知识蒸馏

merge#

paddleslim.dist.merge(teacher_program, student_program, data_name_map, place, scope=fluid.global_scope(), name_prefix='teacher_') [源代码]: merge将teacher_program融合到student_program中。在融合的program中，可以为其中合适的teacher特征图和student特征图添加蒸馏损失函数，从而达到用teacher模型的暗知识（Dark Knowledge）指导student模型学习的目的。

参数：

teacher_program(Program)-定义了teacher模型的 paddle program
student_program(Program)-定义了student模型的 paddle program
data_name_map(dict)-teacher输入接口名与student输入接口名的映射，其中dict的 key 为teacher的输入名，value 为student的输入名
place(fluid.CPUPlace()|fluid.CUDAPlace(N))-该参数表示程序运行在何种设备上，这里的N为GPU对应的ID
scope(Scope)-该参数表示程序使用的变量作用域，如果不指定将使用默认的全局作用域。默认值：fluid.global_scope()
name_prefix(str)-merge操作将统一为teacher的Variables添加的名称前缀name_prefix。默认值：'teacher_'

返回： 无

Note

data_name_map 是 teacher_var name到student_var name的映射，如果写反可能无法正确进行merge

使用示例：

import paddle.fluid as fluid
import paddleslim.dist as dist
student_program = fluid.Program()
with fluid.program_guard(student_program):
    x = fluid.layers.data(name='x', shape=[1, 28, 28])
    conv = fluid.layers.conv2d(x, 32, 1)
    out = fluid.layers.conv2d(conv, 64, 3, padding=1)
teacher_program = fluid.Program()
with fluid.program_guard(teacher_program):
    y = fluid.layers.data(name='y', shape=[1, 28, 28])
    conv = fluid.layers.conv2d(y, 32, 1)
    conv = fluid.layers.conv2d(conv, 32, 3, padding=1)
    out = fluid.layers.conv2d(conv, 64, 3, padding=1)
data_name_map = {'y':'x'}
USE_GPU = False
place = fluid.CUDAPlace(0) if USE_GPU else fluid.CPUPlace()
dist.merge(teacher_program, student_program,
                          data_name_map, place)

fsp_loss#

paddleslim.dist.fsp_loss(teacher_var1_name, teacher_var2_name, student_var1_name, student_var2_name, program=fluid.default_main_program()) [源代码]: fsp_loss为program内的teacher var和student var添加fsp loss，出自论文<<A Gift from Knowledge Distillation: Fast Optimization, Network Minimization and Transfer Learning>>

参数：

teacher_var1_name(str): teacher_var1的名称. 对应的variable是一个形为[batch_size, x_channel, height, width]的4-D特征图Tensor，数据类型为float32或float64
teacher_var2_name(str): teacher_var2的名称. 对应的variable是一个形为[batch_size, y_channel, height, width]的4-D特征图Tensor，数据类型为float32或float64。只有y_channel可以与teacher_var1的x_channel不同，其他维度必须与teacher_var1相同
student_var1_name(str): student_var1的名称. 对应的variable需与teacher_var1尺寸保持一致，是一个形为[batch_size, x_channel, height, width]的4-D特征图Tensor，数据类型为float32或float64
student_var2_name(str): student_var2的名称. 对应的variable需与teacher_var2尺寸保持一致，是一个形为[batch_size, y_channel, height, width]的4-D特征图Tensor，数据类型为float32或float64。只有y_channel可以与student_var1的x_channel不同，其他维度必须与student_var1相同
program(Program): 用于蒸馏训练的fluid program。默认值：fluid.default_main_program()

返回： 由teacher_var1, teacher_var2, student_var1, student_var2组合得到的fsp_loss

使用示例：

import paddle.fluid as fluid
import paddleslim.dist as dist
student_program = fluid.Program()
with fluid.program_guard(student_program):
    x = fluid.layers.data(name='x', shape=[1, 28, 28])
    conv = fluid.layers.conv2d(x, 32, 1, name='s1')
    out = fluid.layers.conv2d(conv, 64, 3, padding=1, name='s2')
teacher_program = fluid.Program()
with fluid.program_guard(teacher_program):
    y = fluid.layers.data(name='y', shape=[1, 28, 28])
    conv = fluid.layers.conv2d(y, 32, 1, name='t1')
    conv = fluid.layers.conv2d(conv, 32, 3, padding=1)
    out = fluid.layers.conv2d(conv, 64, 3, padding=1, name='t2')
data_name_map = {'y':'x'}
USE_GPU = False
place = fluid.CUDAPlace(0) if USE_GPU else fluid.CPUPlace()
merge(teacher_program, student_program, data_name_map, place)
with fluid.program_guard(student_program):
    distillation_loss = dist.fsp_loss('teacher_t1.tmp_1', 'teacher_t2.tmp_1',
                                      's1.tmp_1', 's2.tmp_1', main_program)

l2_loss#

paddleslim.dist.l2_loss(teacher_var_name, student_var_name, program=fluid.default_main_program())[源代码]: l2_loss为program内的teacher var和student var添加l2 loss

参数：

teacher_var_name(str): teacher_var的名称.
student_var_name(str): student_var的名称.
program(Program): 用于蒸馏训练的fluid program。默认值：fluid.default_main_program()

返回： 由teacher_var, student_var组合得到的l2_loss

使用示例：

import paddle.fluid as fluid
import paddleslim.dist as dist
student_program = fluid.Program()
with fluid.program_guard(student_program):
    x = fluid.layers.data(name='x', shape=[1, 28, 28])
    conv = fluid.layers.conv2d(x, 32, 1, name='s1')
    out = fluid.layers.conv2d(conv, 64, 3, padding=1, name='s2')
teacher_program = fluid.Program()
with fluid.program_guard(teacher_program):
    y = fluid.layers.data(name='y', shape=[1, 28, 28])
    conv = fluid.layers.conv2d(y, 32, 1, name='t1')
    conv = fluid.layers.conv2d(conv, 32, 3, padding=1)
    out = fluid.layers.conv2d(conv, 64, 3, padding=1, name='t2')
data_name_map = {'y':'x'}
USE_GPU = False
place = fluid.CUDAPlace(0) if USE_GPU else fluid.CPUPlace()
merge(teacher_program, student_program, data_name_map, place)
with fluid.program_guard(student_program):
    distillation_loss = dist.l2_loss('teacher_t2.tmp_1', 's2.tmp_1',
                                     main_program)

soft_label_loss#

paddleslim.dist.soft_label_loss(teacher_var_name, student_var_name, program=fluid.default_main_program(), teacher_temperature=1., student_temperature=1.)[源代码]: soft_label_loss为program内的teacher var和student var添加soft label loss，出自论文<<Distilling the Knowledge in a Neural Network>>

参数：

teacher_var_name(str): teacher_var的名称.
student_var_name(str): student_var的名称.
program(Program): 用于蒸馏训练的fluid program。默认值：fluid.default_main_program()
teacher_temperature(float): 对teacher_var进行soft操作的温度值，温度值越大得到的特征图越平滑
student_temperature(float): 对student_var进行soft操作的温度值，温度值越大得到的特征图越平滑

返回： 由teacher_var, student_var组合得到的soft_label_loss

使用示例：

import paddle.fluid as fluid
import paddleslim.dist as dist
student_program = fluid.Program()
with fluid.program_guard(student_program):
    x = fluid.layers.data(name='x', shape=[1, 28, 28])
    conv = fluid.layers.conv2d(x, 32, 1, name='s1')
    out = fluid.layers.conv2d(conv, 64, 3, padding=1, name='s2')
teacher_program = fluid.Program()
with fluid.program_guard(teacher_program):
    y = fluid.layers.data(name='y', shape=[1, 28, 28])
    conv = fluid.layers.conv2d(y, 32, 1, name='t1')
    conv = fluid.layers.conv2d(conv, 32, 3, padding=1)
    out = fluid.layers.conv2d(conv, 64, 3, padding=1, name='t2')
data_name_map = {'y':'x'}
USE_GPU = False
place = fluid.CUDAPlace(0) if USE_GPU else fluid.CPUPlace()
merge(teacher_program, student_program, data_name_map, place)
with fluid.program_guard(student_program):
    distillation_loss = dist.soft_label_loss('teacher_t2.tmp_1',
                                             's2.tmp_1', main_program, 1., 1.)

loss#

paddleslim.dist.loss(loss_func, program=fluid.default_main_program(), **kwargs) [源代码]: loss函数支持对任意多对teacher_var和student_var使用自定义损失函数

参数：

loss_func(python function): 自定义的损失函数，输入为teacher var和student var，输出为自定义的loss
program(Program): 用于蒸馏训练的fluid program。默认值：fluid.default_main_program()
**kwargs: loss_func输入名与对应variable名称

返回：自定义的损失函数loss

使用示例：

import paddle.fluid as fluid
import paddleslim.dist as dist
student_program = fluid.Program()
with fluid.program_guard(student_program):
    x = fluid.layers.data(name='x', shape=[1, 28, 28])
    conv = fluid.layers.conv2d(x, 32, 1, name='s1')
    out = fluid.layers.conv2d(conv, 64, 3, padding=1, name='s2')
teacher_program = fluid.Program()
with fluid.program_guard(teacher_program):
    y = fluid.layers.data(name='y', shape=[1, 28, 28])
    conv = fluid.layers.conv2d(y, 32, 1, name='t1')
    conv = fluid.layers.conv2d(conv, 32, 3, padding=1)
    out = fluid.layers.conv2d(conv, 64, 3, padding=1, name='t2')
data_name_map = {'y':'x'}
USE_GPU = False
place = fluid.CUDAPlace(0) if USE_GPU else fluid.CPUPlace()
merge(teacher_program, student_program, data_name_map, place)
def adaptation_loss(t_var, s_var):
    teacher_channel = t_var.shape[1]
    s_hint = fluid.layers.conv2d(s_var, teacher_channel, 1)
    hint_loss = fluid.layers.reduce_mean(fluid.layers.square(s_hint - t_var))
    return hint_loss
with fluid.program_guard(student_program):
    distillation_loss = dist.loss(main_program, adaptation_loss,
            t_var='teacher_t2.tmp_1', s_var='s2.tmp_1')

注意事项

在添加蒸馏loss时会引入新的variable，需要注意新引入的variable不要与student variables命名冲突。这里建议两种用法（两种方法任选其一即可）：

建议与student_program使用同一个命名空间，以避免一些未指定名称的variables(例如tmp_0, tmp_1...)多次定义为同一名称出现命名冲突
建议在添加蒸馏loss时指定一个命名空间前缀，具体用法请参考Paddle官方文档fluid.name_scope