SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2106.01949 · 2021

Active learning of reactive Bayesian force fields: Application to heterogeneous hydrogen-platinum catalysis dynamics

arXiv · PDF · Open in the Atlas

Code that ran

6 of 8 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, 1 violated it and 5 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/2106.01949.json · badge: /api/ran/2106.01949.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
mir-group/flare_pp canonical (classifier label)
classifier basis: census
5 ran · 2 no recorded run
mir-group/flare canonical (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
FunctionStatusWhere it lives
compute_negative_likelihood Ran
ran on a synthesized input
mir-group/flare_pp/flare_pp/sparse_gp.py
code served (permissive licence) · get_code("d9d15a4d3e08fdb2")
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 Tensor, 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
import torch.nn as nn
import numpy as np

torch.manual_seed(0)

class MockSparseGP(nn.Module):
    def __init__(self):
        super().__init__()
        self.hyperparameters = torch.tensor([1.0, 0.5, 0.1])
        self.log_marginal_likelihood = torch.tensor(0.0)
        self._current_hyperparameters = None
    
    def set_hyperparameters(self, hyperparameters):
        """Set the hyperparameters of the GP."""
        self._current_hyperparameters = hyperparameters
    
    def compute_likelihood(self):
        """Compute the log marginal likelihood based on current hyperparameters."""
        if self._current_hyperparameters is not None:
            self.log_marginal_likelihood = torch.sum(self._current_hyperparameters) - 1.0
        else:
            self.log_marginal_likelihood = torch.tensor(0.5)

sparse_gp = MockSparseGP()
hyperparameters = torch.tensor([1.0, 0.5, 0.1])

ARGS = {
    'hyperparameters': hyperparameters,
    'sparse_gp': sparse_gp,
    'print_vals': False
}
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 Tensor, 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
compute_negative_likelihood_grad Ran
ran on a synthesized input
mir-group/flare_pp/flare_pp/sparse_gp.py
code served (permissive licence) · get_code("913d8524d1362e92")
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. 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 torch
import numpy as np

torch.manual_seed(0)

class MockSparseGP:
    """Mock sparse GP model with required attributes and methods."""
    
    def __init__(self):
        self.hyperparameters = [1.0, 0.5, 2.0]
        self._likelihood_grad = np.array([0.1, -0.2, 0.15])
    
    def compute_likelihood_gradient(self, hyperparameters):
        """Mock method that computes and stores likelihood gradient."""
        # Store the gradient for later retrieval
        self.likelihood_gradient = self._likelihood_grad
        # Return a mock likelihood value
        return 5.0
    
    @property
    def likelihood_gradient(self):
        """Property to store/retrieve the gradient."""
        if not hasattr(self, '_stored_grad'):
            self._stored_grad = self._likelihood_grad
        return self._stored_grad
    
    @likelihood_gradient.setter
    def likelihood_gradient(self, value):
        self._stored_grad = value

sparse_gp = MockSparseGP()
hyperparameters = [1.0, 0.5, 2.0]

ARGS = {
    'hyperparameters': hyperparameters,
    'sparse_gp': sparse_gp,
    'print_vals': False
}
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 tuple — a tuple output has no array shape of its own; none is 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
compute_negative_likelihood_grad_stable Ran
ran on a synthesized input
mir-group/flare_pp/flare_pp/sparse_gp.py
code served (permissive licence) · get_code("d7ec88038d224335")
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. 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 torch
import torch.nn as nn
import numpy as np

torch.manual_seed(0)

class SparseGP(nn.Module):
    def __init__(self):
        super().__init__()
        self.hyperparameters = torch.tensor([1.0, 0.5, 0.2])
        self.likelihood_gradient = None
        self._current_hyperparameters = None
    
    def set_hyperparameters(self, hyperparameters):
        """Set the hyperparameters of the sparse GP."""
        self._current_hyperparameters = hyperparameters
    
    def compute_likelihood_gradient_stable(self, precomputed=False):
        """Compute likelihood gradient in a numerically stable way."""
        # Return a scalar negative log likelihood
        # The gradient is stored in self.likelihood_gradient
        self.likelihood_gradient = torch.tensor([-0.1, 0.05, -0.02])
        return torch.tensor(-2.5)

hyperparameters = torch.tensor([1.2, 0.6, 0.25])
sparse_gp = SparseGP()

ARGS = {
    'hyperparameters': hyperparameters,
    'sparse_gp': sparse_gp,
    'precomputed': False,
}
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 tuple — a tuple output has no array shape of its own; none is 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
dumpcompare Ran
ran, violated the contract we drafted
mir-group/flare/tests/test_gp.py
code served (permissive licence) · get_code("9db2b4284972ba6e")
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 numpy as np

# Test case 1: Two identical dictionaries with nested structures
obj1 = {
    'model_weights': np.array([1.0, 2.0, 3.0]),
    'config': {
        'learning_rate': 0.001,
        'batch_size': 32
    },
    'data': [np.array([1, 2]), np.array([3, 4])],
    'name_field': 'ignore_me'
}

obj2 = {
    'model_weights': np.array([1.0, 2.0, 3.0]),
    'config': {
        'learning_rate': 0.001,
        'batch_size': 32
    },
    'data': [np.array([1, 2]), np.array([3, 4])],
    'name_field': 'different_value'
}

ARGS = {
    'obj1': obj1,
    'obj2': obj2
}
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 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
organize_data Ran
ran on a synthesized input
mir-group/flare_pp/flare_pp/parsers.py
code served (permissive licence) · get_code("249818c3a9a8f3fe")
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 torch
import numpy as np
from copy import deepcopy

torch.manual_seed(0)

# The function organize_data expects a dict with:
# - 'calc_types': a list/array of calculation types (0 for DFT, 1 for GP)
# - Other keys with list values that will be partitioned
# - 'store_hyps', 'store_atoms', 'store_natoms' are skipped during partitioning

all_data = {
    'calc_types': [0, 1, 0, 1],  # Mix of DFT (0) and GP (1) calculations
    'energies': [1.0, 2.0, 3.0, 4.0],
    'forces': [[0.1, 0.2], [0.3, 0.4], [0.5, 0.6], [0.7, 0.8]],
    'stresses': [0.01, 0.02, 0.03, 0.04],
    'store_hyps': [10, 20, 30, 40],  # Should not be partitioned
    'store_atoms': ['a', 'b', 'c', 'd'],  # Should not be partitioned
    'store_natoms': [1, 2, 3, 4],  # Should not be partitioned
}

ARGS = {
    'all_data': all_data,
}
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 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
sort_variances Ran
ran on a synthesized input
mir-group/flare_pp/flare_pp/sparse_gp_calculator.py
code served (permissive licence) · get_code("e7226ad66660e127")
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 (6,) 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

class DescriptorValues:
    def __init__(self, atom_indices, n_types):
        self.atom_indices = atom_indices
        self.n_types = n_types

class StructureDescriptor:
    def __init__(self, noa, descriptor_values):
        self.noa = noa
        self.descriptors = [descriptor_values]

# Create atom_indices: list of lists, where each inner list contains atom indices for that type
# For 3 atom types with atoms distributed among them
atom_indices = [
    [0, 1],      # type 0: atoms at indices 0, 1
    [2],         # type 1: atom at index 2
    [3, 4, 5]    # type 2: atoms at indices 3, 4, 5
]

n_types = 3
noa = 6  # total number of atoms

descriptor_values = DescriptorValues(atom_indices=atom_indices, n_types=n_types)
structure_descriptor = StructureDescriptor(noa=noa, descriptor_values=descriptor_values)

# Create variances array with one variance value per atom
variances = np.array([0.1, 0.2, 0.3, 0.4, 0.5, 0.6])

ARGS = {
    "structure_descriptor": structure_descriptor,
    "variances": variances
}
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 float, shape (6,) — shape (6,) 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
parse_otf Not yet run
unverified (harvested, no recorded run)
mir-group/flare_pp/flare_pp/parsers.py
code served (permissive licence) · get_code("7a939cd586dcd1c3")
How it ran

no invocation recorded.

parse_outfiles Not yet run
unverified (harvested, no recorded run)
mir-group/flare_pp/flare_pp/parsers.py
code served (permissive licence) · get_code("ffff3f5ab4848044")
How it ran

no invocation recorded.

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 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 8 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

Accurate modeling of chemically reactive systems has traditionally relied on either expensive ab initio approaches or flexible bond-order force fields such as ReaxFF that require considerable time, effort, and expertise to parameterize. Here, we introduce FLARE++, a Bayesian active learning method for training reactive many-body force fields on the fly during molecular dynamics (MD) simulations. During the automated training loop, the predictive uncertainties of a sparse Gaussian process (SGP) force field are evaluated at each timestep of an MD simulation to determine whether additional ab initio data are needed. Once trained, the SGP is mapped onto an equivalent and much faster model that is polynomial in the local environment descriptors and whose prediction cost is independent of the training set size. We apply our method to a canonical reactive system in the field of heterogeneous catalysis, hydrogen splitting and recombination on a platinum (111) surface, obtaining a trained model within three days of wall time that is twice as fast as a recent Pt/H ReaxFF force field and considerably more accurate. Our method is fully open source and is expected to reduce the time and effort required to train fast and accurate reactive force fields for complex systems.

For agents

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

get_harvested_code_for_paper("2106.01949")
get_code_for_paper("2106.01949")
have("2106.01949")

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

curl https://syntology.ai/api/ran/2106.01949.json

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

[![Syntology run record](https://syntology.ai/api/ran/2106.01949.svg)](https://syntology.ai/paper/2106.01949)

Connect an agent — have() is free.