add mnist autoencoder example necessities (sigmoid cross entropy loss

layer, sparse gaussian filler)

4 files changed, 418 insertions, 0 deletions

diff --git a/examples/mnist/mnist_autoencoder_solver.prototxt b/examples/mnist/mnist_autoencoder_solver.prototxt
new file mode 100644
index 00000000..b11b2c46
--- /dev/null
+++ b/examples/mnist/mnist_autoencoder_solver.prototxt
@@ -0,0 +1,14 @@
+train_net: "mnist_autoencoder_train.prototxt"
+test_net: "mnist_autoencoder_test.prototxt"
+test_iter: 50
+test_interval: 100
+base_lr: 0.0001
+lr_policy: "fixed"
+display: 20
+max_iter: 4500000
+weight_decay: 0.0005
+snapshot: 10000
+snapshot_prefix: "alexnet_train"
+momentum: 0.9
+solver_mode: 1
+device_id: 1
diff --git a/examples/mnist/mnist_autoencoder_test.prototxt b/examples/mnist/mnist_autoencoder_test.prototxt
new file mode 100644
index 00000000..bec7a3c2
--- /dev/null
+++ b/examples/mnist/mnist_autoencoder_test.prototxt
@@ -0,0 +1,164 @@
+name: "MNISTAutoencoder"
+layers {
+  top: "data"
+  top: "label"
+  name: "data"
+  type: DATA
+  data_param {
+    source: "mnist-test-leveldb"
+    scale: 0.0039215684
+    batch_size: 100
+  }
+}
+layers {
+  bottom: "data"
+  top: "flatdata"
+  name: "flatdata"
+  type: FLATTEN
+}
+layers {
+  bottom: "data"
+  top: "encode1"
+  name: "encode1"
+  type: INNER_PRODUCT
+  blobs_lr: 1
+  blobs_lr: 1
+  weight_decay: 1
+  weight_decay: 0
+  inner_product_param {
+    num_output: 1000
+  }
+}
+layers {
+  bottom: "encode1"
+  top: "encode1neuron"
+  name: "encode1neuron"
+  type: SIGMOID
+}
+layers {
+  bottom: "encode1neuron"
+  top: "encode2"
+  name: "encode2"
+  type: INNER_PRODUCT
+  blobs_lr: 1
+  blobs_lr: 1
+  weight_decay: 1
+  weight_decay: 0
+  inner_product_param {
+    num_output: 500
+  }
+}
+layers {
+  bottom: "encode2"
+  top: "encode2neuron"
+  name: "encode2neuron"
+  type: SIGMOID
+}
+layers {
+  bottom: "encode2neuron"
+  top: "encode3"
+  name: "encode3"
+  type: INNER_PRODUCT
+  blobs_lr: 1
+  blobs_lr: 1
+  weight_decay: 1
+  weight_decay: 0
+  inner_product_param {
+    num_output: 250
+  }
+}
+layers {
+  bottom: "encode3"
+  top: "encode3neuron"
+  name: "encode3neuron"
+  type: SIGMOID
+}
+layers {
+  bottom: "encode3neuron"
+  top: "encode4"
+  name: "encode4"
+  type: INNER_PRODUCT
+  blobs_lr: 1
+  blobs_lr: 1
+  weight_decay: 1
+  weight_decay: 0
+  inner_product_param {
+    num_output: 30
+  }
+}
+layers {
+  bottom: "encode4"
+  top: "decode4"
+  name: "decode4"
+  type: INNER_PRODUCT
+  blobs_lr: 1
+  blobs_lr: 1
+  weight_decay: 1
+  weight_decay: 0
+  inner_product_param {
+    num_output: 250
+  }
+}
+layers {
+  bottom: "decode4"
+  top: "decode4neuron"
+  name: "decode4neuron"
+  type: SIGMOID
+}
+layers {
+  bottom: "decode4neuron"
+  top: "decode3"
+  name: "decode3"
+  type: INNER_PRODUCT
+  blobs_lr: 1
+  blobs_lr: 1
+  weight_decay: 1
+  weight_decay: 0
+  inner_product_param {
+    num_output: 500
+  }
+}
+layers {
+  bottom: "decode3"
+  top: "decode3neuron"
+  name: "decode3neuron"
+  type: SIGMOID
+}
+layers {
+  bottom: "decode3neuron"
+  top: "decode2"
+  name: "decode2"
+  type: INNER_PRODUCT
+  blobs_lr: 1
+  blobs_lr: 1
+  weight_decay: 1
+  weight_decay: 0
+  inner_product_param {
+    num_output: 1000
+  }
+}
+layers {
+  bottom: "decode2"
+  top: "decode2neuron"
+  name: "decode2neuron"
+  type: SIGMOID
+}
+layers {
+  bottom: "decode2neuron"
+  top: "decode1"
+  name: "decode1"
+  type: INNER_PRODUCT
+  blobs_lr: 1
+  blobs_lr: 1
+  weight_decay: 1
+  weight_decay: 0
+  inner_product_param {
+    num_output: 784
+  }
+}
+layers {
+  bottom: "decode1"
+  bottom: "flatdata"
+  name: "loss"
+  type: EUCLIDEAN_LOSS
+}
diff --git a/examples/mnist/mnist_autoencoder_train.prototxt b/examples/mnist/mnist_autoencoder_train.prototxt
new file mode 100644
index 00000000..d5201eb9
--- /dev/null
+++ b/examples/mnist/mnist_autoencoder_train.prototxt
@@ -0,0 +1,236 @@
+name: "MNISTAutoencoder"
+layers {
+  top: "data"
+  top: "label"
+  name: "data"
+  type: DATA
+  data_param {
+    source: "mnist-train-leveldb"
+    scale: 0.0039215684
+    batch_size: 100
+  }
+}
+layers {
+  bottom: "data"
+  top: "flatdata"
+  name: "flatdata"
+  type: FLATTEN
+}
+layers {
+  bottom: "data"
+  top: "encode1"
+  name: "encode1"
+  type: INNER_PRODUCT
+  blobs_lr: 1
+  blobs_lr: 1
+  weight_decay: 1
+  weight_decay: 0
+  inner_product_param {
+    num_output: 1000
+    weight_filler {
+      type: "gaussian"
+      std: 1
+      sparse: 15
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layers {
+  bottom: "encode1"
+  top: "encode1neuron"
+  name: "encode1neuron"
+  type: SIGMOID
+}
+layers {
+  bottom: "encode1neuron"
+  top: "encode2"
+  name: "encode2"
+  type: INNER_PRODUCT
+  blobs_lr: 1
+  blobs_lr: 1
+  weight_decay: 1
+  weight_decay: 0
+  inner_product_param {
+    num_output: 500
+    weight_filler {
+      type: "gaussian"
+      std: 1
+      sparse: 15
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layers {
+  bottom: "encode2"
+  top: "encode2neuron"
+  name: "encode2neuron"
+  type: SIGMOID
+}
+layers {
+  bottom: "encode2neuron"
+  top: "encode3"
+  name: "encode3"
+  type: INNER_PRODUCT
+  blobs_lr: 1
+  blobs_lr: 1
+  weight_decay: 1
+  weight_decay: 0
+  inner_product_param {
+    num_output: 250
+    weight_filler {
+      type: "gaussian"
+      std: 1
+      sparse: 15
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layers {
+  bottom: "encode3"
+  top: "encode3neuron"
+  name: "encode3neuron"
+  type: SIGMOID
+}
+layers {
+  bottom: "encode3neuron"
+  top: "encode4"
+  name: "encode4"
+  type: INNER_PRODUCT
+  blobs_lr: 1
+  blobs_lr: 1
+  weight_decay: 1
+  weight_decay: 0
+  inner_product_param {
+    num_output: 30
+    weight_filler {
+      type: "gaussian"
+      std: 1
+      sparse: 15
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layers {
+  bottom: "encode4"
+  top: "decode4"
+  name: "decode4"
+  type: INNER_PRODUCT
+  blobs_lr: 1
+  blobs_lr: 1
+  weight_decay: 1
+  weight_decay: 0
+  inner_product_param {
+    num_output: 250
+    weight_filler {
+      type: "gaussian"
+      std: 1
+      sparse: 15
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layers {
+  bottom: "decode4"
+  top: "decode4neuron"
+  name: "decode4neuron"
+  type: SIGMOID
+}
+layers {
+  bottom: "decode4neuron"
+  top: "decode3"
+  name: "decode3"
+  type: INNER_PRODUCT
+  blobs_lr: 1
+  blobs_lr: 1
+  weight_decay: 1
+  weight_decay: 0
+  inner_product_param {
+    num_output: 500
+    weight_filler {
+      type: "gaussian"
+      std: 1
+      sparse: 15
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layers {
+  bottom: "decode3"
+  top: "decode3neuron"
+  name: "decode3neuron"
+  type: SIGMOID
+}
+layers {
+  bottom: "decode3neuron"
+  top: "decode2"
+  name: "decode2"
+  type: INNER_PRODUCT
+  blobs_lr: 1
+  blobs_lr: 1
+  weight_decay: 1
+  weight_decay: 0
+  inner_product_param {
+    num_output: 1000
+    weight_filler {
+      type: "gaussian"
+      std: 1
+      sparse: 15
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layers {
+  bottom: "decode2"
+  top: "decode2neuron"
+  name: "decode2neuron"
+  type: SIGMOID
+}
+layers {
+  bottom: "decode2neuron"
+  top: "decode1"
+  name: "decode1"
+  type: INNER_PRODUCT
+  blobs_lr: 1
+  blobs_lr: 1
+  weight_decay: 1
+  weight_decay: 0
+  inner_product_param {
+    num_output: 784
+    weight_filler {
+      type: "gaussian"
+      std: 1
+      sparse: 15
+    }
+    bias_filler {
+      type: "constant"
+      value: 0
+    }
+  }
+}
+layers {
+  bottom: "decode1"
+  bottom: "flatdata"
+  name: "loss"
+  type: SIGMOID_CROSS_ENTROPY_LOSS
+}
diff --git a/examples/mnist/train_mnist_autoencoder.sh b/examples/mnist/train_mnist_autoencoder.sh
new file mode 100755
index 00000000..af2245e0
--- /dev/null
+++ b/examples/mnist/train_mnist_autoencoder.sh
@@ -0,0 +1,4 @@
+#!/bin/bash
+TOOLS=../../build/tools
+
+GLOG_logtostderr=1 $TOOLS/train_net.bin mnist_autoencoder_solver.prototxt