SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2212.04450 · 2022

GAUCHE: A Library for Gaussian Processes in Chemistry

arXiv · PDF · Open in the Atlas

Code that ran

6 of 10 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 6 have no contract verdict; 0 failures of our instrument; 4 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/2212.04450.json · badge: /api/ran/2212.04450.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
leojklarner/gauche canonical (classifier label)
classifier basis not recorded on the link
6 ran · 4 no recorded run
FunctionStatusWhere it lives
bag_of_characters Ran
ran on a synthesized input
leojklarner/gauche/gauche/representations/strings.py
code served (permissive licence) · get_code("56b8ff9a52d03c6e")
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 int, shape (4, 32) 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
from typing import List, Optional

# Construct valid arguments for bag_of_characters function
strings = [
    "CCO",
    "CC(C)O",
    "c1ccccc1",
    "CC(=O)O"
]

max_ngram = 3

selfies = False

ARGS = {
    "strings": strings,
    "max_ngram": max_ngram,
    "selfies": selfies
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, selfies, sklearn — 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 int, shape (4, 32) — shape (4, 32) 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
batch_dice_sim Ran
ran on a synthesized input
leojklarner/gauche/gauche/kernels/fingerprint_kernels/dice_kernel.py
code served (permissive licence) · get_code("70cae3ac2ee3b98c")
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 3, dtype family float, shape (2, 3, 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 torch

torch.manual_seed(0)

# Create small batched tensors for x1 and x2
# x1: [b x n x d] where b=2 (batch), n=3 (sequence length), d=4 (feature dim)
# x2: [b x m x d] where b=2 (batch), m=5 (different sequence length), d=4 (feature dim)
x1 = torch.randn(2, 3, 4)
x2 = torch.randn(2, 5, 4)

# eps for numerical stability
eps = 1e-6

ARGS = {
    "x1": x1,
    "x2": x2,
    "eps": eps,
}
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 Tensor, ndim 3, dtype family float, shape (2, 3, 5) — shape (2, 3, 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
load_class Ran
ran on a synthesized input
leojklarner/gauche/gauche/gp.py
code served (permissive licence) · get_code("620718beeccc9a43")
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 type. 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 importlib
import torch
import torch.nn as nn

torch.manual_seed(0)

# The function load_class dynamically imports a module and retrieves a class from it.
# It uses:
# - importlib.import_module(module_name) to load the module
# - getattr(module, class_name) to get the class from the module
# 
# Valid arguments would be:
# - module_name: a string naming an existing module (e.g., 'torch.nn')
# - class_name: a string naming a class that exists in that module (e.g., 'Linear')

ARGS = {
    "module_name": "torch.nn",
    "class_name": "Linear"
}
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 type — a type output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a type 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
one_hot Ran
ran on a synthesized input
leojklarner/gauche/gauche/representations/fingerprints.py
code served (permissive licence) · get_code("27570ebcbb961e1c")
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 (3, 7) 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 pandas as pd

# Create a small pandas DataFrame with various reaction parameters
# The function will apply one-hot encoding to categorical columns
df = pd.DataFrame({
    'reactant': ['A', 'B', 'A'],
    'reagent': ['NaOH', 'HCl', 'NaOH'],
    'temperature': [25, 50, 25],
    'solvent': ['water', 'DMSO', 'water']
})

ARGS = {
    'df': df
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, pandas — 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 (3, 7) — shape (3, 7) 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
rxnfp Ran
ran on a synthesized input
leojklarner/gauche/gauche/representations/fingerprints.py
code served (permissive licence) · get_code("34585a1677e2c03a")
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, 256) 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
from typing import List

# Create a minimal mock of the RXNBERTFingerprintGenerator and related components
class MockTokenizer:
    def __call__(self, text, **kwargs):
        # Mock tokenizer that returns a dict with input_ids
        return {"input_ids": [[1, 2, 3, 4]]}

class MockRXNModel:
    def __call__(self, *args, **kwargs):
        # Mock model that returns embeddings
        return {"last_hidden_state": np.random.randn(1, 256).astype(np.float32)}

class MockRXNBERTFingerprintGenerator:
    def __init__(self, model, tokenizer):
        self.model = model
        self.tokenizer = tokenizer
    
    def convert(self, smiles):
        # Mock convert method that returns a 256-dim fingerprint
        return np.random.randn(256).astype(np.float64)

# Mock the rxnfp module functions
def mock_get_default_model_and_tokenizer():
    return MockRXNModel(), MockTokenizer()

# Patch the imports in the function's namespace
import sys
from types import ModuleType

rxnfp_module = ModuleType("rxnfp")
transformer_fingerprints_module = ModuleType("transformer_fingerprints")

transformer_fingerprints_module.RXNBERTFingerprintGenerator = MockRXNBERTFingerprintGenerator
transformer_fingerprints_module.get_default_model_and_tokenizer = mock_get_default_model_and_tokenizer

rxnfp_module.transformer_fingerprints = transformer_fingerprints_module

sys.modules["rxnfp"] = rxnfp_module
sys.modules["rxnfp.transformer_fingerprints"] = transformer_fingerprints_module

# Prepare valid arguments for the rxnfp function
ARGS = {
    "reaction_smiles": [
        "CC(C)Cc1ccc(cc1)[C@@H](C)C(=O)O.CCO>>CC(C)Cc1ccc(cc1)[C@@H](C)C(=O)OCC",
        "c1ccccc1Br.c1ccccc1B(O)O>>c1ccccc1c2ccccc2"
    ]
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, rxnfp — 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, 256) — shape (2, 256) 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
transform_data Ran
ran on a synthesized input
leojklarner/gauche/gauche/dataloader/data_utils.py
code served (permissive licence) · get_code("7e96d38a7b1e0e8e")
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 numpy as np
from sklearn.decomposition import PCA
from sklearn.preprocessing import StandardScaler

# Create small sample data with distinct dimensions
np.random.seed(42)

# Training set: 2 samples, 4 features
X_train = np.random.randn(2, 4).astype(np.float32)
# Training targets: 2 samples, 1 target dimension
y_train = np.random.randn(2, 1).astype(np.float32)

# Test set: 2 samples, 4 features
X_test = np.random.randn(2, 4).astype(np.float32)
# Test targets: 2 samples, 1 target dimension
y_test = np.random.randn(2, 1).astype(np.float32)

ARGS = {
    "X_train": X_train,
    "y_train": y_train,
    "X_test": X_test,
    "y_test": y_test,
    "use_pca": False,
    "n_components": 2
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, sklearn — 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
batch_braun_blanquet_sim Not yet run
unverified (harvested, no recorded run)
leojklarner/gauche/gauche/kernels/fingerprint_kernels/braun_blanquet_kernel.py
code served (permissive licence) · get_code("383e81358c3bad15")
How it ran

no invocation recorded.

batch_faith_sim Not yet run
unverified (harvested, no recorded run)
leojklarner/gauche/gauche/kernels/fingerprint_kernels/faith_kernel.py
code served (permissive licence) · get_code("42e1e3d4f7b2bb6c")
How it ran

no invocation recorded.

drfp Not yet run
unverified (harvested, no recorded run)
leojklarner/gauche/gauche/representations/fingerprints.py
code served (permissive licence) · get_code("0e4f0b8f22be50a0")
How it ran

no invocation recorded.

molecular_graphs Not yet run
unverified (harvested, no recorded run)
leojklarner/gauche/gauche/representations/graphs.py
code served (permissive licence) · get_code("455665881c64b556")
How it ran

no invocation recorded.

Also present in other papers' code

Of 10 samples lifted from this paper's linked repositories, 10 carry a function identity and 0 do not. Of the 10 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 10 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

We introduce GAUCHE, a library for GAUssian processes in CHEmistry. Gaussian processes have long been a cornerstone of probabilistic machine learning, affording particular advantages for uncertainty quantification and Bayesian optimisation. Extending Gaussian processes to chemical representations, however, is nontrivial, necessitating kernels defined over structured inputs such as graphs, strings and bit vectors. By defining such kernels in GAUCHE, we seek to open the door to powerful tools for uncertainty quantification and Bayesian optimisation in chemistry. Motivated by scenarios frequently encountered in experimental chemistry, we showcase applications for GAUCHE in molecular discovery and chemical reaction optimisation. The codebase is made available at https://github.com/leojklarner/gauche

For agents

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

get_harvested_code_for_paper("2212.04450")
get_code_for_paper("2212.04450")
have("2212.04450")

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

curl https://syntology.ai/api/ran/2212.04450.json

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

[![Syntology run record](https://syntology.ai/api/ran/2212.04450.svg)](https://syntology.ai/paper/2212.04450)

Connect an agent — have() is free.