SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2010.00979 · 2020

BOSS: Bayesian Optimization over String Spaces

arXiv · PDF · Open in the Atlas

Code that ran

8 of 9 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 8 have no contract verdict; 0 failures of our instrument; 1 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2010.00979.json · badge: /api/ran/2010.00979.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
henrymoss/BOSS canonical (classifier label)
classifier basis: census
8 ran · 1 no recorded run
FunctionStatusWhere it lives
choice Ran
ran on a synthesized input
henrymoss/BOSS/boss/code/CFG/CFG.py
code served (permissive licence) · get_code("188e88957d3e32bd")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned str. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

np.random.seed(0)

ARGS = {
    "options": ["apple", "banana", "cherry"],
    "probs": [0.2, 0.3, 0.5]
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
count_codons Ran
ran on a synthesized input
henrymoss/BOSS/boss/code/emukit_models/emukit_bio_features_model.py
code served (permissive licence) · get_code("484b51c1fa63a636")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned defaultdict. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import collections

def count_codons(seq):
    codon_table = {'ttt': 'F', 'ttc': 'F', 'tta': 'L', 'ttg': 'L',
                   'tct': 'S', 'tcc': 'S', 'tca': 'S', 'tcg': 'S',
                   'tat': 'Y', 'tac': 'Y', 'taa': '*', 'tag': '*',
                   'tgt': 'C', 'tgc': 'C', 'tga': '*', 'tgg': 'W',
                   'ctt': 'L', 'ctc': 'L', 'cta': 'L', 'ctg': 'L',
                   'cct': 'P', 'ccc': 'P', 'cca': 'P', 'ccg': 'P',
                   'cat': 'H', 'cac': 'H', 'caa': 'Q', 'cag': 'Q',
                   'cgt': 'R', 'cgc': 'R', 'cga': 'R', 'cgg': 'R',
                   'att': 'I', 'atc': 'I', 'ata': 'I', 'atg': 'M',
                   'act': 'T', 'acc': 'T', 'aca': 'T', 'acg': 'T',
                   'aat': 'N', 'aac': 'N', 'aaa': 'K', 'aag': 'K',
                   'agt': 'S', 'agc': 'S', 'aga': 'R', 'agg': 'R',
                   'gtt': 'V', 'gtc': 'V', 'gta': 'V', 'gtg': 'V',
                   'gct': 'A', 'gcc': 'A', 'gca': 'A', 'gcg': 'A',
                   'gat': 'D', 'gac': 'D', 'gaa': 'E', 'gag': 'E',
                   'ggt': 'G', 'ggc': 'G', 'gga': 'G', 'ggg': 'G'}
    counts = collections.defaultdict(int)
    for codon in codon_table.keys():
        counts[codon] = 0
    for i in range(0, len(seq) - len(seq) % 3, 3):
        codon = seq[i:i + 3]
        counts[codon] += 1
    return counts

seq = 'atgatgatgtttcccaaaggg'

ARGS = {
    'seq': seq
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type defaultdict — a defaultdict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a defaultdict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
count_number_basis Ran
ran on a synthesized input
henrymoss/BOSS/boss/code/emukit_models/emukit_bio_features_model.py
code served (permissive licence) · get_code("9232fb37be581cf9")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned int, ndim 0, dtype family int, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)

# The function count_number_basis takes a sequence (seq) parameter
# and returns the length of that sequence using len(seq)
# So seq needs to be any object that supports len()

# Create a simple sequence - could be a list, tuple, or any iterable with len
seq = [1, 2, 3, 4, 5]

ARGS = {
    "seq": seq
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type int, ndim 0, dtype family int, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
gc_content Ran
ran on a synthesized input
henrymoss/BOSS/boss/code/emukit_models/emukit_bio_features_model.py
code served (permissive licence) · get_code("b0743369c99e1e3e")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned float, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

# Analyze the function to understand what it needs
# gc_content(seq, loc=0) expects:
# - seq: a string-like object with methods .count() and slicing support
# - loc: an integer (default 0)

# The function:
# 1. Slices seq based on loc value (0, 1, 2, or 3)
# 2. Counts 'g' and 'c' characters in the (possibly sliced) seq
# 3. Divides by the total length using count_number_basis(seq)

# seq should be a DNA sequence string containing 'g', 'c', and other bases

ARGS = {
    "seq": "gctagctagcta",  # A simple DNA sequence string with g, c, t, a
    "loc": 0  # Default location parameter
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
map_to_feature_space Ran
ran on a synthesized input
henrymoss/BOSS/boss/code/emukit_models/emukit_bow_model.py
code served (permissive licence) · get_code("fcb870b56796f525")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 2, dtype family float, shape (2, 3) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import re
import numpy as np

# Analyze the function:
# - x: needs to be an array-like with shape (batch_size, ...) where x[i][0] is a string
# - features: a list of strings (patterns) to search for

# Create sample data
batch_size = 2
x = np.array([
    ["hello world"],
    ["foo bar baz"]
], dtype=object)

features = ["o", "a", "l"]

ARGS = {
    "x": x,
    "features": features
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 2, dtype family float, shape (2, 3) — shape (2, 3) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
map_to_feature_space Ran
ran on a synthesized input
henrymoss/BOSS/boss/code/emukit_models/emukit_linear_model.py
code served (permissive licence) · get_code("8a64da8f8aaa442c")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 2, dtype family float, shape (2, 12) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Analyze the function:
# - x: iterable of rows, each row is a tuple/list where row[0] is a string with space-separated characters
# - encoding: dict mapping characters to lists of numbers (feature vectors)

# Create a small encoding dictionary
# Map single characters to small feature vectors
encoding = {
    'a': [0.1, 0.2, 0.3, 0.4],
    'b': [0.5, 0.6, 0.7, 0.8],
    'c': [0.9, 1.0, 1.1, 1.2],
    'd': [1.3, 1.4, 1.5, 1.6]
}

# Create x as a list of rows
# Each row is a tuple where the first element is a space-separated string of characters
x = [
    ('a b c',),
    ('d a b',)
]

ARGS = {
    'x': x,
    'encoding': encoding
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 2, dtype family float, shape (2, 12) — shape (2, 12) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
rand_subtree Ran
ran on a synthesized input
henrymoss/BOSS/boss/code/optimizers/GrammarGeneticAlgorithmAcquisitionOptimizer.py
code served (permissive licence) · get_code("9c753533b7f39f53")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

class Grammar:
    def __init__(self):
        self.swappable_nonterminals = {'NP', 'VP', 'S', 'ADJ'}

tree = "(S (NP (DET the) (N cat)) (VP (V sat)))"
grammar = Grammar()

ARGS = {
    'tree': tree,
    'grammar': grammar
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
rand_subtree_fixed_head Ran
ran on a synthesized input
henrymoss/BOSS/boss/code/optimizers/GrammarGeneticAlgorithmAcquisitionOptimizer.py
code served (permissive licence) · get_code("b6df4d6267e817e3")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned int, ndim 0, dtype family int, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Analyze the function:
# - tree: a string that will be split by spaces
# - head_node: a string used for comparison with split_tree[i][1:]

# Create a simple tree string where elements have heads (indicated by first character)
# The function looks for elements where tree.split(' ')[i][1:] == head_node
tree = "xA yB zA wC vA"

# head_node should match some of the suffixes (after first character) in the split tree
head_node = "A"

ARGS = {
    "tree": tree,
    "head_node": head_node
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type int, ndim 0, dtype family int, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
unparse Not yet run
unverified (harvested, no recorded run)
henrymoss/BOSS/boss/code/optimizers/GrammarGeneticAlgorithmAcquisitionOptimizer.py
code served (permissive licence) · get_code("a194776a5aa6f109")
How it ran

no invocation recorded.

Also present in other papers' code

Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 9 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

This article develops a Bayesian optimization (BO) method which acts directly over raw strings, proposing the first uses of string kernels and genetic algorithms within BO loops. Recent applications of BO over strings have been hindered by the need to map inputs into a smooth and unconstrained latent space. Learning this projection is computationally and data-intensive. Our approach instead builds a powerful Gaussian process surrogate model based on string kernels, naturally supporting variable length inputs, and performs efficient acquisition function maximization for spaces with syntactical constraints. Experiments demonstrate considerably improved optimization over existing approaches across a broad range of constraints, including the popular setting where syntax is governed by a context-free grammar.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2010.00979")
get_code_for_paper("2010.00979")
have("2010.00979")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2010.00979.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2010.00979.svg)](https://syntology.ai/paper/2010.00979)

Connect an agent — have() is free.