SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2007.01332 · NeurIPS · 2020

Meta-Learning Stationary Stochastic Process Prediction with Convolutional Neural Processes

Andrew Foong, Wessel Bruinsma, Jonathan Gordon, Yann Dubois, James Requeima, Richard Turner

arXiv · PDF · Open in the Atlas

Code that ran

5 of 5 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 1 violated it and 4 have no contract verdict; 0 failures of our instrument; 0 no recorded run), measured 2026-11-21; 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/2007.01332.json · badge: /api/ran/2007.01332.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
YannDubs/Neural-Process-Family canonical (classifier label)
classifier basis: census
5 ran · 0 no recorded run
FunctionStatusWhere it lives
discard_ith_arg Ran
ran on a synthesized input
YannDubs/Neural-Process-Family/npf/architectures/encoders.py
code served (permissive licence) · get_code("0de02fe344a7769c")
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 function. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

# Create a simple module to pass as the `module` parameter
class SimpleModule(nn.Module):
    def __init__(self, *args, **kwargs):
        super().__init__()
        self.linear = nn.Linear(4, 3)
    
    def forward(self, *args, **kwargs):
        # The module receives filtered args
        if len(args) > 0:
            return self.linear(args[0])
        return torch.tensor([1.0, 2.0, 3.0])

# Define arguments for discard_ith_arg
ARGS = {
    "module": SimpleModule,
    "i": 1,
}
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 function — a function output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a function 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
get_exponential_decay_gamma Ran
ran on a synthesized input
YannDubs/Neural-Process-Family/utils/helpers.py
code served (permissive licence) · get_code("72e4538978e857c6")
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 float, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

ARGS = {
    "scheduling_factor": 10.0,
    "max_epochs": 100,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
load_all_results Ran
ran on a synthesized input
YannDubs/Neural-Process-Family/utils/helpers.py
code served (permissive licence) · get_code("3d99c27a829c98ab")
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 DataFrame. 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 os
import tempfile
import pandas as pd

# Create a temporary directory structure with test data
temp_dir = tempfile.mkdtemp()

# Create the directory structure: data/model/run_N/eval.csv
os.makedirs(os.path.join(temp_dir, "data1", "model1", "run_1"), exist_ok=True)
os.makedirs(os.path.join(temp_dir, "data1", "model1", "run_2"), exist_ok=True)
os.makedirs(os.path.join(temp_dir, "data1", "model2", "run_1"), exist_ok=True)
os.makedirs(os.path.join(temp_dir, "data2", "model1", "run_1"), exist_ok=True)

# Create eval.csv files with sample data
# The function reads these and computes .mean()[0]
pd.DataFrame([1.5]).to_csv(os.path.join(temp_dir, "data1", "model1", "run_1", "eval.csv"), header=False, index=False)
pd.DataFrame([2.5]).to_csv(os.path.join(temp_dir, "data1", "model1", "run_2", "eval.csv"), header=False, index=False)
pd.DataFrame([3.5]).to_csv(os.path.join(temp_dir, "data1", "model2", "run_1", "eval.csv"), header=False, index=False)
pd.DataFrame([4.5]).to_csv(os.path.join(temp_dir, "data2", "model1", "run_1", "eval.csv"), header=False, index=False)

ARGS = {
    "folder": temp_dir
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
pandas — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type DataFrame — a DataFrame output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a DataFrame output the recorded ndim 2 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
mean Ran
ran, violated the contract we drafted
YannDubs/Neural-Process-Family/utils/helpers.py
code served (permissive licence) · get_code("1ba480fa4f850c03")
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 float, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# The function `mean` takes a list-like object `l` and computes the average
# It uses: sum(l) and len(l)
# So `l` should be an array-like that supports sum() and len()

# Simple test case: a list of numbers
l = [1.0, 2.0, 3.0, 4.0, 5.0]

ARGS = {
    "l": l
}
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 float, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
sample_gp_dataset_like Ran
ran on a synthesized input
YannDubs/Neural-Process-Family/utils/ntbks_helpers.py
code served (permissive licence) · get_code("893d9056de856c0b")
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 MockGPDataset. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import copy

torch.manual_seed(0)

class MockGPDataset:
    """Mock GP dataset with required methods."""
    def __init__(self):
        self.samples = None
        self.data = torch.randn(2, 4)
    
    def get_samples(self, **kwargs):
        """Return mock samples as tuple of tensors."""
        return (torch.randn(2, 4), torch.randn(2, 3))
    
    def set_samples_(self, *args):
        """Set samples from arguments."""
        self.samples = args
    
    def __deepcopy__(self, memo):
        """Support deepcopy."""
        new_obj = MockGPDataset()
        new_obj.samples = self.samples
        new_obj.data = self.data.clone()
        return new_obj


dataset = MockGPDataset()

ARGS = {
    "dataset": dataset,
}
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 MockGPDataset — a MockGPDataset output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a MockGPDataset 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

Also present in other papers' code

Of 5 samples lifted from this paper's linked repositories, 5 carry a function identity and 0 do not. Of the 5 distinct functions with an identity, 1 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 1 later) and 4 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
mean 5 other papers
1 dated earlier, 4 later
1909.13584 (2019-09)
Interpretations are useful: penalizing explanations to align neural networks with prior knowledge
2009.12789 (2020-09)
Learning Optimal Representations with the Decodable Information Bottleneck
aaai_16202 (2021 (year only))
Graph and Temporal Convolutional Networks for 3D Multi-person Pose Estimation in Monocular Videos
and 2 more not named here

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 2026-11-21.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Stationary stochastic processes (SPs) are a key component of many probabilistic models, such as those for off-the-grid spatio-temporal data. They enable the statistical symmetry of underlying physical phenomena to be leveraged, thereby aiding generalization. Prediction in such models can be viewed as a translation equivariant map from observed data sets to predictive SPs, emphasizing the intimate relationship between stationarity and equivariance. Building on this, we propose the Convolutional Neural Process (ConvNP), which endows Neural Processes (NPs) with translation equivariance and extends convolutional conditional NPs to allow for dependencies in the predictive distribution. The latter enables ConvNPs to be deployed in settings which require coherent samples, such as Thompson sampling or conditional image completion. Moreover, we propose a new maximum-likelihood objective to replace the standard ELBO objective in NPs, which conceptually simplifies the framework and empirically improves performance. We demonstrate the strong performance and generalization capabilities of ConvNPs on 1D regression, image completion, and various tasks with real-world spatio-temporal data.

For agents

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

get_harvested_code_for_paper("2007.01332")
get_code_for_paper("2007.01332")
have("2007.01332")

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

curl https://syntology.ai/api/ran/2007.01332.json

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

[![Syntology run record](https://syntology.ai/api/ran/2007.01332.svg)](https://syntology.ai/paper/2007.01332)

Connect an agent — have() is free.