SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2201.07788 · CVPR · 2022

ConDor: Self-Supervised Canonicalization of 3D Pose for Partial Shapes

Leonidas Guibas, Srinath Sridhar, Adrien Poulenard, Jivitesh Jain, Rahul Sajnani, Radhika Dua, Figure Condor

arXiv · PDF · Open in the Atlas

Code that ran

18 of 19 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 18 have no contract verdict; 0 failures of our instrument; 1 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2201.07788.json · badge: /api/ran/2201.07788.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
brown-ivl/ConDor canonical (classifier label)
classifier basis not recorded on the link
18 ran · 1 no recorded run
FunctionStatusWhere it lives
apply_mlp Ran
ran on a synthesized input
brown-ivl/ConDor/ConDor/auto_encoder/tfn_atlasnet.py
code served (permissive licence) · get_code("a43929f6265ac2c2")
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 EagerTensor, ndim 2, dtype family float, shape (2, 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 tensorflow as tf
import numpy as np
from tensorflow.keras.layers import Dense, BatchNormalization, Activation, Dropout

# Set seed for reproducibility
tf.random.set_seed(0)
np.random.seed(0)

# Create a small input tensor (batch_size=2, features=4)
x = tf.constant(np.random.randn(2, 4).astype(np.float32))

# Create MLP structure: list of dicts with 'layer' and 'bn_layer' keys
mlp = [
    {
        'layer': Dense(3, activation=None),  # Output dim=3
        'bn_layer': BatchNormalization()
    },
    {
        'layer': Dense(5, activation=None),  # Output dim=5
        'bn_layer': BatchNormalization()
    }
]

# Build the layers by calling them once to initialize weights
# This is necessary for Keras layers to be properly initialized
temp = x
for block in mlp:
    temp = block['layer'](temp)
    temp = block['bn_layer'](temp)

ARGS = {
    'x': x,
    'mlp': mlp
}
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 EagerTensor, ndim 2, dtype family float, shape (2, 5) — shape (2, 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
eqvar_heads_fuzion Ran
ran on a synthesized input
brown-ivl/ConDor/ConDor/network_utils/convolution.py
code served (permissive licence) · get_code("95ec15b1571be826")
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 dict. 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

# Analyze the function:
# - Takes X as input
# - Iterates over X (so X is iterable)
# - For each x in X, iterates over x (so x is iterable)
# - For each l in x, accesses x[l] (so x is dict-like)
# - Concatenates lists of tensors using Concatenate() layer
# - Returns a dict of concatenated tensors

# Build sample data: X should be a list of dicts
# Each dict maps string keys to tensors
# The function will group tensors by key and concatenate them

# Create sample tensors (small, CPU only)
tensor1_key_a = tf.constant([[1.0, 2.0, 3.0, 4.0]], dtype=tf.float32)  # shape (1, 4)
tensor1_key_b = tf.constant([[5.0, 6.0, 7.0, 8.0]], dtype=tf.float32)  # shape (1, 4)

tensor2_key_a = tf.constant([[9.0, 10.0, 11.0, 12.0]], dtype=tf.float32)  # shape (1, 4)
tensor2_key_b = tf.constant([[13.0, 14.0, 15.0, 16.0]], dtype=tf.float32)  # shape (1, 4)

# Create list of dicts
dict1 = {
    'a': tensor1_key_a,
    'b': tensor1_key_b
}

dict2 = {
    'a': tensor2_key_a,
    'b': tensor2_key_b
}

X = [dict1, dict2]

ARGS = {
    'X': X
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
grid_sampler Ran
ran on a synthesized input
brown-ivl/ConDor/ConDor/network_utils/pooling.py
code served (permissive licence) · get_code("183d074ff7b98277")
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 tensorflow as tf
import numpy as np

# Set seed for reproducibility
tf.random.set_seed(0)

# Create small test tensors
batch_size = 2
num_points = 4
num_points_target = 2

# Create x: shape (batch_size, num_points, 3) - 3D coordinates
# Using small values so they round to distinct grid cells
x = tf.constant([
    [[1.1, 2.2, 3.3], [1.5, 2.5, 3.5], [5.1, 5.2, 5.3], [5.5, 5.6, 5.7]],
    [[10.1, 10.2, 10.3], [10.5, 10.6, 10.7], [15.1, 15.2, 15.3], [15.5, 15.6, 15.7]]
], dtype=tf.float32)

cell_size = 1.0

ARGS = {
    "x": x,
    "cell_size": cell_size,
    "num_points_target": num_points_target
}
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 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
grid_sampler_ Ran
ran on a synthesized input
brown-ivl/ConDor/ConDor/network_utils/pooling.py
code served (permissive licence) · get_code("87f376a5a795f16c")
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 tensorflow as tf
import numpy as np

# Set seed for reproducibility
tf.random.set_seed(0)
np.random.seed(0)

# Create small test tensors
batch_size = 2
num_points = 4
spatial_dims = 3

# x: (batch_size, num_points, spatial_dims) - 3D point coordinates
x = tf.constant(np.random.uniform(0, 10, (batch_size, num_points, spatial_dims)), dtype=tf.float32)

# cell_size: scalar controlling grid resolution
cell_size = 2.0

# num_points_target: target number of points to sample
num_points_target = 2

ARGS = {
    "x": x,
    "cell_size": cell_size,
    "num_points_target": num_points_target
}
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 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
mask Ran
ran on a synthesized input
brown-ivl/ConDor/ConDor/network_utils/convolution.py
code served (permissive licence) · get_code("8b561ee6fdb8f3fd")
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 dict. 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 small mask tensor (batch=2, spatial dims)
mask_value = tf.constant([[1.0, 0.0, 1.0],
                          [0.0, 1.0, 1.0]], dtype=tf.float32)

# Create a dict with integer keys mapping to tensors
# The function multiplies mask_ (expanded to [..., 1, 1]) with x[l]
# So x[l] should be a tensor with at least 2 extra dimensions for the expand_dims to work
x_dict = {
    0: tf.constant(np.random.randn(2, 3, 4, 5), dtype=tf.float32),
    1: tf.constant(np.random.randn(2, 3, 6, 7), dtype=tf.float32),
}

ARGS = {
    'x': x_dict,
    'mask': mask_value
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
norml2 Ran
ran on a synthesized input
brown-ivl/ConDor/ConDor/network_utils/activations.py
code served (permissive licence) · get_code("83c97d74407b4755")
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 EagerTensor, ndim 1, dtype family float, shape (2,) 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

# Create a small test tensor
x = tf.constant([[1.0, 2.0, 3.0, 4.0],
                 [5.0, 6.0, 7.0, 8.0]], dtype=tf.float32)

ARGS = {
    "x": x,
    "axis": -1,
    "keepdims": False,
    "eps": 1e-07
}
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 EagerTensor, ndim 1, dtype family float, shape (2,) — shape (2,) 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
patches_radius Ran
ran on a synthesized input
brown-ivl/ConDor/ConDor/network_utils/group_points.py
code served (permissive licence) · get_code("c01f0222f247ed28")
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 EagerTensor, ndim 3, dtype family float, shape (2, 1, 1) 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

# Create a small sq_norm tensor with shape (batch_size, num_points, 1)
# batch_size = 2, num_points = 4
batch_size = 2
num_points = 4
sq_norm = tf.constant(np.random.rand(batch_size, num_points, 1), dtype=tf.float32)

# Test with different radius types
radius_float = 0.5

ARGS = {
    "radius": radius_float,
    "sq_norm": sq_norm,
}
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 EagerTensor, ndim 3, dtype family float, shape (2, 1, 1) — shape (2, 1, 1) 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
set_mlp Ran
ran on a synthesized input
brown-ivl/ConDor/ConDor/auto_encoder/tfn_atlasnet.py
code served (permissive licence) · get_code("f1475b36486b7bbe")
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 tensorflow as tf
from tensorflow.keras.layers import Dense, BatchNormalization, Activation, Dropout

# Define arguments for set_mlp function
units = [64, 32, 16]  # List of unit sizes for each layer
momentum = 0.99       # Momentum value for BatchNormalization

ARGS = {
    'units': units,
    'momentum': momentum
}
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 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
set_sphere_weights Ran
ran on a synthesized input
brown-ivl/ConDor/ConDor/auto_encoder/tfn_atlasnet.py
code served (permissive licence) · get_code("34d3cd3a8304196e")
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 dict. 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
from tensorflow.keras.layers import Dense, BatchNormalization, Activation, Dropout

def set_sphere_weights(units, types):
    weights = dict()
    for l in types:
        if int(l) == 0:
            weights[l] = Dense(units=units)
        else:
            weights[l] = Dense(units=units, use_bias=False)
    return weights

ARGS = {
    "units": 16,
    "types": [0, 1, 2]
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
sq_dist_mat Ran
ran on a synthesized input
brown-ivl/ConDor/ConDor/evaluators/train_pointnet_cls2.py
code served (permissive licence) · get_code("bc0a1952c0374318")
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 EagerTensor, 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 tensorflow as tf
import numpy as np

# Set seed for reproducibility
tf.random.set_seed(0)
np.random.seed(0)

# Create small tensors with distinct dimensions
# x: shape (batch=2, num_points=3, dim=4)
# y: shape (batch=2, num_points=5, dim=4)
x = tf.constant(np.random.randn(2, 3, 4).astype(np.float32))
y = tf.constant(np.random.randn(2, 5, 4).astype(np.float32))

ARGS = {
    "x": x,
    "y": y,
}
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 EagerTensor, 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
squash Ran
ran on a synthesized input
brown-ivl/ConDor/ConDor/network_utils/activations.py
code served (permissive licence) · get_code("6ee731af0229a9ba")
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 EagerTensor, ndim 3, dtype family float, shape (2, 3, 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

# Create a small tensor for testing
# Shape: (batch_size=2, feature_dim=4, reduce_dim=3)
# The function reduces over axis=-2, so we need at least 2 dimensions
# Making it (2, 3, 4) so axis=-2 reduces over dimension of size 3
x = tf.constant(np.random.randn(2, 3, 4).astype(np.float32))

b = 1.0

eps = 1e-07

bias = tf.constant(np.array([0.1, 0.2, 0.3]).astype(np.float32), shape=(1, 3, 1))

ARGS = {
    "x": x,
    "b": b,
    "eps": eps,
    "bias": bias,
}
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 EagerTensor, ndim 3, dtype family float, shape (2, 3, 4) — shape (2, 3, 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
std Ran
ran on a synthesized input
brown-ivl/ConDor/ConDor/evaluators/train_pointnet_cls2.py
code served (permissive licence) · get_code("95a1e02c41de556a")
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 tensorflow as tf
import numpy as np

# Create a small test tensor with distinct dimensions
# Using shape (2, 4) - batch size 2, feature dimension 4
x = tf.constant([[1.0, 2.0, 3.0, 4.0],
                 [5.0, 6.0, 7.0, 8.0]], dtype=tf.float32)

ARGS = {
    'x': x,
    'axis_mean': 0,
    'axis_norm': 1
}
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 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
tf_heaviside Ran
ran on a synthesized input
brown-ivl/ConDor/ConDor/network_utils/sh_conv.py
code served (permissive licence) · get_code("492443ff0f1a5a31")
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 EagerTensor, 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 tensorflow as tf
import numpy as np

def tf_heaviside(X):
    return tf.maximum(0.0, tf.sign(X))

# Create a small TensorFlow tensor with both positive and negative values
# to test the heaviside function behavior
X = tf.constant([[-2.0, -1.0, 0.5, 3.0],
                  [-0.5, 1.5, -3.0, 2.0]], dtype=tf.float32)

ARGS = {
    "X": X
}
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 EagerTensor, ndim 2, dtype family float — no output shape is stated: observe_signature.py recorded SymbolicTensor ndim 2, which disagrees with the fixture run
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
tf_segment_indicator_ Ran
ran on a synthesized input
brown-ivl/ConDor/ConDor/network_utils/sh_conv.py
code served (permissive licence) · get_code("b7f5b304241048ea")
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 EagerTensor, 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

def tf_heaviside(X):
    return tf.maximum(0.0, tf.sign(X))

def tf_segment_indicator_(X, a, b):
    return tf_heaviside(X - a) - tf_heaviside(X - b)

# Create small test tensors
X = tf.constant([[-2.0, -1.0, 0.5, 2.0],
                 [-1.5, 0.0, 1.5, 3.0]], dtype=tf.float32)

a = tf.constant(-1.0, dtype=tf.float32)
b = tf.constant(1.5, dtype=tf.float32)

ARGS = {
    "X": X,
    "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, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EagerTensor, 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
tf_segment_indictor Ran
ran on a synthesized input
brown-ivl/ConDor/ConDor/network_utils/sh_conv.py
code served (permissive licence) · get_code("645fc9c3ed7714e6")
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 EagerTensor, 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

# Create small test tensors
X = tf.constant([[-2.0, -1.0, 0.0, 1.0], [2.0, 3.0, 4.0, 5.0]], dtype=tf.float32)
r = tf.constant(1.5, dtype=tf.float32)
sigma = tf.constant(0.5, dtype=tf.float32)

ARGS = {
    "X": X,
    "r": r,
    "sigma": sigma
}
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 EagerTensor, 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
tf_unique_with_inverse Ran
ran on a synthesized input
brown-ivl/ConDor/ConDor/network_utils/group_points.py
code served (permissive licence) · get_code("cb175242ae058123")
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 tensorflow as tf

# The function takes a single parameter x, which should be a 1D tensor
# The function calls tf.unique on x, which requires a tensor

# Create a small 1D tensor with some duplicate values
x = tf.constant([1, 2, 2, 3, 1, 4], dtype=tf.int32)

ARGS = {
    "x": x
}
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 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
tf_unique_with_inverse Ran
ran on a synthesized input
brown-ivl/ConDor/ConDor/network_utils/pooling.py
code served (permissive licence) · get_code("30fc09c6037ff867")
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 tensorflow as tf

# Create a small 1D tensor with some repeated values
x = tf.constant([1, 2, 3, 2, 1, 4], dtype=tf.int32)

ARGS = {
    "x": x
}
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 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
var Ran
ran on a synthesized input
brown-ivl/ConDor/ConDor/evaluators/train_pointnet_cls2.py
code served (permissive licence) · get_code("9fb46aab78071109")
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 tensorflow as tf
import numpy as np

# Create a small test tensor
# Shape: (2, 4) - batch size 2, feature dimension 4
x = tf.constant([
    [1.0, 2.0, 3.0, 4.0],
    [5.0, 6.0, 7.0, 8.0]
], dtype=tf.float32)

ARGS = {
    "x": x,
    "axis_mean": 0,
    "axis_norm": 1
}
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 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_patches_grid Not yet run
unverified (harvested, no recorded run)
brown-ivl/ConDor/ConDor/network_utils/group_points.py
code served (permissive licence) · get_code("83e06b900f5f086d")
How it ran

no invocation recorded.

Also present in other papers' code

Of 19 samples lifted from this paper's linked repositories, 19 carry a function identity and 0 do not. Of the 19 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 19 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

is a self-supervised method that learns to Canonicalize the 3D orientation and position (3D pose) for full and partial shapes. (left) Our method takes un-canonicalized 3D point clouds (gray) from different categories as input and produces consistently canonicalized outputs (colored). (right) Our method can also operate on partial point clouds (missing part of shape shown only for visualization). In addition, ConDor can also learn consistent co-segmentation of shapes without supervision, visualized as colored parts.

For agents

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

get_harvested_code_for_paper("2201.07788")
get_code_for_paper("2201.07788")
have("2201.07788")

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

curl https://syntology.ai/api/ran/2201.07788.json

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

[![Syntology run record](https://syntology.ai/api/ran/2201.07788.svg)](https://syntology.ai/paper/2201.07788)

Connect an agent — have() is free.