This extension provides a new data type and some functions to deal with linear codes.
A collection of questions created using this extension is available to reuse.
This extension adds two new JME data types, Numbas.jme.types.codeword and Numbas.jme.types.code.
Create a codeword from a list or vector of digits, in Z_{field_size}. For example, codeword([1,1,0,1],2)
Create a codeword from a string of digits, in Z_{field_size}. For example, codeword("11001",2)
The zero word of the given length in the field Z_{field_size}.
Is word a zero word (are all its digits 0)?
A LaTeX rendering of the given codeword.
Add two codewords.
Subtract one codeword from another.
Multiply a codeword by a scalar.
nth digit of codeword.
List of mth to nth digits of codeword.
Hamming weight of the word.
Generate a list of all codewords of given length in Z_{field_size}.
Pick a random word of given length in Z_{field_size}.
A random linear combination of the given words (from the field Z_p), i.e. a_0*w_0 + a_1*w_1 + ... + a_n*w_n where the a_i are elements of the field Z_p.
Returns the set of codewords generated by the given list of basis codewords.
Can all of the given words be written as a linear combination of the words in basis?
Are the given codewords linearly independent? (it checks that the set generated by the basis contains (field_size)^(number of basis words) words)
Generate the coset containing the given word, with respect to the given generating set.
Generate the Slepian array corresponding to the given basis set. Each row in the result is a coset, sorted by weight.
Is the given word a coset leader in its coset? That is, does it have the minimum weight?
A minimal set of generators for the linear code generated by the given words. If the words are linearly independent, you'll get the same number of words back, otherwise you'll get fewer.
Create a parity check matrix for the given generating set, by putting it in reduced row-echelon form I_n|A and returning -A_transpose|I_(m).
Create a parity check matrix for the given generating set, with columns in lexicographic order.
Create a parity check matrix for the Hamming code Ham_p(r). (p must be prime)
Create a generating matrix for the Hamming code Ham_p(r). (p must be prime)
Compute the syndrome of the given word with respect to the given parity check matrix.
Put the given list of words (interpreted as a matrix) into reduced row-echelon form.
Returns a matrix whose rows are the given codewords.
Encode word using Hamming's square code.
Decode (and correct up to one error in) word using Hamming's square code.
Encode word using the general Hamming code.
Decode (and correct up to one error in) word using the general Hamming code.
Hamming distance between two codewords.
All words within given Hamming distance of the given codeword.
All words at given Hamming distance from the given codeword.
Concatenate a list of codewords into one codeword.
Length of a codeword.
Introduce an error (change a digit) at the given position. Returns a new codeword.
Returns a LaTeX check-array for the given word, from a Hamming square code.
String representation of a codeword.
LaTeX representation of a codeword.
Create a code from a complete set of codewords.
All words belonging to the given code.
Minimum Hamming distance between words in code.
A code's information rate
Number of words in code.
nth word in code. (in the order used when you created the code object)
List of the mth to nth words in code.
Permute the positions of the digits in code's words, following the given order.
Permute the symbols in code's words, following the given order. symbols is a list giving a permutation for each digit. That is, symbols[i][j] says what symbol j should change to when in position i.
Can code2 be obtained from code1 by a combination of positional and symbolic permutations?
If code1 is equivalent to code2, find a combination of positional and symbolic permutations that maps code1 to code2.
Returns a dictionary with keys "positional_permutation" and "symbolic_permutation" when the codes are equivalent, and nothing if they're not.
If code1 is positionally equivalent to code2, find a positional permutation that maps code1 to code2.
Returns a list representing a permutation of codeword digits when the codes are positionally equivalent, and nothing if they're not.
If code1 is symbolically equivalent to code2, find a symbolic permutation that maps code1 to code2.
Returns a list representing a permutation of the symbols for each codeword digits when the codes are symbolically equivalent, and nothing if they're not.
Is there an isomorphism between the Hamming distance tables of code1 and code2?
Returns true if there is a permutation of the rows and columns of the Hamming distance table of code1 that produces the table for code2.
Hamming bound on the maximum number of codewords in a code with the given parameters.
Singleton bound on the maximum number of codewords in a code with the given parameters.
Gilbert-Varshamov bound on the minimum number of codewords in a code with the given parameters.
Everything lives under Numbas.extensions.codewords, which I'll omit from now on.
Create a codeword with the given digits, belonging to the field Z_{field_size}.
String representation of the word.
LaTeX representation of the word.
JME representation of the word.
Is the word zero (are all its digits 0)?
Is this word the same as word2?
Return a new word which is the sum of this word and word2.
Subtract word2 from this word (returns a new codeword object).
Negate this word: w.add(w.negate()) = 0.
Scale this word by n - multiply every digit by n.
Hamming weight of this word - number of non-zero digits.
LaTeX rendering of a Hamming square code check array for this word (only makes sense if this word is a 9-digit binary word)
Find all words within radius Hamming distance of this word.
Find all words with Hamming distance exactly radius from this word.
Create a codeword object from a string representation, e.g. Codeword.fromString("01001",2).
Comparison function to sort codewords lexicographically.
Equivalent to w1.eq(w2).
Create a zero word with the given length in the field Z_{field_size}.
Get all words of the given length in the field Z_{field_size}.
Get a random word of the given length in the field Z_{field_size}.
A random linear combination of the given words (from the field Z_p), i.e. a_0*w_0 + a_1*w_1 + ... + a_n*w_n where the a_i are elements of the field Z_p.
Get all words generated by the given basis set.
Can all of the given words be written as a linear combination of the words in basis?
Are all of the given words linearly independent of each other?
Words are linearly independent if l1*w1 + l2*w2 + .. +ln*wn = 0 has no solution other than l1=l2=...=ln=0.
Generate the coset containing the given word, with respect to the given generating set.
Generate the Slepian array corresponding to the given basis set. Each row in the result is a coset, sorted by weight.
Is the given word a coset leader in its coset? That is, does it have the minimum weight?
Hamming distance between two words.
Put the given list of words (interpreted as a matrix) into reduced row-echelon form by reordering and taking linear combinations.
A minimal set of generators for the linear code generated by the given words. If the words are linearly independent, you'll get the same number of words back, otherwise you'll get fewer.
A parity check matrix for the given generating set (which should be linearly independent)
A parity check matrix for the given generating set, with columns in lexicographic order.
Encode word using Hamming's square code.
Decode (and correct up to one error in) word using Hamming's square code.
Encode word using the general Hamming code.
Decode (and correct up to one error in) word using the general Hamming code.
Compute the syndrome of the given word with respect to the given parity check matrix.
Create a parity check matrix for the Hamming code Ham_p(r). (p must be prime)
Create a generating matrix for the Hamming code Ham_p(r). (p must be prime)
Ham_p(r) is the p-ary Hamming code whose PCM has r rows
A wrapper object for the code containing the given words.
String representation of the code - list all its words.
LaTeX representation of the code.
JME representation of the code.
Is this code the same as code2? True if they have exactly the same words.
Does this code contain word?
If this code is equivalent to code2, return an object {positional_permutation, symbolic_permutation}. If not, return null.
Two codes are equivalent if we can get from one to the other by a combination of positional and symbolic permutations.
Is this code equivalent to code2 - can you get from one to the other by performing positional and symbolic permutations on the digits of the codewords?
The minimum Hamming distance between any pair of words in this code.
Information rate of the code: log(number of words)/(log(2)*word length)
Perform a positional permutation on the words in the code. order is a list, where column i is sent to order[i]. Returns a new Code object.
Perform a symbolic permutation on the words in the code. symbols is a list of permutations for each digit, where symbol j in position i is changed to symbols[i][j]. Returns a new Code object.
Hamming bound on the maximum number of codewords in a code with the given parameters.
Singleton bound on the maximum number of codewords in a code with the given parameters.
Gilbert-Varshamov bound on the minimum number of codewords in a code with the given parameters.
To be used on a "match text pattern" part, where the student's answer is a list of codewords separated by commas.
Checks that the student's answer is valid, and parses it to a list of Codeword objects in the given field, then calls fn(words). In fn, this refers to the given part object.
Validate a part marked by the above function. Shows a warning message if the student's answer can't be interpreted as a list of codewords separated by commas.