SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1703.06103 · 2017

Modeling Relational Data with Graph Convolutional Networks

Max Welling, Rianne Van Den Berg, Thomas Kipf, Michael Schlichtkrull, Ivan Titov, Peter Bloem

arXiv · PDF · Open in the Atlas

Code that ran

18 of 32 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (3 of them constructed an object rather than computing a result; 1 honoured the contract we drafted, 1 violated it and 11 have no contract verdict; 5 failures of our instrument; 14 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/1703.06103.json · badge: /api/ran/1703.06103.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
tkipf/relational-gcn canonical (classifier label)
classifier basis: census
5 ran · 0 no recorded run
susurrant/spatial-interaction-modeling community
Papers with Code archive
5 ran · 2 no recorded run
MichSchli/RelationPrediction community (classifier label)
classifier basis not recorded on the link
3 ran · 1 no recorded run
QustKcz/relational-GCN community
Papers with Code archive
2 ran · 2 no recorded run
parkererickson/crunchBaseGraph community
Papers with Code archive
1 ran · 2 no recorded run
tkipf/gae community
Papers with Code archive
1 ran · 1 no recorded run
copy not recorded community
not in the paper's declared repositories
1 ran · 0 no recorded run
masakicktashiro/rgcn_pytorch_implementation community
Papers with Code archive
0 ran · 1 no recorded run
kracr/document-level-relation-extraction community
Papers with Code archive
0 ran · 1 no recorded run
giuseppefutia/semi community
Papers with Code archive
0 ran · 1 no recorded run
INK-USC/RE-Net community
Papers with Code archive
0 ran · 1 no recorded run
berlincho/RGCN-pytorch community
Papers with Code archive
0 ran · 1 no recorded run
thiviyanT/torch-rgcn community
Papers with Code archive
0 ran · 1 no recorded run
FunctionStatusWhere it lives
Layer Ran
ran on a synthesized input · constructed an object; no result was computed
tkipf/gae/gae/layers.py
code served (permissive licence) · get_code("c101076233bfcdee")
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 an instance of its own class, Layer: the run constructed the object and did not call it, so no result was computed. 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 tensorflow as tf

_LAYER_UIDS = {}

def get_layer_uid(layer_name=''):
    """Helper function, assigns unique layer IDs
    """
    if layer_name not in _LAYER_UIDS:
        _LAYER_UIDS[layer_name] = 1
        return 1
    else:
        _LAYER_UIDS[layer_name] += 1
        return _LAYER_UIDS[layer_name]

class Layer(object):
    """Base layer class. Defines basic API for all layer objects.

    # Properties
        name: String, defines the variable scope of the layer.

    # Methods
        _call(inputs): Defines computation graph of layer
            (i.e. takes input, returns output)
        __call__(inputs): Wrapper for _call()
    """

    def __init__(self, **kwargs):
        allowed_kwargs = {'name', 'logging'}
        for kwarg in kwargs.keys():
            assert kwarg in allowed_kwargs, 'Invalid keyword argument: ' + kwarg
        name = kwargs.get('name')
        if not name:
            layer = self.__class__.__name__.lower()
            name = layer + '_' + str(get_layer_uid(layer))
        self.name = name
        self.vars = {}
        logging = kwargs.get('logging', False)
        self.logging = logging
        self.issparse = False

    def _call(self, inputs):
        return inputs

    def __call__(self, inputs):
        with tf.name_scope(self.name):
            outputs = self._call(inputs)
            return outputs


ARGS = {
    'name': 'test_layer',
    'logging': True
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Layer — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a Layer 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
Model Ran
ran on a synthesized input · constructed an object; no result was computed
susurrant/spatial-interaction-modeling/SI-GCN/code/encoders/message_gcns/gcn_basis.py
pointer only (licence: NONE) · get_code("2c3abf305bb94378")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, susurrant/spatial-interaction-modeling when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

Model Ran
ran on a synthesized input · constructed an object; no result was computed
QustKcz/relational-GCN/code/encoders/message_gcns/gcn_basis.py
code served (permissive licence) · get_code("e38cd476b8281a12")
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 an instance of its own class, Model: the run constructed the object and did not call it, so no result was computed. 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 tensorflow as tf
import numpy as np

# Create a minimal next_component that implements required methods
class MinimalComponent:
    def needs_graph(self):
        return False
    
    def get_weights(self):
        return []
    
    def get_train_input_variables(self):
        return [tf.placeholder(tf.float32, shape=[None, 6])]
    
    def get_test_input_variables(self):
        return [tf.placeholder(tf.float32, shape=[None, 6])]
    
    def get_loss(self, mode='train'):
        return tf.constant(0.0)
    
    def get_regularization(self, base=0):
        return base
    
    def get_all_subject_codes(self, mode='train'):
        return tf.constant([[1.0, 2.0], [3.0, 4.0]])
    
    def get_all_object_codes(self, mode='train'):
        return tf.constant([[5.0, 6.0], [7.0, 8.0]])
    
    def get_all_codes(self, mode='train'):
        return tf.constant([[1.0, 2.0], [3.0, 4.0]])
    
    def predict(self):
        return tf.constant([0.5, 0.6])
    
    def predict_all_subject_scores(self):
        return tf.constant([[0.1, 0.2], [0.3, 0.4]])
    
    def predict_all_object_scores(self):
        return tf.constant([[0.5, 0.6], [0.7, 0.8]])
    
    def get_graph(self):
        return tf.constant([1.0, 2.0])
    
    def get_additional_ops(self):
        return []
    
    def initialize_train(self):
        pass
    
    def set_variable(self, name, value):
        pass

# Create settings dictionary
settings = {
    'EntityCount': '10',
    'RelationCount': '5',
    'EdgeCount': '20'
}

# Create next_component instance
next_component = MinimalComponent()

ARGS = {
    'next_component': next_component,
    'settings': settings
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Model — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a Model 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
csc_zero_cols Ran
ran on a synthesized input
tkipf/relational-gcn/rgcn/utils.py
code served (permissive licence) · get_code("e1a1d86eb911298c")
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 csc_matrix. 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 scipy import sparse

# Create a small sparse CSC matrix
# Using a 5x4 matrix with some non-zero elements
data = np.array([1.0, 2.0, 3.0, 4.0, 5.0], dtype=np.float64)
indices = np.array([0, 2, 1, 3, 4], dtype=np.int32)
indptr = np.array([0, 2, 3, 4, 5], dtype=np.int32)
csc_matrix = sparse.csc_matrix((data, indices, indptr), shape=(5, 4))

# Columns to zero out
cols_to_zero = np.array([1, 3])

ARGS = {
    "csc": csc_matrix,
    "cols_to_zero": cols_to_zero
}
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 csc_matrix — a csc_matrix output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a csc_matrix 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
csr_zero_rows Ran
ran on a synthesized input
tkipf/relational-gcn/rgcn/utils.py
code served (permissive licence) · get_code("d62c184c10723fda")
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 csr_matrix. 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 scipy import sparse

# Create a small sparse CSR matrix
# Let's create a 5x4 matrix with some non-zero elements
data = np.array([1, 2, 3, 4, 5, 6], dtype=np.float64)
indices = np.array([0, 2, 1, 3, 0, 2], dtype=np.int32)
indptr = np.array([0, 2, 3, 4, 6, 6], dtype=np.int32)
csr = sparse.csr_matrix((data, indices, indptr), shape=(5, 4))

# Specify which rows to zero out
rows_to_zero = np.array([1, 3])

ARGS = {
    "csr": csr,
    "rows_to_zero": rows_to_zero,
}
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 csr_matrix — a csr_matrix output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a csr_matrix 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
dot_or_lookup Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
susurrant/spatial-interaction-modeling/SI-GCN/code/encoders/message_gcns/gcn_basis.py
pointer only (licence: NONE) · get_code("d8d50ae992bdb3c9")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, susurrant/spatial-interaction-modeling when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

gcn_reduce Ran
ran; our contract draft was wrong, not the code
parkererickson/crunchBaseGraph/py_scripts/gcn.py
pointer only (licence: NONE) · get_code("4e2d70eb57ed4788")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, parkererickson/crunchBaseGraph when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_layer_uid Ran
ran, honoured the contract we drafted
this paper's copy was not recorded; identical code first harvested from sigvae/SIGraphVAE
pointer only; the stored copy get_code serves, from sigvae/SIGraphVAE, is pointer only (licence: NONE) · get_code("addac9c28c065096")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

glorot_variance Ran
ran, violated the contract we drafted
susurrant/spatial-interaction-modeling/SI-GCN/code/encoders/message_gcns/gcn_basis.py
pointer only (licence: NONE) · get_code("a9a4285637586425")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, susurrant/spatial-interaction-modeling when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

load_sparse_csr Ran
ran on a synthesized input
tkipf/relational-gcn/rgcn/data_utils.py
code served (permissive licence) · get_code("02de651d9efcfd65")
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 csr_matrix. 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
import scipy.sparse as sp
import tempfile
import os

# Create a temporary directory and file for the sparse matrix
temp_dir = tempfile.mkdtemp()
temp_file = os.path.join(temp_dir, 'sparse_matrix.npz')

# Create a small sparse CSR matrix
data = np.array([1.0, 2.0, 3.0, 4.0], dtype=np.float32)
indices = np.array([0, 2, 1, 3], dtype=np.int32)
indptr = np.array([0, 2, 3, 4], dtype=np.int32)
shape = (3, 4)

# Save the sparse matrix in the format expected by load_sparse_csr
np.savez(temp_file, data=data, indices=indices, indptr=indptr, shape=shape)

ARGS = {
    'filename': temp_file
}
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 csr_matrix — a csr_matrix output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a csr_matrix 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
make_tf_bias Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
susurrant/spatial-interaction-modeling/SI-GCN/code/encoders/message_gcns/gcn_basis.py
pointer only (licence: NONE) · get_code("2ff4c1185681f679")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, susurrant/spatial-interaction-modeling when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

make_tf_variable Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
susurrant/spatial-interaction-modeling/SI-GCN/code/encoders/message_gcns/gcn_basis.py
pointer only (licence: NONE) · get_code("ce564e1ae279d09c")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, susurrant/spatial-interaction-modeling when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

make_tf_variable Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
QustKcz/relational-GCN/code/encoders/message_gcns/gcn_basis.py
code served (permissive licence) · get_code("eee67eafc4b13ee4")
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 ResourceVariable, ndim 2, dtype family float, shape (2, 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 tensorflow as tf
import numpy as np

np.random.seed(0)

ARGS = {
    'mean': 0.0,
    'variance': 1.0,
    'shape': (2, 4),
    'init': 'normal'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ResourceVariable, ndim 2, dtype family float, shape (2, 4) — shape (2, 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
parse Ran
ran on a synthesized input
tkipf/relational-gcn/rgcn/data_utils.py
code served (permissive licence) · get_code("241fcba67ca4a269")
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 torch

def parse(symbol):
    if symbol.startswith('<'):
        return symbol[1:-1]
    return symbol

ARGS = {
    'symbol': '<test>'
}
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 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
read Ran
ran on a synthesized input
MichSchli/RelationPrediction/code/common/settings_reader.py
code served (permissive licence) · get_code("aecde0e9a66b3279")
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 Settings. 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 tempfile
import os

# Create a temporary settings file
temp_dir = tempfile.mkdtemp()
settings_file = os.path.join(temp_dir, "test_settings.txt")

# Write a sample settings file with various content
settings_content = """key1=value1
key2=value2
[section1]
	nested_key1=nested_value1
	nested_key2=nested_value2
[section2]
	another_key=another_value
"""

with open(settings_file, 'w') as f:
    f.write(settings_content)

ARGS = {
    "filename": settings_file
}
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 Settings — a Settings output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Settings 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
read_triplet_file Ran
ran on a synthesized input
MichSchli/RelationPrediction/code/common/io.py
code served (permissive licence) · get_code("d945735b7bb56f17")
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 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 tempfile
import os

# Create a temporary file with tab-separated triplet data
temp_file = tempfile.NamedTemporaryFile(mode='w', delete=False, suffix='.txt')
temp_file.write("entity1\trelation1\tentity2\n")
temp_file.write("entity3\trelation2\tentity4\n")
temp_file.write("entity5\trelation3\tentity6\n")
temp_file.close()

filename = temp_file.name

ARGS = {
    'filename': filename
}
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 2, 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
read_triplets_as_list Ran
ran on a synthesized input
MichSchli/RelationPrediction/code/common/io.py
code served (permissive licence) · get_code("536bd3fa07b27fed")
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 tempfile
import os

# Create temporary files for the function to read
temp_dir = tempfile.mkdtemp()

# Create entity dictionary file (id \t entity_name format)
entity_dict_file = os.path.join(temp_dir, "entities.txt")
with open(entity_dict_file, 'w') as f:
    f.write("0\tAlice\n")
    f.write("1\tBob\n")
    f.write("2\tCharlie\n")

# Create relation dictionary file (id \t relation_name format)
relation_dict_file = os.path.join(temp_dir, "relations.txt")
with open(relation_dict_file, 'w') as f:
    f.write("0\tknows\n")
    f.write("1\tlikes\n")

# Create triplets file (entity_name \t relation_name \t entity_name format)
triplets_file = os.path.join(temp_dir, "triplets.txt")
with open(triplets_file, 'w') as f:
    f.write("Alice\tknows\tBob\n")
    f.write("Bob\tlikes\tCharlie\n")

ARGS = {
    "filename": triplets_file,
    "entity_dict": entity_dict_file,
    "relation_dict": relation_dict_file,
}
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 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
sp_vec_from_idx_list Ran
ran on a synthesized input
tkipf/relational-gcn/rgcn/utils.py
code served (permissive licence) · get_code("24a38c97cf6b6580")
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 csr_matrix. 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 scipy.sparse as sp
import numpy as np

# Set up arguments for sp_vec_from_idx_list
idx_list = [0, 2, 4, 6]  # list of indices
dim = 10  # dimensionality of the sparse vector

ARGS = {
    "idx_list": idx_list,
    "dim": dim
}
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 csr_matrix — a csr_matrix output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a csr_matrix 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
BasisGcn Not yet run
unverified (harvested, no recorded run)
susurrant/spatial-interaction-modeling/SI-GCN/code/encoders/message_gcns/gcn_basis.py
pointer only (licence: NONE) · get_code("7e784123e5ac077d")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, susurrant/spatial-interaction-modeling when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

BasisGcn Not yet run
unverified (harvested, no recorded run)
QustKcz/relational-GCN/code/encoders/message_gcns/gcn_basis.py
code served (permissive licence) · get_code("6fc2ea51851a4f5d")
How it ran

no invocation recorded.

GCNLayer Not yet run
unverified (harvested, no recorded run)
parkererickson/crunchBaseGraph/py_scripts/gcn.py
pointer only (licence: NONE) · get_code("2bfb21c319767958")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, parkererickson/crunchBaseGraph when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

GraphConvolution Not yet run
unverified (harvested, no recorded run)
tkipf/gae/gae/layers.py
code served (permissive licence) · get_code("c97818a941d8a0b9")
How it ran

no invocation recorded.

GraphConvolution Not yet run
unverified (harvested, no recorded run)
masakicktashiro/rgcn_pytorch_implementation/layers.py
pointer only (licence: NONE) · get_code("8d26f1fa141a632b")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, masakicktashiro/rgcn_pytorch_implementation when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

MessageGcn Not yet run
unverified (harvested, no recorded run)
susurrant/spatial-interaction-modeling/SI-GCN/code/encoders/message_gcns/gcn_basis.py
pointer only (licence: NONE) · get_code("c0687ad3ab3730c9")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, susurrant/spatial-interaction-modeling when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

MessageGcn Not yet run
unverified (harvested, no recorded run)
QustKcz/relational-GCN/code/encoders/message_gcns/gcn_basis.py
code served (permissive licence) · get_code("1560b6e47bad3798")
How it ran

no invocation recorded.

RGCNConv Not yet run
unverified (harvested, no recorded run)
kracr/document-level-relation-extraction/DocRE-CLiP/code/link prediction/models.py
code served (permissive licence) · get_code("b858c1faaadce1a7")
How it ran

no invocation recorded.

RGCNLayer Not yet run
unverified (harvested, no recorded run)
giuseppefutia/semi/src/link_prediction/layers.py
pointer only (licence: GPL-3.0) · get_code("aff7c6441824509d")
How it ran

The body of this sample is withheld: GPL-3.0, as GitHub detects it on this paper's repository, giuseppefutia/semi, is not in the permissive set Syntology serves inline. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

RGCNLayer Not yet run
unverified (harvested, no recorded run)
INK-USC/RE-Net/RGCN.py
pointer only (licence: NONE) · get_code("e1fa7a1467322bab")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, INK-USC/RE-Net when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

RelationalGraphConvModel Not yet run
unverified (harvested, no recorded run)
berlincho/RGCN-pytorch/models.py
pointer only (licence: NONE) · get_code("396309e66058552e")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, berlincho/RGCN-pytorch when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

RelationalGraphConvolutionNC Not yet run
unverified (harvested, no recorded run)
thiviyanT/torch-rgcn/torch_rgcn/layers.py
code served (permissive licence) · get_code("c746dc3cd7142fb7")
How it ran

no invocation recorded.

gcn_message Not yet run
unverified (harvested, no recorded run)
parkererickson/crunchBaseGraph/py_scripts/gcn.py
pointer only (licence: NONE) · get_code("a8e4cdc01a9f23a6")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, parkererickson/crunchBaseGraph when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

read_dictionary Not yet run
unverified (harvested, no recorded run)
MichSchli/RelationPrediction/code/common/io.py
code served (permissive licence) · get_code("00587b6b3a684969")
How it ran

no invocation recorded.

Also present in other papers' code

Of 32 samples lifted from this paper's linked repositories, 32 carry a function identity and 0 do not. Of the 32 distinct functions with an identity, 4 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 1 the same month, 4 later) and 28 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.

FunctionAlso present inEarliest-datedOthers named
get_layer_uid
brick of this paper's assembly: helper
33 other papers
4 dated earlier, 1 the same month, 28 later · 5 only as an assembly brick
1603.08861 (2016-03)
Revisiting Semi-Supervised Learning with Graph Embeddings
1606.09375 (2016-06)
Convolutional Neural Networks on Graphs with Fast Localized Spectral Filtering
1609.02907 (2016-09)
Semi-Supervised Classification with Graph Convolutional Networks
and 30 more not named here
RGCNConv
brick of this paper's assembly: entry
1 other paper
1 later
2005.00791 (2020-05)
KinGDOM: Knowledge-Guided DOMain Adaptation for Sentiment Analysis
gcn_message
brick of this paper's assembly: helper
1 other paper
1 later
2306.12327 (2023-06)
Learning the galaxy-environment connection with graph neural networks
gcn_reduce
brick of this paper's assembly: helper
1 other paper
1 later
2306.12327 (2023-06)
Learning the galaxy-environment connection with graph neural networks

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

Knowledge graphs enable a wide variety of applications, including question answering and information retrieval. Despite the great effort invested in their creation and maintenance, even the largest (e.g., Yago, DBPedia or Wikidata) remain incomplete. We introduce Relational Graph Convolutional Networks (R-GCNs) and apply them to two standard knowledge base completion tasks: Link prediction (recovery of missing facts, i.e. subject-predicate-object triples) and entity classification (recovery of missing entity attributes). R-GCNs are related to a recent class of neural networks operating on graphs, and are developed specifically to deal with the highly multi-relational data characteristic of realistic knowledge bases. We demonstrate the effectiveness of R-GCNs as a stand-alone model for entity classification. We further show that factorization models for link prediction such as DistMult can be significantly improved by enriching them with an encoder model to accumulate evidence over multiple inference steps in the relational graph, demonstrating a large improvement of 29.8% on FB15k-237 over a decoder-only baseline. * Equal contribution.

For agents

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

get_harvested_code_for_paper("1703.06103")
get_code_for_paper("1703.06103")
have("1703.06103")

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

curl https://syntology.ai/api/ran/1703.06103.json

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

[![Syntology run record](https://syntology.ai/api/ran/1703.06103.svg)](https://syntology.ai/paper/1703.06103)

Connect an agent — have() is free.