A python module for exploring look and say sequences in the spirit of John H Conway.
The following assumes familiarity with the terminology and notation for look and sequences introduced by Conway in his delightful article The Weird and Wonderful Chemistry of Audioactive Decay.
This module can be used to recover several of Conway’s results on standard look and say sequences. Additionally, it can be used to discover new results on various nonstandard look and say sequences. For an introduction to nonstandard look and say sequences, see the notes here.
pip install look-and-say
Full documentation for the module can be found here.
You can read more details about all the objects and methods in the look_and_say
module here. The following example sessions skip those details and get straight to some results. More example sessions can be found here.
The following session illustrates how the look_and_say
module can be used to recover some of Conway’s results.
>>> from look_and_say import *
>>>
>>> # The default LookAndSay object uses the standard decimal number system:
... decimal = LookAndSay()
>>>
>>> # Perform the fundamental look and say operation:
... decimal.say_what_you_see('1222111')
'113231'
>>>
>>> # Generate a look and say sequence
... decimal.generate_sequence(seed='1', terms=5)
>>> decimal.get_sequence()
['1', '11', '21', '1211', '111221']
>>>
>>> # Use Conway's splitting theorem to search for all the elements
... # in the look and say sequence generated from the seed '1'.
... # This will generate Conway's 92 common elements:
... chem = Chemistry(decimal)
>>> chem.generate_elements('1')
>>> chem.get_elements()
[H, He, Li, Be, B, C, N, O, F, Ne, Na, Mg, Al, Si, P, S, Cl, Ar, K, Ca, Sc, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Ga, Ge, As, Se, Br, Kr, Rb, Sr, Y, Zr, Nb, Mo, Tc, Ru, Rh, Pd, Ag, Cd, In, Sn, Sb, Te, I, Xe, Cs, Ba, La, Ce, Pr, Nd, Pm, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu, Hf, Ta, W, Re, Os, Ir, Pt, Au, Hg, Tl, Pb, Bi, Po, At, Rn, Fr, Ra, Ac, Th, Pa, U]
>>>
>>> # The periodic table is a dictionary holding the chemical properties of each element.
... pt = chem.get_periodic_table(abundance_sum=10**6)
>>> print('Hydrogen:', pt['H'])
Hydrogen: {'string': '22', 'abundance': 91790.383216, 'decay': [H]}
>>> print('Thulium:', pt['Tm'])
Thulium: {'string': '11131221133112', 'abundance': 1204.9083841, 'decay': [Er, Ca, Co]}
>>>
>>> # Conway's constant can be found as the dominant (i.e. maximal real) eigenvalue of the decay matrix:
... chem.get_dom_eigenvalue()
1.3035772690342982
>>> # Conway's constant is root of the degree 71 factor of the characteristic polynomial:
... chem.get_char_poly()
lambda**18*(lambda - 1)**2*(lambda + 1)*(lambda**71 - lambda**69 - 2*lambda**68 - lambda**67 + 2*lambda**66 + 2*lambda**65 + lambda**64 - lambda**63 - lambda**62 - lambda**61 - lambda**60 - lambda**59 + 2*lambda**58 + 5*lambda**57 + 3*lambda**56 - 2*lambda**55 - 10*lambda**54 - 3*lambda**53 - 2*lambda**52 + 6*lambda**51 + 6*lambda**50 + lambda**49 + 9*lambda**48 - 3*lambda**47 - 7*lambda**46 - 8*lambda**45 - 8*lambda**44 + 10*lambda**43 + 6*lambda**42 + 8*lambda**41 - 5*lambda**40 - 12*lambda**39 + 7*lambda**38 - 7*lambda**37 + 7*lambda**36 + lambda**35 - 3*lambda**34 + 10*lambda**33 + lambda**32 - 6*lambda**31 - 2*lambda**30 - 10*lambda**29 - 3*lambda**28 + 2*lambda**27 + 9*lambda**26 - 3*lambda**25 + 14*lambda**24 - 8*lambda**23 - 7*lambda**21 + 9*lambda**20 + 3*lambda**19 - 4*lambda**18 - 10*lambda**17 - 7*lambda**16 + 12*lambda**15 + 7*lambda**14 + 2*lambda**13 - 12*lambda**12 - 4*lambda**11 - 2*lambda**10 + 5*lambda**9 + lambda**7 - 7*lambda**6 + 7*lambda**5 - 4*lambda**4 + 12*lambda**3 - 6*lambda**2 + 3*lambda - 6)
The following session shows how to use the module to explore a nonstandard look and say sequence. We use the binary number system known as Gray code to generate the sequence. The corresponding LookAndSay object depends on the say function which converts a positive integer into its Gray code.
>>> from look_and_say import *
>>>
>>> # Define the "say function"
... def gray(num):
... '''Returns the binary Gray code of an integer from 1 to 7.'''
... assert num < 8, "This say function can only count to 7."
... gray_code = {1:'1', 2:'11', 3:'10', 4:'110', 5:'111', 6:'101', 7:'100'}
... return gray_code[num]
...
>>> # Create the LookAndSay object and generate a look and say sequence
... gray_ls = LookAndSay(gray)
>>> gray_ls.generate_sequence(seed='0', terms=6)
>>> gray_ls.get_sequence()
['0', '10', '1110', '10110', '111011110', '10110110110']
>>>
>>> # Use a BinaryChemistry object to determine the chemical properties
... gray_chem = BinaryChemistry(gray_ls)
>>> gray_chem.generate_elements('0')
>>> gray_chem.print_periodic_table()
element string abundance decay
E1 10 0.0 [E3]
E2 110 58.5786438 [E4]
E3 1110 0.0 [E1, E2]
E4 11110 41.4213562 [E2, E2]
>>>
>>> # The dominant eigenvalue of the decay matrix gives the long term
... # growth rate of look and say sequences.
... gray_chem.get_dom_eigenvalue()
1.4142135623730958
>>>
>>> # The growth rate is the maximal real root of the characteristic polynomial
... gray_chem.get_char_poly()
(lambda - 1)*(lambda + 1)*(lambda**2 - 2)
The following session illustrates how to use the module to explore look and say sequences using the standard ternary number system (i.e. using base 3 with digits 0, 1, and 2). The results are similar to those discussed here.
To construct the corresponding LookAndSay object requires a say function which, in this case, is a function that converts an integer to it’s standard ternary representation. The construction of the Chemistry object requires both the LookAndSay object as well as a splitting function which determines when the terms of the look and say sequences split. In this case, one can show that the terms will always split after a run of 0’s as well as between a 2 (on the left) and either a 10 or a 1110 (on the right). We use a SplitFuncFactory object to create the splitting function.
>>> from look_and_say import *
>>>
>>> # Define a "say function"
... def ternary(num):
... '''Returns the ternary representation of a nonnegative integer'''
... if num < 3:
... return str(num)
... return ternary(num // 3) + str(num % 3)
...
>>> # Use the Split Function Factory to create a split function:
... sff = SplitFuncFactory()
>>> sff.declare_split_after('0')
>>> sff.declare_splitting_pairs(('2', '1110'), ('2', '10'))
>>> split = sff.get_split()
>>>
>>> # Instantiate the LookAndSay and Chemistry objects:
... ternary_ls = LookAndSay(ternary)
>>> ternary_chem = Chemistry(ternary_ls, split)
>>>
>>> # Generate elements and order them according to relative abundances:
... ternary_chem.generate_elements('0', '1', '2')
>>> ternary_chem.order_elements('abundance')
>>>
>>> # Print chemical properties:
... ternary_chem.print_periodic_table()
element string abundance decay
E1 10 18.5037375 [E4]
E2 22110 13.9680582 [E5]
E3 2110 13.9680582 [E6]
E4 1110 13.9680582 [E1, E7]
E5 222110 10.5441752 [E1, E2]
E6 122110 10.5441752 [E8]
E7 110 10.5441752 [E3]
E8 11222110 7.9595623 [E3, E2]
E9 222112 0.0 [E1, E10]
E10 22112 0.0 [E9]
E11 212221 0.0 [E16, E4, E13]
E12 2112 0.0 [E14]
E13 211 0.0 [E15]
E14 122112 0.0 [E17]
E15 1221 0.0 [E18]
E16 12 0.0 [E20]
E17 11222112 0.0 [E3, E10]
E18 112211 0.0 [E11]
E19 112 0.0 [E12]
E20 1112 0.0 [E1, E19]
>>>
>>> # Show the dominant eigenvalue which gives the generic growth rate of the look and say sequences.
... print(ternary_chem.get_dom_eigenvalue())
1.3247179572447458
>>>
>>> # The characteristic polynomial formatted in latex:
... print(ternary_chem.get_char_poly(latex=True))
\lambda^{6} \left(\lambda - 1\right)^{2} \left(\lambda + 1\right)^{2} \left(\lambda^{2} + 1\right) \left(\lambda^{3} - \lambda - 1\right) \left(\lambda^{5} - \lambda^{3} + 1\right)
Stuttering look and say sequences and a challenger to Conway’s most complicated algebraic number from the silliest source (2022) preprint on the arXiv.
D3 visualizations of Conway’s chemistry:
A D3 force graph showing the decay of Conway’s elements.
A D3 collapsable tree showing the descendants of Methuselum.