SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1608.06194 · 2016

Understanding molecular representations in machine learning: The role of uniqueness and target similarity

arXiv · PDF · Open in the Atlas

Code that ran

10 of 12 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 10 have no contract verdict; 0 failures of our instrument; 2 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/1608.06194.json · badge: /api/ran/1608.06194.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
binghuang2018/aqml community (classifier label)
classifier basis not recorded on the link
10 ran · 2 no recorded run
FunctionStatusWhere it lives
cmp Ran
ran on a synthesized input
binghuang2018/aqml/cheminfo/algo/amon_mpi.py
code served (permissive licence) · get_code("ae24b3d5a6652184")
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

torch.manual_seed(0)

# The function cmp compares two values a and b
# It uses the < and > operators on both parameters
# So we need two comparable objects

# Simple case: use integers
a = 5
b = 10

ARGS = {
    "a": a,
    "b": b
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
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
generate_cm Ran
ran on a synthesized input
binghuang2018/aqml/cheminfo/algo/amon.py
code served (permissive licence) · get_code("c4079565f813de5a")
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 1, dtype family float, shape (25,) 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
import scipy.spatial.distance as ssd

np.random.seed(0)

# Create valid arguments for generate_cm
# zs: atomic numbers (charges), shape (na,)
# coords: atomic coordinates, shape (na, 3)
# ds: optional distance matrix, shape (na, na)
# inorm: boolean flag
# wz: boolean flag
# rpower: float power for distance term

na = 5
zs = np.array([6, 8, 1, 1, 1])  # Example: C, O, H, H, H
coords = np.random.rand(na, 3) * 10  # Random coordinates, na x 3
ds = ssd.squareform(ssd.pdist(coords))  # Precomputed distance matrix

ARGS = {
    'zs': zs,
    'coords': coords,
    'ds': ds,
    'inorm': False,
    'wz': False,
    'rpower': 1.0
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 1, dtype family float, shape (25,) — shape (25,) 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
get_compl Ran
ran on a synthesized input
binghuang2018/aqml/cheminfo/math.py
code served (permissive licence) · get_code("c1a7b05eca4f9e44")
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 list, ndim 1, dtype family int; 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 torch
import numpy as np

torch.manual_seed(0)

# Analyzing the function:
# - get_compl(a, b) takes two arguments
# - It iterates over elements in 'a'
# - It checks if each element 'ai' is in 'b' using 'not in'
# - It appends elements to a_compl list
# - Both a and b should be iterable collections that support membership testing
# - The elements should be comparable/hashable for the 'in' operator

# Simple approach: use lists of integers or strings
a = [1, 2, 3, 4, 5]
b = [2, 4, 6]

ARGS = {
    "a": a,
    "b": b
}
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 list, ndim 1, dtype family int — a list 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
get_compl_u Ran
ran on a synthesized input
binghuang2018/aqml/cheminfo/math.py
code served (permissive licence) · get_code("a662f1035f9ca7ce")
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 list, ndim 2, dtype family int; 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 torch
import numpy as np

def get_compl_u(a, b):
    a_compl = []
    bu = [set(bi) for bi in b]
    for ai in a:
        if set(ai) not in bu:
            a_compl.append(ai)
    return a_compl

# Construct test arguments
# 'a' should be a list of iterables (to convert to sets)
# 'b' should be a list of iterables (to convert to sets)
# The function finds elements in 'a' whose set representation is not in the set representations of 'b'

a = [
    [1, 2, 3],
    [4, 5, 6],
    [1, 2, 3],
    [7, 8, 9]
]

b = [
    [1, 2, 3],
    [4, 5, 6]
]

ARGS = {
    'a': a,
    'b': b
}
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 list, ndim 2, dtype family int — a list 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
get_index Ran
ran on a synthesized input
binghuang2018/aqml/calculators/mopac.py
code served (permissive licence) · get_code("f2aae5d63f46ee02")
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

# Construct arguments for get_index(lines, pattern)
# The function takes a list of strings and searches for a pattern in each line

lines = [
    "hello world",
    "this is a test",
    "python code here",
    "pattern found here"
]

pattern = "pattern"

ARGS = {
    "lines": lines,
    "pattern": pattern
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
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
get_pl Ran
ran on a synthesized input
binghuang2018/aqml/cheminfo/graph.py
code served (permissive licence) · get_code("2ba2caed4e08cf53")
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 networkx as nx

# Create a simple directed graph
gnx = nx.DiGraph()
gnx.add_edges_from([(0, 1), (1, 2), (2, 3), (0, 2)])

# Define node indices
i = 0
j = 3

# Construct the arguments dictionary
ARGS = {
    "ipt": (gnx, i, j)
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
networkx — 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
is_subset Ran
ran on a synthesized input
binghuang2018/aqml/cheminfo/algo/amon.py
code served (permissive licence) · get_code("3a403e7536e68c98")
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 bool, 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.

import torch

torch.manual_seed(0)

# Based on the function signature and implementation:
# - `a` should be a list (the subset candidate)
# - `b` should be a list of lists (the superset collection)
# The function checks if `a` as a set matches any element in `b` as a set

a = [1, 2]
b = [[2, 4], [2, 1], [3, 9, 10]]

ARGS = {
    "a": a,
    "b": b
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type bool, 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
lowdin Ran
ran on a synthesized input
binghuang2018/aqml/calculators/pscf.py
code served (permissive licence) · get_code("94c1e4a4c652b1fe")
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 (4, 4) 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

# Create a small symmetric positive definite matrix
# This is a valid overlap matrix (S_mm) that would be used in quantum chemistry
s_mm = np.array([
    [2.0, 0.5, 0.1, 0.05],
    [0.5, 1.8, 0.3, 0.1],
    [0.1, 0.3, 1.5, 0.2],
    [0.05, 0.1, 0.2, 1.2]
], dtype=np.complex128)

ARGS = {
    "s_mm": s_mm
}
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 (4, 4) — shape (4, 4) 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
read_output Ran
ran on a synthesized input
binghuang2018/aqml/calculators/turbomole.py
code served (permissive licence) · get_code("a08beb896bc4426a")
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 list, 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 os
import re
import tempfile
import shutil

# Create a temporary directory for test files
test_dir = tempfile.mkdtemp()
original_dir = os.getcwd()
os.chdir(test_dir)

# Create test output files with content matching a pattern
with open('job.001.out', 'w') as f:
    f.write("Processing step 1\n")
    f.write("Result: value_123\n")
    f.write("Status: complete\n")

with open('job.002.out', 'w') as f:
    f.write("Processing step 2\n")
    f.write("Result: value_456\n")

with open('output.out', 'w') as f:
    f.write("Final Result: value_789\n")

with open('ignored_file.txt', 'w') as f:
    f.write("This should be ignored\n")

# Define the regex pattern to extract values
# This pattern captures numbers after "Result: " or "Final Result: "
regex_pattern = r"Result:\s*(\w+)"

ARGS = {
    "regex": regex_pattern
}

# Cleanup will happen after function execution
# os.chdir(original_dir)
# shutil.rmtree(test_dir)
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 list, ndim 1, dtype family float — a list 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
sort_h Ran
ran on a synthesized input
binghuang2018/aqml/cheminfo/algo/amon.py
code served (permissive licence) · get_code("045a61560dea4c27")
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 1, dtype family int, shape (5,) 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

# Construct a small list of atomic numbers (zs parameter)
# This represents a molecule with some heavy atoms and hydrogen atoms
# The function sorts atoms by pushing hydrogens (atomic number 1) to the end
zs = np.array([6, 8, 1, 1, 1], dtype=int)  # Carbon, Oxygen, three Hydrogens

ARGS = {
    "zs": zs
}
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 1, dtype family int, shape (5,) — shape (5,) 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
get_nsheav Not yet run
unverified (harvested, no recorded run)
binghuang2018/aqml/cheminfo/core.py
code served (permissive licence) · get_code("5e588b857d571c5c")
How it ran

no invocation recorded.

products Not yet run
unverified (harvested, no recorded run)
binghuang2018/aqml/cheminfo/math.py
code served (permissive licence) · get_code("e45a23a8d309ce4d")
How it ran

no invocation recorded.

Also present in other papers' code

Of 12 samples lifted from this paper's linked repositories, 12 carry a function identity and 0 do not. Of the 12 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 12 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

The predictive accuracy of Machine Learning (ML) models of molecular properties depends on the choice of the molecular representation. Based on the postulates of quantum mechanics, we introduce a hierarchy of representations which meet uniqueness and target similarity criteria. To systematically control target similarity, we rely on interatomic many body expansions, as implemented in universal force-fields, including Bonding, Angular, and higher order terms (BA). Addition of higher order contributions systematically increases similarity to the true potential energy and predictive accuracy of the resulting ML models. We report numerical evidence for the performance of BAML models trained on molecular properties pre-calculated at electron-correlated and density functional theory level of theory for thousands of small organic molecules. Properties studied include enthalpies and free energies of atomization, heatcapacity, zero-point vibrational energies, dipole-moment, polarizability, HOMO/LUMO energies and gap, ionization potential, electron affinity, and electronic excitations. After training, BAML predicts energies or electronic properties of out-of-sample molecules with unprecedented accuracy and speed.

For agents

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

get_harvested_code_for_paper("1608.06194")
get_code_for_paper("1608.06194")
have("1608.06194")

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

curl https://syntology.ai/api/ran/1608.06194.json

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

[![Syntology run record](https://syntology.ai/api/ran/1608.06194.svg)](https://syntology.ai/paper/1608.06194)

Connect an agent — have() is free.