神经网络压缩的研究近三年十分热门,笔者查阅到相关的两篇博客,博主们非常奉献的提供了源代码,但是发发现在使用gpu训练添加mask的网络上,稍微有些不顺,特此再进行详细说明。
此文是在 基于Caffe的CNN剪枝[1]和 Deep Compression阅读理解及Caffe源码修改[2] 的基础上修改的。
mask的结构?
[1]中使用的blob,存储mask。blob是一块数据块,在初始化时,需要为gpu上的数据块申请一块空间,故有Addmask()函数。AddMask()是blob.hpp中的blob的成员方法,需要在blob.cpp中实现。使用时将Addmask()添加在innerproduct.cpp和base_conv.cpp中,使得网络在setuplayer的过程中,为fc层和conv层多开辟一块存放mask的syncedmemory。blob有一系列需要实现的cpu_data()/mutable_cpu_data()等,初始化中改变mask的值时需要注意使用合理的方式。
InnerProductLayer.cpp
1 void InnerProductLayer<Dtype>::LayerSetUp(const vector<Blob<Dtype>*>& bottom, 2 const vector<Blob<Dtype>*>& top) { 3 ... 4 this->blobs_[0].reset(new Blob<Dtype>(weight_shape)); 5 this->blobs_[0]->Addmask(); 6 ...}
base_conv.cpp:
1 template <typename Dtype> 2 void BaseConvolutionLayer<Dtype>::LayerSetUp(const vector<Blob<Dtype>*>& bottom, 3 const vector<Blob<Dtype>*>& top) { 4 ... 5 this->blobs_[0].reset(new Blob<Dtype>(weight_shape)); 6 this->blobs_[0]->Addmask(); 7 ...}
修改blob.hpp和blob.cpp,添加成员mask_和相关的方法,在[1]文章的评论里作者已给出源代码。
[2]中使用layer结构定义mask,layer是相当于数据的一系列操作,或者说是blob的组合方法。
但是,想要实现在gpu上的操作,数据需要有gpu有关的操作。故此处采用[1]中的方法,将mask_添加到blob class中,实现mask_属性。
mask的初始化?
在Caffe框架下,网络的初始化有两种方式,一种是调用filler,按照模型中定义的初始化方式进行初始化,第二种是从已有的caffemodel或者snapshot中读取相应参数矩阵进行初始化[1]。
1、filler的方法
在程序开始时,网络使用net.cpp中的Init()进行初始化,由输入至输出,依次调用各个层的layersetup,建立网络结构。如下所示是caffe中使用xavier方法进行填充的操作。
1 virtual void Fill(Blob<Dtype>* blob) { 2 CHECK(blob->count()); 3 int fan_in = blob->count() / blob->num(); 4 int fan_out = blob->count() / blob->channels(); 5 Dtype n = fan_in; // default to fan_in 6 if (this->filler_param_.variance_norm() == 7 FillerParameter_VarianceNorm_AVERAGE) { 8 n = (fan_in + fan_out) / Dtype(2); 9 } else if (this->filler_param_.variance_norm() == 10 FillerParameter_VarianceNorm_FAN_OUT) { 11 n = fan_out; 12 } 13 Dtype scale = sqrt(Dtype(3) / n); 14 caffe_rng_uniform<Dtype>(blob->count(), -scale, scale, 15 blob->mutable_cpu_data()); 16 //Filler<Dtype>:: FillMask(blob); 17 CHECK_EQ(this->filler_param_.sparse(), -1) 18 << "Sparsity not supported by this Filler."; 19 }
filler的作用是,为建立的网络结构产生随机初始化值。
即使是从snapshot或caffemodel中读入数据,也执行随机填充操作。
2、从snapshot或caffemodel中读入数据
tools/caffe.cpp 中的phase:train可以从snapshot或caffemodel中提取参数,进行finetune。phase:test则可以从提取的参数中建立网络,进行预测过程。
这里笔者的网络结构是在pycaffe中进行稀疏化的,因此读入网络的proto文件是一个连接数不变、存在部分连接权值为零的网络。需要在读入参数的同时初始化mask_。因此修改blob.cpp中的fromproto函数:
1 template <typename Dtype> 2 void Blob<Dtype>::FromProto(const BlobProto& proto, bool reshape) { 3 if (reshape) { 4 vector<int> shape; 5 if (proto.has_num() || proto.has_channels() || 6 proto.has_height() || proto.has_width()) { 7 // Using deprecated 4D Blob dimensions -- 8 // shape is (num, channels, height, width). 9 shape.resize(4); 10 shape[0] = proto.num(); 11 shape[1] = proto.channels(); 12 shape[2] = proto.height(); 13 shape[3] = proto.width(); 14 } else { 15 shape.resize(proto.shape().dim_size()); 16 for (int i = 0; i < proto.shape().dim_size(); ++i) { 17 shape[i] = proto.shape().dim(i); 18 } 19 } 20 Reshape(shape); 21 } else { 22 CHECK(ShapeEquals(proto)) << "shape mismatch (reshape not set)"; 23 } 24 // copy data 25 Dtype* data_vec = mutable_cpu_data(); 26 if (proto.double_data_size() > 0) { 27 CHECK_EQ(count_, proto.double_data_size()); 28 for (int i = 0; i < count_; ++i) { 29 data_vec[i] = proto.double_data(i); 30 } 31 } else { 32 CHECK_EQ(count_, proto.data_size()); 33 for (int i = 0; i < count_; ++i) { 34 data_vec[i] = proto.data(i); 35 } 36 } 37 if (proto.double_diff_size() > 0) { 38 CHECK_EQ(count_, proto.double_diff_size()); 39 Dtype* diff_vec = mutable_cpu_diff(); 40 for (int i = 0; i < count_; ++i) { 41 diff_vec[i] = proto.double_diff(i); 42 } 43 } else if (proto.diff_size() > 0) { 44 CHECK_EQ(count_, proto.diff_size()); 45 Dtype* diff_vec = mutable_cpu_diff(); 46 for (int i = 0; i < count_; ++i) { 47 diff_vec[i] = proto.diff(i); 48 } 49 } 50 if(shape_.size()==4||shape_.size()==2){ 51 Dtype* mask_vec = mutable_cpu_data(); 52 CHECK(count_); 53 for(int i=0;i<count_;i++) 54 mask_vec[i]=data_vec[i]?1:0; 55 }
在读入proto文件的同时,如果层的大小是4D——conv层、或2D——fc层时,初始化mask_为data_vec[i]?1:0。当层的大小是1Ds——pool或relu层时,不进行mask的初始化。
反向传播的修改?
1、修改blob的更新方式,添加math_funcion.hpp头文件。
1 template <typename Dtype> 2 void Blob<Dtype>::Update() { 3 // We will perform update based on where the data is located. 4 switch (data_->head()) { 5 case SyncedMemory::HEAD_AT_CPU: 6 // perform computation on CPU 7 caffe_axpy<Dtype>(count_, Dtype(-1), 8 static_cast<const Dtype*>(diff_->cpu_data()), 9 static_cast<Dtype*>(data_->mutable_cpu_data())); 10 caffe_mul<Dtype>(count_, 11 static_cast<const Dtype*>(mask_->cpu_data()), 12 static_cast<const Dtype*>(data_->cpu_data()), 13 static_cast<Dtype*>(data_->mutable_cpu_data())); 14 break; 15 case SyncedMemory::HEAD_AT_GPU: 16 case SyncedMemory::SYNCED: 17 #ifndef CPU_ONLY 18 // perform computation on GPU 19 caffe_gpu_axpy<Dtype>(count_, Dtype(-1), 20 static_cast<const Dtype*>(diff_->gpu_data()), 21 static_cast<Dtype*>(data_->mutable_gpu_data())); 22 caffe_gpu_mul<Dtype>(count_, 23 static_cast<const Dtype*>(mask_->gpu_data()), 24 static_cast<const Dtype*>(data_->gpu_data()), 25 static_cast<Dtype*>(data_->mutable_gpu_data())); 26 #else 27 NO_GPU; 28 #endif 29 break; 30 default: 31 LOG(FATAL) << "Syncedmem not initialized."; 32 } 33 }
2、为cpu下的计算和gpu下的计算分别添加形如weight[i]*=mask[i];的运算方式。
inner_product_layer.cpp:
1 void InnerProductLayer<Dtype>::Backward_cpu(const vector<Blob<Dtype>*>& top, 2 const vector<bool>& propagate_down, 3 const vector<Blob<Dtype>*>& bottom) { 4 if (this->param_propagate_down_[0]) { 5 const Dtype* top_diff = top[0]->cpu_diff(); 6 const Dtype* bottom_data = bottom[0]->cpu_data(); 7 // Gradient with respect to weight 8 Dtype* weight_diff = this->blobs_[0]->mutable_cpu_diff(); 9 vector<int> weight_shape(2); 10 if (transpose_) { 11 weight_shape[0] = K_; 12 weight_shape[1] = N_; 13 } else { 14 weight_shape[0] = N_; 15 weight_shape[1] = K_; 16 } 17 int count = weight_shape[0]*weight_shape[1]; 18 const Dtype* mask = this->blobs_[0]->cpu_mask(); 19 for(int j=0;j<count;j++) 20 weight_diff[j]*=mask[j]; 21 22 if (transpose_) { 23 caffe_cpu_gemm<Dtype>(CblasTrans, CblasNoTrans, 24 K_, N_, M_, 25 (Dtype)1., bottom_data, top_diff, 26 (Dtype)1., weight_diff); 27 } else { 28 caffe_cpu_gemm<Dtype>(CblasTrans, CblasNoTrans, 29 N_, K_, M_, 30 (Dtype)1., top_diff, bottom_data, 31 (Dtype)1., weight_diff); 32 } 33 } 34 if (bias_term_ && this->param_propagate_down_[1]) { 35 const Dtype* top_diff = top[0]->cpu_diff(); 36 // Gradient with respect to bias 37 caffe_cpu_gemv<Dtype>(CblasTrans, M_, N_, (Dtype)1., top_diff, 38 bias_multiplier_.cpu_data(), (Dtype)1., 39 this->blobs_[1]->mutable_cpu_diff()); 40 } 41 if (propagate_down[0]) { 42 const Dtype* top_diff = top[0]->cpu_diff(); 43 // Gradient with respect to bottom data 44 if (transpose_) { 45 caffe_cpu_gemm<Dtype>(CblasNoTrans, CblasTrans, 46 M_, K_, N_, 47 (Dtype)1., top_diff, this->blobs_[0]->cpu_data(), 48 (Dtype)0., bottom[0]->mutable_cpu_diff()); 49 } else { 50 caffe_cpu_gemm<Dtype>(CblasNoTrans, CblasNoTrans, 51 M_, K_, N_, 52 (Dtype)1., top_diff, this->blobs_[0]->cpu_data(), 53 (Dtype)0., bottom[0]->mutable_cpu_diff()); 54 } 55 } 56 }
inner_product_layer.cu:
1 template <typename Dtype> 2 void InnerProductLayer<Dtype>::Backward_gpu(const vector<Blob<Dtype>*>& top, 3 const vector<bool>& propagate_down, 4 const vector<Blob<Dtype>*>& bottom) { 5 if (this->param_propagate_down_[0]) { 6 const Dtype* top_diff = top[0]->gpu_diff(); 7 const Dtype* bottom_data = bottom[0]->gpu_data(); 8 vector<int> weight_shape(2); 9 if (transpose_) { 10 weight_shape[0] = K_; 11 weight_shape[1] = N_; 12 } else { 13 weight_shape[0] = N_; 14 weight_shape[1] = K_; 15 } 16 int count = weight_shape[0]*weight_shape[1]; 17 caffe_gpu_mul<Dtype>(count,static_cast<const Dtype*>(this->blobs_[0]->mutable_gpu_diff()),static_cast<const Dtype*>(this->blobs_[0]->gpu_mask()),static_cast<Dtype*>(this->blobs_[0]->mutable_gpu_diff())); 18 Dtype* weight_diff = this->blobs_[0]->mutable_gpu_diff(); 19 //for(int j=0;j<count;j++) 20 //weight_diff[j]*=this->masks_[j]; 21 // Gradient with respect to weight 22 if (transpose_) { 23 caffe_gpu_gemm<Dtype>(CblasTrans, CblasNoTrans, 24 K_, N_, M_, 25 (Dtype)1., bottom_data, top_diff, 26 (Dtype)1., weight_diff); 27 } else { 28 caffe_gpu_gemm<Dtype>(CblasTrans, CblasNoTrans, 29 N_, K_, M_, 30 (Dtype)1., top_diff, bottom_data, 31 (Dtype)1., weight_diff); 32 } 33 } 34 if (bias_term_ && this->param_propagate_down_[1]) { 35 const Dtype* top_diff = top[0]->gpu_diff(); 36 // Gradient with respect to bias 37 caffe_gpu_gemv<Dtype>(CblasTrans, M_, N_, (Dtype)1., top_diff, 38 bias_multiplier_.gpu_data(), (Dtype)1., 39 this->blobs_[1]->mutable_gpu_diff()); 40 } 41 if (propagate_down[0]) { 42 const Dtype* top_diff = top[0]->gpu_diff(); 43 // Gradient with respect to bottom data 44 if (transpose_) { 45 caffe_gpu_gemm<Dtype>(CblasNoTrans, CblasTrans, 46 M_, K_, N_, 47 (Dtype)1., top_diff, this->blobs_[0]->gpu_data(), 48 (Dtype)0., bottom[0]->mutable_gpu_diff()); 49 } else { 50 caffe_gpu_gemm<Dtype>(CblasNoTrans, CblasNoTrans, 51 M_, K_, N_, 52 (Dtype)1., top_diff, this->blobs_[0]->gpu_data(), 53 (Dtype)0., bottom[0]->mutable_gpu_diff()); 54 } 55 } 56 }
至此修改完毕。
另外,caffe在新的版本中已添加sparse_参数,参考 https://github.com/BVLC/caffe/pulls?utf8=%E2%9C%93&q=sparse