Check that no common problems in two datasets Parameters: - ds1 -- 1-st dataset as dictionary of its problems dictio

Make convolutional model Parameters: - convolutions -- specification of convolutional layers as l

Compute l1 and l2 regulizer Parameters: - regular -- regularizer as a pair (l1, l2) or None if no reg

Make batched TF dataset from samples generator, Samples are defined with list of sequences Drop samples to make the dataset t

Compute loss function as function of the argument Parameters: - loss -- loss function as a string

* Allocates a fresh unused token from the token pool. */

Cluster samples into n clusters Parameters: - n -- number of clusters

Print memory usage Parameter: - point -- identifier of print: string or number

Make raw dataset having length multiple to batch size Samples are defined with list of sequences Parameters: - batch

Read problems of the dataset Parameters: -fname -- file name Returns: - initialized dictionary * key - problem name

Read samples of dataset Parameters; - cvs_file -- csv file defining dataset - problems -- dictionary of problems of the dataset

Printing report of dataset statistics Parameters: - title -- title of the report

Report classification of set of classes Parameters: - classes -- classes to report - f -- file to write repor

Report errors in dissimilarity detection of problem solutions Parameters: - tests -- tests to report

Print report on accuracy of dissimilarity detection of solutions of pairs of problems Parameters: - tests -- list of

Write down accuracy of similarity detection of two solutions of same problem Parameters: - sim_accuracy -- list of

Distance from sample to set of vectors

Undoes the push action but only if there is some content.

Resets the token buffer cursor.

Tokenize source code file Parameters: - in_fn -- name of input file with code to tokenize Returns: - either

Update statistics of problem solutions usage Parameters: - problem -- problem name - problem_stat -- Statistic

Assume input is UTF-8 encoded Unicode code points. Will read 1 code point and return that as int. The number of bytes read is len and they are p

Write down information on classification predictions Parameters: - predictions - list of predictions each of which is

Compute and write problem distribution in samples The distribution is computed only for problems used for constructing sampl

Write report on classification and misclassification

Write dataset samples in csv format to report test, train, validation datset Parameters: - samples -- similarity sam

Escape token for output as CSV string.

Escape token for output as JSON string.

Escape token for output as XML text.

Escape token for output as XML text.

Initialize object of classification dataset Parameters: - ds -- object of loaded dataset - start_idx -

Initialize tokenizer object Parameters: - ds -- Directory to write down tokenized solutions - data -- Direc

Fill in dataset numpy array with one solution using categorical coding Parameters: - ds -- numpy array of to

Fill in similarity dataset with one solution Parameters: - ds -- numpy array of to fill in with one part

Register problem solution Parameters: - reg_dict -- dictionary where to register Key: problem name. V

Print registered problem solutions and their lengths Parameters: - reg_dict -- Dictionary with registered problems

Add to the dnn a dense layer with classifier Classifier is either softmax for more than 2 classes or sigmoid for 2 classes

Add to the dnn a stack of convolutional layers Parameters: - dnn -- dnn to buld - convolutions -- specific

Add a stack of dense layers to the dnn Parameters: - dnn -- dnn to buld - dense -- specification of de

Add embedding layer to the DNN Parameters: - dnn -- dnn to buld - trainable -- flag defining that

Check loaded dataset with respect to source datset directory Parameters: - ds -- dictionary defining dataset to be checked - ds_

Compute classification function Parameters: - labels -- number of labels (classes)

Make sequential DNN building block of stack of convolutional layers Parameters: - convolutions -- specification of

Decode the coordinates attribute. */

Make DNN with dense (fully connected) layers Parameters: - dense -- list of hidden layers widths for

Determine correct coordinates for all tokens (leaf nodes) and adjust all labels by removing leading and trailing white-space. */

Make TF dataset from samples generator, defined with list of sequences Parameters: - samples -- list of samples (s

Complete debugging

Enter a parse tree produced by {@link CymbolParser#block}. @param ctx the parse tree

Enter a parse tree produced by {@link CymbolParser#expr}. @param ctx the parse tree

Enter a parse tree produced by {@link CymbolParser#exprList}. @param ctx the parse tree

Enter a parse tree produced by {@link CymbolParser#file}. @param ctx the parse tree

Enter a parse tree produced by {@link CymbolParser#formalParameter}. @param ctx the parse tree

Enter a parse tree produced by {@link CymbolParser#formalParameters}. @param ctx the parse tree

Enter a parse tree produced by {@link CymbolParser#functionDecl}. @param ctx the parse tree

Enter a parse tree produced by {@link CymbolParser#stat}. @param ctx the parse tree

Enter a parse tree produced by {@link CymbolParser#type}. @param ctx the parse tree

Enter a parse tree produced by {@link CymbolParser#varDecl}. @param ctx the parse tree

Exit a parse tree produced by {@link CymbolParser#block}. @param ctx the parse tree

Exit a parse tree produced by {@link CymbolParser#expr}. @param ctx the parse tree

Exit a parse tree produced by {@link CymbolParser#exprList}. @param ctx the parse tree

Exit a parse tree produced by {@link CymbolParser#file}. @param ctx the parse tree

Exit a parse tree produced by {@link CymbolParser#formalParameter}. @param ctx the parse tree

Exit a parse tree produced by {@link CymbolParser#formalParameters}. @param ctx the parse tree

Exit a parse tree produced by {@link CymbolParser#functionDecl}. @param ctx the parse tree

Exit a parse tree produced by {@link CymbolParser#stat}. @param ctx the parse tree

Exit a parse tree produced by {@link CymbolParser#type}. @param ctx the parse tree

Exit a parse tree produced by {@link CymbolParser#varDecl}. @param ctx the parse tree

Make a stack of dense layers with classifier Classifier is either softmax for more than 2 classes or sigmoid for 2 classes

Checks whether token is a string or primitive. */

Get list of errors in dissimilarity detection of problem solutions Parameters: - tests -- list of accuracy of dissi