Maps xs to consumers. Returns a dict mapping each element of xs to any of its consumers that are indirectly consumed by ys. Args: ys:

Codepoints not allowed for interchange are: C0 (ASCII) controls: U+0000 to U+001F excluding Space (SP, U+0020), Horizontal Tab (HT, U+0009), Line-Feed

Returns true if 'str' is the start of a structurally valid UTF-8 sequence and is not a surrogate codepoint. Returns false if str.empty() or if str.len

For an array of paths of length count, append them all together, ensuring that the proper path separators are inserted between them.

Extracts the label from a given parser action. If the action is SHIFT, returns -1.

Finds a named component in the registry.

A factory function for creating a UnicodeText from a buffer of UTF-8 data. The new UnicodeText takes ownership of the buffer. (It is an "owner.") Eac

A factory function for creating a UnicodeText from a buffer of UTF-8 data. The new UnicodeText does not take ownership of the buffer. (It is an "alias

Return the MS-SSIM score between `img1` and `img2`. This function implements Multi-Scale Structural Similarity (MS-SSIM) Image Quality Assessment

Pad or trim list to len length. Args: ids: list of ints to pad pad_id: what to pad with length: length to pad or trim to Returns:

Process a single XML file containing a bounding box.

Reads the vocabulary file into a map from token to vector index.

Resets the output files. Must be called once before Write().

Resets the offset vectors required for a single run because we're starting a new matrix of scores.

Restores the model from the given checkpoint path into the session. Args: checkpoint_path: File pathname of the checkpoint. sess:

Returns the special value to represent a root token.

Sets this tag on the token at index.

Sets this tag on the token at index.

Sets up the feature extractor. Note that only top-level functions exist until Setup() is called. This does not take ownership over the context, which

Return the number of elements.

Generates consecutive snippets between start and end tokens. Args: text: a string start_tok: a string denoting the start of snippets en

Evaluate a model in softmax mode. Adds char, word recall and sequence error rate events to the sw summary writer, and returns them as well

Returns the prefix for nested feature functions. This is the prefix of this feature function concatenated with the feature function name.

Returns the tag assigned to a given token.

Returns the tag assigned to a given token.

Train MNIST for a number of steps. Args: mnist_train_file: path of MNIST train data file. mnist_test_file: path of MNIST test data file.

Builds and trains the network. Args: sess: tensorflow session to use. num_actions: number of possible golden actions. feature_sizes: si

Copies a substring of a Unicode text to a string.

Writes the reference and decoded outputs to RKV files. Args: reference: The human (correct) result. decode: The machine-generated res

Cross entropy plus L2 loss on weights and biases of the hidden layers.

Adds an embedding matrix and passes the `features` vector through it.

Add the appropriate loss function. Args: logits: Pre-softmax/logistic fully-connected output shaped to out_dims. height_in: Height

Adds an optimizer with learning rate decay to minimize self.loss. Args: optimizer_type: One of 'GradientDescent', 'AdaGrad', 'Momentum', 'A

Add the fully-connected logits and SoftMax/Logistic output Layer. Args: prev_layer: Output of last layer of main network. out_dims:

Adds the output layer and loss function. Args: prev_layer: Output of last layer of main network. out_dims: Number of output dimen

Restore a checkpoint and decode it. Args: saver: Tensorflow checkpoint saver. sess: Tensorflow session. Returns: If success

Convert id to words and writing results. Args: article: The original article string. abstract: The human (correct) abstract string.

Dump the softmax weights and word embeddings to files. Args: vocab: Vocabulary. Contains vocabulary size and converts word to ids.

Predict next words using the given prefix words. Args: sentence: Sentence words. vocab: Vocabulary. Contains max word chard id length and c

Runs model eval.

Evaluate model perplexity using provided dataset. Args: dataset: LM1BDataset object.

Function to mimic the 'fspecial' gaussian MATLAB function.

Convert a PNG string into an input tensor. We allow for fixed and variable sizes. Does fixed conversion to floats in the range [-1.28, 1.27]. A

Reads the decoder file and initializes self.decoder from it. Args: filename: Name of text file mapping codes to utf8 strings. Raises:

Pads or slices 3-d labels to match logits. Covers the case of 2-d softmax output, when labels is [batch, height, width] and logits is [batch, hei

Parses input_spec and returns the numbers obtained therefrom. Args: input_spec: Specification of the input layer. See Build. Returns: s

Parses the output spec. Args: output_spec: Output layer definition. See Build. Returns: out_dims: 2|1|0 for 2-d, 1-d, 0-d. out_f

conv2d run separately per example, to help compute per-example gradients. Args: input_: tensor containing a minibatch of images / feature m

Builds network input tensor ops for TF Example. Args: filename_queue: Queue of filenames, from tf.train.string_input_producer shape:

Return the Structural Similarity Map between `img1` and `img2`. This function attempts to match the functionality of ssim_index_new.m by Zhou Wan

Predict next words using the given prefix words. Args: prefix_words: Prefix words. vocab: Vocabulary. Contains max word chard id length and

Skips any leading whitespace in the model description. Args: index: Position in model_str to start parsing Returns: end ind

Generates article and abstract text from tf.Example.

Runs model training.

Initialize a MomentsAccountant. Args: total_examples: total number of examples. moment_orders: the order of moments to keep.

Initialize vocabulary. Args: filename: Vocabulary file name.

Helper to create summaries for activations. Creates a summary that provides a histogram of activations. Creates a summary that measure the sparsi

Add histogram summaries to gradients. Note: The summaries are also added to the SUMMARIES collection. Args: grads_and_vars: A list of gradie

Inputs to be fed to the graph.