Layer definitions and declarations re-organization and documentation

include/caffe/data_layers.hpp

@@ -0,0 +1,218 @@
+// Copyright 2014 BVLC and contributors.
+
+#ifndef CAFFE_DATA_LAYERS_HPP_
+#define CAFFE_DATA_LAYERS_HPP_
+
+#include <string>
+#include <utility>
+#include <vector>
+
+#include "leveldb/db.h"
+#include "pthread.h"
+#include "hdf5.h"
+#include "boost/scoped_ptr.hpp"
+
+#include "caffe/blob.hpp"
+#include "caffe/common.hpp"
+#include "caffe/layer.hpp"
+#include "caffe/proto/caffe.pb.h"
+
+namespace caffe {
+
+#define HDF5_DATA_DATASET_NAME "data"
+#define HDF5_DATA_LABEL_NAME "label"
+
+template <typename Dtype>
+class HDF5OutputLayer : public Layer<Dtype> {
+ public:
+  explicit HDF5OutputLayer(const LayerParameter& param);
+  virtual ~HDF5OutputLayer();
+  virtual void SetUp(const vector<Blob<Dtype>*>& bottom,
+      vector<Blob<Dtype>*>* top);
+  inline std::string file_name() const { return file_name_; }
+
+ protected:
+  virtual Dtype Forward_cpu(const vector<Blob<Dtype>*>& bottom,
+      vector<Blob<Dtype>*>* top);
+  virtual Dtype Forward_gpu(const vector<Blob<Dtype>*>& bottom,
+      vector<Blob<Dtype>*>* top);
+  virtual void Backward_cpu(const vector<Blob<Dtype>*>& top,
+      const bool propagate_down, vector<Blob<Dtype>*>* bottom);
+  virtual void Backward_gpu(const vector<Blob<Dtype>*>& top,
+      const bool propagate_down, vector<Blob<Dtype>*>* bottom);
+  virtual void SaveBlobs();
+
+  std::string file_name_;
+  hid_t file_id_;
+  Blob<Dtype> data_blob_;
+  Blob<Dtype> label_blob_;
+};
+
+
+template <typename Dtype>
+class HDF5DataLayer : public Layer<Dtype> {
+ public:
+  explicit HDF5DataLayer(const LayerParameter& param)
+      : Layer<Dtype>(param) {}
+  virtual ~HDF5DataLayer();
+  virtual void SetUp(const vector<Blob<Dtype>*>& bottom,
+      vector<Blob<Dtype>*>* top);
+
+ protected:
+  virtual Dtype Forward_cpu(const vector<Blob<Dtype>*>& bottom,
+      vector<Blob<Dtype>*>* top);
+  virtual Dtype Forward_gpu(const vector<Blob<Dtype>*>& bottom,
+      vector<Blob<Dtype>*>* top);
+  virtual void Backward_cpu(const vector<Blob<Dtype>*>& top,
+      const bool propagate_down, vector<Blob<Dtype>*>* bottom);
+  virtual void Backward_gpu(const vector<Blob<Dtype>*>& top,
+      const bool propagate_down, vector<Blob<Dtype>*>* bottom);
+  virtual void LoadHDF5FileData(const char* filename);
+
+  std::vector<std::string> hdf_filenames_;
+  unsigned int num_files_;
+  unsigned int current_file_;
+  hsize_t current_row_;
+  Blob<Dtype> data_blob_;
+  Blob<Dtype> label_blob_;
+};
+
+// TODO: DataLayer, ImageDataLayer, and WindowDataLayer all have the
+// same basic structure and a lot of duplicated code.
+
+// This function is used to create a pthread that prefetches the data.
+template <typename Dtype>
+void* DataLayerPrefetch(void* layer_pointer);
+
+template <typename Dtype>
+class DataLayer : public Layer<Dtype> {
+  // The function used to perform prefetching.
+  friend void* DataLayerPrefetch<Dtype>(void* layer_pointer);
+
+ public:
+  explicit DataLayer(const LayerParameter& param)
+      : Layer<Dtype>(param) {}
+  virtual ~DataLayer();
+  virtual void SetUp(const vector<Blob<Dtype>*>& bottom,
+      vector<Blob<Dtype>*>* top);
+
+ protected:
+  virtual Dtype Forward_cpu(const vector<Blob<Dtype>*>& bottom,
+      vector<Blob<Dtype>*>* top);
+  virtual Dtype Forward_gpu(const vector<Blob<Dtype>*>& bottom,
+      vector<Blob<Dtype>*>* top);
+  virtual void Backward_cpu(const vector<Blob<Dtype>*>& top,
+      const bool propagate_down, vector<Blob<Dtype>*>* bottom) { return; }
+  virtual void Backward_gpu(const vector<Blob<Dtype>*>& top,
+      const bool propagate_down, vector<Blob<Dtype>*>* bottom) { return; }
+
+  virtual void CreatePrefetchThread();
+  virtual void JoinPrefetchThread();
+  virtual unsigned int PrefetchRand();
+
+  shared_ptr<Caffe::RNG> prefetch_rng_;
+  shared_ptr<leveldb::DB> db_;
+  shared_ptr<leveldb::Iterator> iter_;
+  int datum_channels_;
+  int datum_height_;
+  int datum_width_;
+  int datum_size_;
+  pthread_t thread_;
+  shared_ptr<Blob<Dtype> > prefetch_data_;
+  shared_ptr<Blob<Dtype> > prefetch_label_;
+  Blob<Dtype> data_mean_;
+  bool output_labels_;
+  Caffe::Phase phase_;
+};
+
+// This function is used to create a pthread that prefetches the data.
+template <typename Dtype>
+void* ImageDataLayerPrefetch(void* layer_pointer);
+
+template <typename Dtype>
+class ImageDataLayer : public Layer<Dtype> {
+  // The function used to perform prefetching.
+  friend void* ImageDataLayerPrefetch<Dtype>(void* layer_pointer);
+
+ public:
+  explicit ImageDataLayer(const LayerParameter& param)
+      : Layer<Dtype>(param) {}
+  virtual ~ImageDataLayer();
+  virtual void SetUp(const vector<Blob<Dtype>*>& bottom,
+      vector<Blob<Dtype>*>* top);
+
+ protected:
+  virtual Dtype Forward_cpu(const vector<Blob<Dtype>*>& bottom,
+      vector<Blob<Dtype>*>* top);
+  virtual Dtype Forward_gpu(const vector<Blob<Dtype>*>& bottom,
+      vector<Blob<Dtype>*>* top);
+  virtual void Backward_cpu(const vector<Blob<Dtype>*>& top,
+      const bool propagate_down, vector<Blob<Dtype>*>* bottom) { return; }
+  virtual void Backward_gpu(const vector<Blob<Dtype>*>& top,
+      const bool propagate_down, vector<Blob<Dtype>*>* bottom) { return; }
+
+  virtual void ShuffleImages();
+
+  virtual void CreatePrefetchThread();
+  virtual void JoinPrefetchThread();
+  virtual unsigned int PrefetchRand();
+
+  shared_ptr<Caffe::RNG> prefetch_rng_;
+  vector<std::pair<std::string, int> > lines_;
+  int lines_id_;
+  int datum_channels_;
+  int datum_height_;
+  int datum_width_;
+  int datum_size_;
+  pthread_t thread_;
+  shared_ptr<Blob<Dtype> > prefetch_data_;
+  shared_ptr<Blob<Dtype> > prefetch_label_;
+  Blob<Dtype> data_mean_;
+  Caffe::Phase phase_;
+};
+
+
+// This function is used to create a pthread that prefetches the window data.
+template <typename Dtype>
+void* WindowDataLayerPrefetch(void* layer_pointer);
+
+template <typename Dtype>
+class WindowDataLayer : public Layer<Dtype> {
+  // The function used to perform prefetching.
+  friend void* WindowDataLayerPrefetch<Dtype>(void* layer_pointer);
+
+ public:
+  explicit WindowDataLayer(const LayerParameter& param)
+      : Layer<Dtype>(param) {}
+  virtual ~WindowDataLayer();
+  virtual void SetUp(const vector<Blob<Dtype>*>& bottom,
+      vector<Blob<Dtype>*>* top);
+
+ protected:
+  virtual Dtype Forward_cpu(const vector<Blob<Dtype>*>& bottom,
+      vector<Blob<Dtype>*>* top);
+  virtual Dtype Forward_gpu(const vector<Blob<Dtype>*>& bottom,
+      vector<Blob<Dtype>*>* top);
+  virtual void Backward_cpu(const vector<Blob<Dtype>*>& top,
+      const bool propagate_down, vector<Blob<Dtype>*>* bottom) { return; }
+  virtual void Backward_gpu(const vector<Blob<Dtype>*>& top,
+      const bool propagate_down, vector<Blob<Dtype>*>* bottom) { return; }
+
+  virtual void CreatePrefetchThread();
+  virtual void JoinPrefetchThread();
+  virtual unsigned int PrefetchRand();
+
+  shared_ptr<Caffe::RNG> prefetch_rng_;
+  pthread_t thread_;
+  shared_ptr<Blob<Dtype> > prefetch_data_;
+  shared_ptr<Blob<Dtype> > prefetch_label_;
+  Blob<Dtype> data_mean_;
+  vector<std::pair<std::string, vector<int> > > image_database_;
+  enum WindowField { IMAGE_INDEX, LABEL, OVERLAP, X1, Y1, X2, Y2, NUM };
+  vector<vector<float> > fg_windows_;
+  vector<vector<float> > bg_windows_;
+};
+
+}  // namespace caffe
+
+#endif  // CAFFE_DATA_LAYERS_HPP_

include/caffe/loss_layers.hpp

@@ -0,0 +1,171 @@
+// Copyright 2014 BVLC and contributors.
+
+#ifndef CAFFE_LOSS_LAYERS_HPP_
+#define CAFFE_LOSS_LAYERS_HPP_
+
+#include <string>
+#include <utility>
+#include <vector>
+
+#include "leveldb/db.h"
+#include "pthread.h"
+#include "boost/scoped_ptr.hpp"
+#include "hdf5.h"
+
+#include "caffe/blob.hpp"
+#include "caffe/common.hpp"
+#include "caffe/layer.hpp"
+#include "caffe/neuron_layers.hpp"
+#include "caffe/proto/caffe.pb.h"
+
+namespace caffe {
+
+const float kLOG_THRESHOLD = 1e-20;
+
+/* LossLayer
+  Takes two inputs of same num (a and b), and has no output.
+  The gradient is propagated to a.
+*/
+template <typename Dtype>
+class LossLayer : public Layer<Dtype> {
+ public:
+  explicit LossLayer(const LayerParameter& param)
+     : Layer<Dtype>(param) {}
+  virtual void SetUp(
+      const vector<Blob<Dtype>*>& bottom, vector<Blob<Dtype>*>* top);
+  virtual void FurtherSetUp(
+      const vector<Blob<Dtype>*>& bottom, vector<Blob<Dtype>*>* top) {}
+};
+
+/* SigmoidCrossEntropyLossLayer
+*/
+template <typename Dtype>
+class SigmoidCrossEntropyLossLayer : public LossLayer<Dtype> {
+ public:
+  explicit SigmoidCrossEntropyLossLayer(const LayerParameter& param)
+      : LossLayer<Dtype>(param),
+          sigmoid_layer_(new SigmoidLayer<Dtype>(param)),
+          sigmoid_output_(new Blob<Dtype>()) {}
+  virtual void FurtherSetUp(const vector<Blob<Dtype>*>& bottom,
+      vector<Blob<Dtype>*>* top);
+
+ protected:
+  virtual Dtype Forward_cpu(const vector<Blob<Dtype>*>& bottom,
+      vector<Blob<Dtype>*>* top);
+  virtual Dtype Forward_gpu(const vector<Blob<Dtype>*>& bottom,
+      vector<Blob<Dtype>*>* top);
+  virtual void Backward_cpu(const vector<Blob<Dtype>*>& top,
+      const bool propagate_down, vector<Blob<Dtype>*>* bottom);
+  virtual void Backward_gpu(const vector<Blob<Dtype>*>& top,
+      const bool propagate_down, vector<Blob<Dtype>*>* bottom);
+
+  shared_ptr<SigmoidLayer<Dtype> > sigmoid_layer_;
+  // sigmoid_output stores the output of the sigmoid layer.
+  shared_ptr<Blob<Dtype> > sigmoid_output_;
+  // Vector holders to call the underlying sigmoid layer forward and backward.
+  vector<Blob<Dtype>*> sigmoid_bottom_vec_;
+  vector<Blob<Dtype>*> sigmoid_top_vec_;
+};
+
+/* EuclideanLossLayer
+  Compute the L_2 distance between the two inputs.
+
+  loss = (1/2 \sum_i (a_i - b_i)^2)
+  a' = 1/I (a - b)
+*/
+template <typename Dtype>
+class EuclideanLossLayer : public LossLayer<Dtype> {
+ public:
+  explicit EuclideanLossLayer(const LayerParameter& param)
+      : LossLayer<Dtype>(param), diff_() {}
+  virtual void FurtherSetUp(const vector<Blob<Dtype>*>& bottom,
+      vector<Blob<Dtype>*>* top);
+
+ protected:
+  virtual Dtype Forward_cpu(const vector<Blob<Dtype>*>& bottom,
+      vector<Blob<Dtype>*>* top);
+  virtual void Backward_cpu(const vector<Blob<Dtype>*>& top,
+      const bool propagate_down, vector<Blob<Dtype>*>* bottom);
+
+  Blob<Dtype> diff_;
+};
+
+/* InfogainLossLayer
+*/
+template <typename Dtype>
+class InfogainLossLayer : public LossLayer<Dtype> {
+ public:
+  explicit InfogainLossLayer(const LayerParameter& param)
+      : LossLayer<Dtype>(param), infogain_() {}
+  virtual void FurtherSetUp(const vector<Blob<Dtype>*>& bottom,
+      vector<Blob<Dtype>*>* top);
+
+ protected:
+  virtual Dtype Forward_cpu(const vector<Blob<Dtype>*>& bottom,
+      vector<Blob<Dtype>*>* top);
+  virtual void Backward_cpu(const vector<Blob<Dtype>*>& top,
+      const bool propagate_down, vector<Blob<Dtype>*>* bottom);
+
+  Blob<Dtype> infogain_;
+};
+
+/* HingeLossLayer
+*/
+template <typename Dtype>
+class HingeLossLayer : public LossLayer<Dtype> {
+ public:
+  explicit HingeLossLayer(const LayerParameter& param)
+      : LossLayer<Dtype>(param) {}
+
+ protected:
+  virtual Dtype Forward_cpu(const vector<Blob<Dtype>*>& bottom,
+      vector<Blob<Dtype>*>* top);
+  virtual void Backward_cpu(const vector<Blob<Dtype>*>& top,
+      const bool propagate_down, vector<Blob<Dtype>*>* bottom);
+};
+
+/* MultinomialLogisticLossLayer
+*/
+template <typename Dtype>
+class MultinomialLogisticLossLayer : public LossLayer<Dtype> {
+ public:
+  explicit MultinomialLogisticLossLayer(const LayerParameter& param)
+      : LossLayer<Dtype>(param) {}
+  virtual void FurtherSetUp(const vector<Blob<Dtype>*>& bottom,
+      vector<Blob<Dtype>*>* top);
+
+ protected:
+  virtual Dtype Forward_cpu(const vector<Blob<Dtype>*>& bottom,
+      vector<Blob<Dtype>*>* top);
+  virtual void Backward_cpu(const vector<Blob<Dtype>*>& top,
+      const bool propagate_down, vector<Blob<Dtype>*>* bottom);
+};
+
+/* AccuracyLayer
+  Note: not an actual loss layer! Does not implement backwards step.
+  Computes the accuracy and logprob of a with respect to b.
+*/
+template <typename Dtype>
+class AccuracyLayer : public Layer<Dtype> {
+ public:
+  explicit AccuracyLayer(const LayerParameter& param)
+      : Layer<Dtype>(param) {}
+  virtual void SetUp(const vector<Blob<Dtype>*>& bottom,
+      vector<Blob<Dtype>*>* top);
+
+ protected:
+  virtual Dtype Forward_cpu(const vector<Blob<Dtype>*>& bottom,
+      vector<Blob<Dtype>*>* top);
+  virtual void Backward_cpu(const vector<Blob<Dtype>*>& top,
+      const bool propagate_down, vector<Blob<Dtype>*>* bottom) {
+    NOT_IMPLEMENTED;
+  }
+};
+
+/* Also see
+- SoftmaxWithLossLayer in vision_layers.hpp
+*/
+
+}  // namespace caffe
+
+#endif  // CAFFE_LOSS_LAYERS_HPP_