Converts a N-dimensional array to blob proto. If diff is given, also convert the diff. You need to make sure that arr and diff have the same s

Convert a blob proto to an array. In default, we will just return the data, unless return_diff is True, in which case we will return the diff

A utility function to check the parameters for the data layers.

Check if a given path is a regular file or a path

Define colors for nodes based on the layer type.

Configure crop dimensions and amount of context for cropping. If context is included, make the special input mean for context padding

Extract the crop layer parameters with defaults.

Returns the TestCase object for the test that's currently running, or NULL if no test is running. L < mutex_

Do windowed detection over Selective Search proposals by extracting the crop and warping to the input dimensions of the net.

Draws a caffe net and returns the image string encoded using the given extension. Parameters ---------- caffe_net : a caffe.proto.caf

Creates an image embedded in HTML base64 format.

Controls whether events will be forwarded to listeners_. Set to false in death test child processes.

Yields files in the source tree which are Net prototxts.

Define edge label based on layer type.

Define node label based on layer type. Parameters ---------- layer : caffe_pb2.LayerParameter rankdir : {'LR', 'TB', 'BT'} Di

Get the learning rate multipliers. Get the learning rate multipliers for the given layer. Assumes a Convolution/Deconvolution/InnerProduct la

Get year from log file system timestamp

Parse phase from flags

Get dictionary mapping pooling type number to type name

Create a data structure which represents the `caffe_net`. Parameters ---------- caffe_net : object rankdir : {'LR', 'TB', 'BT'}

Find start time from group of lines

Invert a coord map by de-scaling and un-shifting; this gives the backward mapping for the gradient.

Returns true iff this is NAN (not a number).

Load the next image in a batch.

This code is borrowed from Ross Girshick's FAST-RCNN code (https://github.com/rbgirshick/fast-rcnn). It parses the PASCAL .xml metadata f

Do the file extension and encoding match?

Gets the message associated with the test part.

@brief Input and output blob numbers

Find out the correspondence between layer names and parameter names.

Parse input arguments

Parse log file Returns (train_dict_list, test_dict_list) train_dict_list and test_dict_list are lists of dicts that define the table rows

Output some info regarding the class

Print a simple nicely-aligned table. table must be a list of (equal-length) lists. Columns are space-separated, and as narrow as possible, bu

Usage: caffe_('reset')

Takes an image name, resize it and crop the center square

Save CSV files to output_dir If the input log file is, e.g., caffe.INFO, the names will be caffe.INFO.train and caffe.INFO.test

* @brief Sets the loss associated with a top blob at a given index. */

Set the mean to subtract for centering the data.

STL-style container methods.

Parse command line options and start the server.

Gets the number of successful tests.

Returns the number of the test properties.

Time: benchmark the execution time of a model.

Generate a NetParameter that contains all layers needed to compute this top.

Gets the number of all test parts. This is the sum of the number of successful test parts and the number of failed test parts.