Add partial support for HDF5Data layer in networks

digits/dataset/tasks/create_db.py

@@ -49,6 +49,12 @@ def __init__(self, input_file, db_name, backend, image_dims, **kwargs):
  self.input_file = input_file
  self.db_name = db_name
  self.backend = backend
+ if backend == 'hdf5':
+ # store a textfile that contains a path to the hdf5 file
+ self.textfile = '%s.txt' % self.db_name
+ with open(self.path(self.textfile), 'w') as outfile:
+ # XXX - this works because the model job will be in an adjacent folder
+ outfile.write('../%s/%s' % (self.job_id, self.db_name))
  self.image_dims = image_dims
  if image_dims[2] == 3:
  self.image_channel_order = 'BGR'

digits/model/tasks/caffe_train.py

@@ -140,8 +140,8 @@ def save_files_classification(self):
  loss_layers = []
  accuracy_layers = []
  for layer in self.network.layer:
- assert layer.type not in ['MemoryData', 'HDF5Data', 'ImageData'], 'unsupported data layer type'
- if layer.type == 'Data':
+ assert layer.type not in ['DummyData', 'ImageData', 'MemoryData', 'WindowData'], 'unsupported data layer type'
+ if layer.type in ['Data', 'HDF5Data']:
  for rule in layer.include:
  if rule.phase == caffe_pb2.TRAIN:
  assert train_data_layer is None, 'cannot specify two train data layers'
@@ -192,13 +192,22 @@ def save_files_classification(self):
 
  train_val_network = caffe_pb2.NetParameter()
 
+ dataset_backend = self.dataset.train_db_task().backend
+
  # data layers
  if train_data_layer is not None:
- if train_data_layer.HasField('data_param'):
+ if dataset_backend == 'lmdb':
+ assert train_data_layer.type == 'Data', 'expecting a Data layer'
+ elif dataset_backend == 'hdf5':
+ assert train_data_layer.type == 'HDF5Data', 'expecting an HDF5Data layer'
+ if dataset_backend == 'lmdb' and train_data_layer.HasField('data_param'):
  assert not train_data_layer.data_param.HasField('source'), "don't set the data_param.source"
  assert not train_data_layer.data_param.HasField('backend'), "don't set the data_param.backend"
+ if dataset_backend == 'hdf5' and train_data_layer.HasField('hdf5_data_param'):
+ assert not train_data_layer.hdf5_data_param.HasField('source'), "don't set the hdf5_data_param.source"
  max_crop_size = min(self.dataset.image_dims[0], self.dataset.image_dims[1])
  if self.crop_size:
+ assert dataset_backend != 'hdf5', 'HDF5Data layer does not support cropping'
  assert self.crop_size <= max_crop_size, 'crop_size is larger than the image size'
  train_data_layer.transform_param.crop_size = self.crop_size
  elif train_data_layer.transform_param.HasField('crop_size'):
@@ -211,36 +220,58 @@ def save_files_classification(self):
  train_val_network.layer.add().CopyFrom(train_data_layer)
  train_data_layer = train_val_network.layer[-1]
  if val_data_layer is not None and has_val_set:
- if val_data_layer.HasField('data_param'):
+ if dataset_backend == 'lmdb':
+ assert val_data_layer.type == 'Data', 'expecting a Data layer'
+ elif dataset_backend == 'hdf5':
+ assert val_data_layer.type == 'HDF5Data', 'expecting an HDF5Data layer'
+ if dataset_backend == 'lmdb' and val_data_layer.HasField('data_param'):
  assert not val_data_layer.data_param.HasField('source'), "don't set the data_param.source"
  assert not val_data_layer.data_param.HasField('backend'), "don't set the data_param.backend"
+ if dataset_backend == 'hdf5' and val_data_layer.HasField('hdf5_data_param'):
+ assert not val_data_layer.hdf5_data_param.HasField('source'), "don't set the hdf5_data_param.source"
  if self.crop_size:
  # use our error checking from the train layer
  val_data_layer.transform_param.crop_size = self.crop_size
  train_val_network.layer.add().CopyFrom(val_data_layer)
  val_data_layer = train_val_network.layer[-1]
  else:
- train_data_layer = train_val_network.layer.add(type = 'Data', name = 'data')
+ layer_type = 'Data'
+ if dataset_backend == 'hdf5':
+ layer_type = 'HDF5Data'
+ train_data_layer = train_val_network.layer.add(type = layer_type, name = 'data')
  train_data_layer.top.append('data')
  train_data_layer.top.append('label')
  train_data_layer.include.add(phase = caffe_pb2.TRAIN)
- train_data_layer.data_param.batch_size = constants.DEFAULT_BATCH_SIZE
+ if dataset_backend == 'lmdb':
+ train_data_layer.data_param.batch_size = constants.DEFAULT_BATCH_SIZE
+ elif dataset_backend == 'hdf5':
+ train_data_layer.hdf5_data_param.batch_size = constants.DEFAULT_BATCH_SIZE
  if self.crop_size:
+ assert dataset_backend != 'hdf5', 'HDF5Data layer does not support cropping'
  train_data_layer.transform_param.crop_size = self.crop_size
  if has_val_set:
- val_data_layer = train_val_network.layer.add(type = 'Data', name = 'data')
+ val_data_layer = train_val_network.layer.add(type = layer_type, name = 'data')
  val_data_layer.top.append('data')
  val_data_layer.top.append('label')
  val_data_layer.include.add(phase = caffe_pb2.TEST)
- val_data_layer.data_param.batch_size = constants.DEFAULT_BATCH_SIZE
+ if dataset_backend == 'lmdb':
+ val_data_layer.data_param.batch_size = constants.DEFAULT_BATCH_SIZE
+ elif dataset_backend == 'hdf5':
+ val_data_layer.hdf5_data_param.batch_size = constants.DEFAULT_BATCH_SIZE
  if self.crop_size:
  val_data_layer.transform_param.crop_size = self.crop_size
- train_data_layer.data_param.source = self.dataset.path(self.dataset.train_db_task().db_name)
- train_data_layer.data_param.backend = caffe_pb2.DataParameter.LMDB
- if val_data_layer is not None and has_val_set:
- val_data_layer.data_param.source = self.dataset.path(self.dataset.val_db_task().db_name)
- val_data_layer.data_param.backend = caffe_pb2.DataParameter.LMDB
+ if dataset_backend == 'lmdb':
+ train_data_layer.data_param.source = self.dataset.path(self.dataset.train_db_task().db_name)
+ train_data_layer.data_param.backend = caffe_pb2.DataParameter.LMDB
+ if val_data_layer is not None and has_val_set:
+ val_data_layer.data_param.source = self.dataset.path(self.dataset.val_db_task().db_name)
+ val_data_layer.data_param.backend = caffe_pb2.DataParameter.LMDB
+ elif dataset_backend == 'hdf5':
+ train_data_layer.hdf5_data_param.source = self.dataset.path(self.dataset.train_db_task().textfile)
+ if val_data_layer is not None and has_val_set:
+ val_data_layer.hdf5_data_param.source = self.dataset.path(self.dataset.val_db_task().textfile)
  if self.use_mean:
+ assert dataset_backend != 'hdf5', 'HDF5Data layer does not support mean subtraction'
  mean_pixel = None
  with open(self.dataset.path(self.dataset.train_db_task().mean_file),'rb') as f:
  blob = caffe_pb2.BlobProto()
@@ -259,14 +290,25 @@ def save_files_classification(self):
  for value in mean_pixel:
  val_data_layer.transform_param.mean_value.append(value)
  if self.batch_size:
- train_data_layer.data_param.batch_size = self.batch_size
- if val_data_layer is not None and has_val_set:
- val_data_layer.data_param.batch_size = self.batch_size
+ if dataset_backend == 'lmdb':
+ train_data_layer.data_param.batch_size = self.batch_size
+ if val_data_layer is not None and has_val_set:
+ val_data_layer.data_param.batch_size = self.batch_size
+ elif dataset_backend == 'hdf5':
+ train_data_layer.hdf5_data_param.batch_size = self.batch_size
+ if val_data_layer is not None and has_val_set:
+ val_data_layer.hdf5_data_param.batch_size = self.batch_size
  else:
- if not train_data_layer.data_param.HasField('batch_size'):
- train_data_layer.data_param.batch_size = constants.DEFAULT_BATCH_SIZE
- if val_data_layer is not None and has_val_set and not val_data_layer.data_param.HasField('batch_size'):
- val_data_layer.data_param.batch_size = constants.DEFAULT_BATCH_SIZE
+ if dataset_backend == 'lmdb':
+ if not train_data_layer.data_param.HasField('batch_size'):
+ train_data_layer.data_param.batch_size = constants.DEFAULT_BATCH_SIZE
+ if val_data_layer is not None and has_val_set and not val_data_layer.data_param.HasField('batch_size'):
+ val_data_layer.data_param.batch_size = constants.DEFAULT_BATCH_SIZE
+ elif dataset_backend == 'hdf5':
+ if not train_data_layer.hdf5_data_param.HasField('batch_size'):
+ train_data_layer.hdf5_data_param.batch_size = constants.DEFAULT_BATCH_SIZE
+ if val_data_layer is not None and has_val_set and not val_data_layer.hdf5_data_param.HasField('batch_size'):
+ val_data_layer.hdf5_data_param.batch_size = constants.DEFAULT_BATCH_SIZE
 
  # hidden layers
  train_val_network.MergeFrom(hidden_layers)

digits/templates/datasets/images/classification/summary.html

@@ -34,7 +34,5 @@ <h4>
  {% endif %}
  </dd>
  {% endfor %}
- <dt>Encoding</dt>
- <dd>{{dataset.train_db_task().encoding}}</dd>
 </dl>