输出的通道。卷积层的权重形状为 ``[output_channel, input_channel, kernel_size, kernel_size]`` ,通过剪裁该权重的第一纬度达到剪裁输出通道数的目的。 **参数:** - **criterion** - 评估一个卷积层内通道重要性所参考的指标。目前支持 ``l1_norm`` 和 ``batch_norm_scale`` 。默认为 ``l1_norm`` 。若该参数设为 ``batch_norm_scale`` , 则表示剪枝算法将根据卷积层后连接的BatchNorm层的Scale参数的绝对值大小作为评估卷积层内通道重要性所参考的指标。 在初始化Pruner()类实例时,若没有传入该参数,则表示Pruner()使用criterion默认参数值 ``l1_norm`` ;可以显示地传入criterion的值以改变剪枝算法的剪枝策略。 **返回:** 一个Pruner类的实例 **示例代码:** .. code-block:: python from paddleslim.prune import Pruner pruner = Pruner() .. .. py:method:: paddleslim.prune.Pruner.prune(program, scope, params, ratios, place=None, lazy=False, only_graph=False, param_backup=False, param_shape_backup=False) 对目标网络的一组卷积层的权重进行裁剪。 **参数:** - **program(paddle.fluid.Program)** - 要裁剪的目标网络。更多关于Program的介绍请参考:`Program概念介绍 `_。 - **scope(paddle.fluid.Scope)** - 要裁剪的权重所在的 ``scope`` ,Paddle中用 ``scope`` 实例存放模型参数和运行时变量的值。Scope中的参数值会被 ``inplace`` 的裁剪。更多介绍请参考: `Scope概念介绍 <>`_ - **params(list)** - 需要被裁剪的卷积层的参数的名称列表。可以通过以下方式查看模型中所有参数的名称: .. code-block:: python for block in program.blocks: for param in block.all_parameters(): print("param: {}; shape: {}".format(param.name, param.shape)) - **ratios(list)** - 用于裁剪 ``params`` 的剪切率,类型为列表。该列表长度必须与 ``params`` 的长度一致。 - **place(paddle.fluid.Place)** - 待裁剪参数所在的设备位置,可以是 ``CUDAPlace`` 或 ``CPUPlace`` 。[Place概念介绍]() - **lazy(bool)** - ``lazy`` 为True时,通过将指定通道的参数置零达到裁剪的目的,参数的 ``shape保持不变`` ; ``lazy`` 为False时,直接将要裁的通道的参数删除,参数的 ``shape`` 会发生变化。 - **only_graph(bool)** - 是否只裁剪网络结构。在Paddle中,Program定义了网络结构,Scope存储参数的数值。一个Scope实例可以被多个Program使用,比如定义了训练网络的Program和定义了测试网络的Program是使用同一个Scope实例的。 ``only_graph`` 为True时,只对Program中定义的卷积的通道进行剪裁; ``only_graph`` 为false时,Scope中卷积参数的数值也会被剪裁。默认为False。 - **param_backup(bool)** - 是否返回对参数值的备份。默认为False。 - **param_shape_backup(bool)** - 是否返回对参数 ``shape`` 的备份。默认为False。 **返回:** - **pruned_program(paddle.fluid.Program)** - 被裁剪后的Program。 - **param_backup(dict)** - 对参数数值的备份,用于恢复Scope中的参数数值。 - **param_shape_backup(dict)** - 对参数形状的备份。 **示例:** 点击 `AIStudio `_ 执行以下示例代码。 .. code-block:: python import paddle.fluid as fluid from paddle.fluid.param_attr import ParamAttr from paddleslim.prune import Pruner def conv_bn_layer(input, num_filters, filter_size, name, stride=1, groups=1, act=None): conv = fluid.layers.conv2d( input=input, num_filters=num_filters, filter_size=filter_size, stride=stride, padding=(filter_size - 1) // 2, groups=groups, act=None, param_attr=ParamAttr(name=name + "_weights"), bias_attr=False, name=name + "_out") bn_name = name + "_bn" return fluid.layers.batch_norm( input=conv, act=act, name=bn_name + '_output', param_attr=ParamAttr(name=bn_name + '_scale'), bias_attr=ParamAttr(bn_name + '_offset'), moving_mean_name=bn_name + '_mean', moving_variance_name=bn_name + '_variance', ) 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") place = fluid.CPUPlace() exe = fluid.Executor(place) scope = fluid.Scope() exe.run(startup_program, scope=scope) pruner = Pruner() main_program, _, _ = pruner.prune( main_program, scope, params=["conv4_weights"], ratios=[0.5], place=place, lazy=False, only_graph=False, param_backup=False, param_shape_backup=False) for param in main_program.global_block().all_parameters(): if "weights" in param.name: print("param name: {}; param shape: {}".format(param.name, param.shape)) sensitivity -------------- .. py:function:: paddleslim.prune.sensitivity(program, place, param_names, eval_func, sensitivities_file=None, pruned_ratios=None) `源代码 `_ 计算网络中每个卷积层的敏感度。每个卷积层的敏感度信息统计方法为:依次剪掉当前卷积层不同比例的输出通道数,在测试集上计算剪裁后的精度损失。得到敏感度信息后,可以通过观察或其它方式确定每层卷积的剪裁率。 **参数:** - **program(paddle.fluid.Program)** - 待评估的目标网络。更多关于Program的介绍请参考:`Program概念介绍 `_。 - **place(paddle.fluid.Place)** - 待分析的参数所在的设备位置,可以是 ``CUDAPlace`` 或 ``CPUPlace`` 。[Place概念介绍]() - **param_names(list)** - 待分析的卷积层的参数的名称列表。可以通过以下方式查看模型中所有参数的名称: .. code-block:: python for block in program.blocks: for param in block.all_parameters(): print("param: {}; shape: {}".format(param.name, param.shape)) - **eval_func(function)** - 用于评估裁剪后模型效果的回调函数。该回调函数接受被裁剪后的 ``program`` 为参数,返回一个表示当前program的精度,用以计算当前裁剪带来的精度损失。 - **sensitivities_file(str)** - 保存敏感度信息的本地文件系统的文件。在敏感度计算过程中,会持续将新计算出的敏感度信息追加到该文件中。重启任务后,文件中已有敏感度信息不会被重复计算。该文件可以用 ``pickle`` 加载。 - **pruned_ratios(list)** - 计算卷积层敏感度信息时,依次剪掉的通道数比例。默认为 ``[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9]`` 。 **返回:** - **sensitivities(dict)** - 存放敏感度信息的dict,其格式为: .. code-block:: python {"weight_0": {0.1: 0.22, 0.2: 0.33 }, "weight_1": {0.1: 0.21, 0.2: 0.4 } } 其中, ``weight_0`` 是卷积层参数的名称, ``sensitivities['weight_0']`` 的 ``value`` 为剪裁比例, ``value`` 为精度损失的比例。 **示例:** 点击 `AIStudio `_ 运行以下示例代码。 .. code-block:: python import paddle import numpy as np import paddle.fluid as fluid from paddle.fluid.param_attr import ParamAttr from paddleslim.prune import sensitivity import paddle.dataset.mnist as reader def conv_bn_layer(input, num_filters, filter_size, name, stride=1, groups=1, act=None): conv = fluid.layers.conv2d( input=input, num_filters=num_filters, filter_size=filter_size, stride=stride, padding=(filter_size - 1) // 2, groups=groups, act=None, param_attr=ParamAttr(name=name + "_weights"), bias_attr=False, name=name + "_out") bn_name = name + "_bn" return fluid.layers.batch_norm( input=conv, act=act, name=bn_name + '_output', param_attr=ParamAttr(name=bn_name + '_scale'), bias_attr=ParamAttr(bn_name + '_offset'), moving_mean_name=bn_name + '_mean', moving_variance_name=bn_name + '_variance', ) 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 image_shape = [1,28,28] with fluid.program_guard(main_program, startup_program): image = fluid.data(name='image', shape=[None]+image_shape, dtype='float32') label = fluid.data(name='label', shape=[None, 1], dtype='int64') conv1 = conv_bn_layer(image, 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") out = fluid.layers.fc(conv6, size=10, act="softmax") # cost = fluid.layers.cross_entropy(input=out, label=label) # avg_cost = fluid.layers.mean(x=cost) acc_top1 = fluid.layers.accuracy(input=out, label=label, k=1) # acc_top5 = fluid.layers.accuracy(input=out, label=label, k=5) place = fluid.CPUPlace() exe = fluid.Executor(place) exe.run(startup_program) val_reader = paddle.batch(reader.test(), batch_size=128) val_feeder = feeder = fluid.DataFeeder( [image, label], place, program=main_program) def eval_func(program): acc_top1_ns = [] for data in val_reader(): acc_top1_n = exe.run(program, feed=val_feeder.feed(data), fetch_list=[acc_top1.name]) acc_top1_ns.append(np.mean(acc_top1_n)) return np.mean(acc_top1_ns) param_names = [] for param in main_program.global_block().all_parameters(): if "weights" in param.name: param_names.append(param.name) sensitivities = sensitivity(main_program, place, param_names, eval_func, sensitivities_file="./sensitive.data", pruned_ratios=[0.1, 0.2, 0.3]) print(sensitivities) merge_sensitive ---------------- .. py:function:: paddleslim.prune.merge_sensitive(sensitivities) `源代码 `_ 合并多个敏感度信息。 参数: - **sensitivities(list | list)** - 待合并的敏感度信息,可以是字典的列表,或者是存放敏感度信息的文件的路径列表。 返回: - **sensitivities(dict)** - 合并后的敏感度信息。其格式为: .. code-block:: bash {"weight_0": {0.1: 0.22, 0.2: 0.33 }, "weight_1": {0.1: 0.21, 0.2: 0.4 } } 其中, ``weight_0`` 是卷积层参数的名称, ``sensitivities['weight_0']`` 的 ``value`` 为剪裁比例, ``value`` 为精度损失的比例。 示例: .. code-block:: python from paddleslim.prune import merge_sensitive sen0 = {"weight_0": {0.1: 0.22, 0.2: 0.33 }, "weight_1": {0.1: 0.21, 0.2: 0.4 } } sen1 = {"weight_0": {0.3: 0.41, }, "weight_2": {0.1: 0.10, 0.2: 0.35 } } sensitivities = merge_sensitive([sen0, sen1]) print(sensitivities) load_sensitivities --------------------- .. py:function:: paddleslim.prune.load_sensitivities(sensitivities_file) `源代码 `_ 从文件中加载敏感度信息。 参数: - **sensitivities_file(str)** - 存放敏感度信息的本地文件. 返回: - **sensitivities(dict)** - 敏感度信息。 示例: .. code-block:: python import pickle from paddleslim.prune import load_sensitivities sen = {"weight_0": {0.1: 0.22, 0.2: 0.33 }, "weight_1": {0.1: 0.21, 0.2: 0.4 } } sensitivities_file = "sensitive_api_demo.data" with open(sensitivities_file, 'wb') as f: pickle.dump(sen, f) sensitivities = load_sensitivities(sensitivities_file) print(sensitivities) get_ratios_by_loss ------------------- .. py:function:: paddleslim.prune.get_ratios_by_loss(sensitivities, loss) `源代码 `_ 根据敏感度和精度损失阈值计算出一组剪切率。对于参数 ``w`` , 其剪裁率为使精度损失低于 ``loss`` 的最大剪裁率。 **参数:** - **sensitivities(dict)** - 敏感度信息。 - **loss** - 精度损失阈值。 **返回:** - **ratios(dict)** - 一组剪切率。 ``key`` 是待剪裁参数的名称。 ``value`` 是对应参数的剪裁率。 **示例:** .. code-block:: python from paddleslim.prune import get_ratios_by_loss sen = {"weight_0": {0.1: 0.22, 0.2: 0.33 }, "weight_1": {0.1: 0.21, 0.2: 0.4 } } ratios = get_ratios_by_loss(sen, 0.3) print(ratios)