SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · aaai_25976 · AAAI · 2023

Spatio-Temporal Meta-Graph Learning for Traffic Forecasting

Renhe Jiang, Xuan Song, Zhaonan Wang, Toyotaro Suzumura, Jiawei Yong, Shintaro Fukushima, Puneet Jeph, Quanjun Chen, Yasumasa Kobayashi

Open in the Atlas

Code that ran

13 of 17 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, 2 violated it and 7 have no contract verdict; 4 failures of our instrument; 4 no recorded run), measured 2026-11-22; 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/aaai_25976.json · badge: /api/ran/aaai_25976.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
deepkashiwa20/MegaCRN canonical (classifier label)
classifier basis: census
13 ran · 4 no recorded run
FunctionStatusWhere it lives
MSE Ran
ran, violated the contract we drafted
deepkashiwa20/MegaCRN/model/metrics.py
pointer only (licence: MIT); the stored copy get_code serves, from xdzhelheim/himnet, is pointer only (licence: NONE) · get_code("0a7d2992d05685eb")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from xdzhelheim/himnet, and GitHub detected no licence in that repository 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.

MSE Ran
ran on a synthesized input
deepkashiwa20/MegaCRN/model_EXPYTKY/metrics.py
code served (permissive licence) · get_code("1ede8e979a80e0d2")
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 float32, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Create small test arrays for y_true and y_pred
# The function performs element-wise operations and masking
y_true = np.array([
    [1.5, 0.0, 2.3, 0.00001],
    [0.5, 1.2, 0.00001, 3.1]
], dtype=np.float32)

y_pred = np.array([
    [1.4, 0.0, 2.5, 0.00002],
    [0.6, 1.1, 0.00001, 3.0]
], dtype=np.float32)

ARGS = {
    'y_true': y_true,
    'y_pred': y_pred
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float32, 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
RMSE Ran
ran, violated the contract we drafted
deepkashiwa20/MegaCRN/model/metrics.py
pointer only (licence: MIT); the stored copy get_code serves, from bdi-lab/spot-mamba, is pointer only (licence: NOASSERTION) · get_code("602913d134538363")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from bdi-lab/spot-mamba, and GitHub could not classify the licence of that repository (NOASSERTION). 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.

RMSE Ran
ran on a synthesized input
deepkashiwa20/MegaCRN/model_EXPYTKY/metrics.py
code served (permissive licence) · get_code("10ccb12f6ac2f295")
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 float32, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Create small numpy arrays for y_true and y_pred
# The function expects numpy arrays and performs element-wise operations
np.random.seed(0)

y_true = np.random.rand(4, 3).astype(np.float32)
y_pred = np.random.rand(4, 3).astype(np.float32)

ARGS = {
    "y_true": y_true,
    "y_pred": y_pred
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float32, 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
evaluate Ran
ran on a synthesized input
deepkashiwa20/MegaCRN/model/metrics.py
code served (permissive licence) · get_code("1ff1a78719215bef")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Create sample data for y_true and y_pred
np.random.seed(0)

# Create y_true with some non-zero values and some zeros
y_true = np.array([
    [1.0, 2.0, 0.0, 3.0],
    [0.0, 1.5, 2.5, 0.0]
], dtype=np.float32)

# Create y_pred with similar shape but different values
y_pred = np.array([
    [1.1, 1.9, 0.5, 3.2],
    [0.2, 1.6, 2.3, 0.1]
], dtype=np.float32)

ARGS = {
    'y_true': y_true,
    'y_pred': y_pred,
    'precision': 10
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
evaluate Ran
ran on a synthesized input
deepkashiwa20/MegaCRN/model_EXPYTKY/metrics.py
code served (permissive licence) · get_code("62b92919279383bc")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

# Create sample data for evaluation
np.random.seed(0)

# Create y_true and y_pred as numpy arrays with some variation
# Using shape (8,) to keep it simple and small
y_true = np.random.rand(8).astype(np.float32) * 10
y_pred = np.random.rand(8).astype(np.float32) * 10

# Ensure some non-zero values for the mask to work properly
y_true = np.array([0.5, 1.2, 3.4, 0.0, 2.1, 0.8, 4.5, 0.0], dtype=np.float32)
y_pred = np.array([0.6, 1.3, 3.2, 0.1, 2.0, 0.9, 4.6, 0.05], dtype=np.float32)

ARGS = {
    'y_true': y_true.copy(),
    'y_pred': y_pred.copy(),
    'precision': 10
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, dtype family float — a tuple output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
getDayTimestamp Ran
ran on a synthesized input
deepkashiwa20/MegaCRN/model/utils.py
code served (permissive licence) · get_code("d4b5951aefb94d6f")
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 (10, 3) 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 DataFrame with a DatetimeIndex for the data parameter
dates = pd.date_range('2023-01-02 00:00', periods=10, freq='5min')
data_values = np.random.randn(10, 3)
data = pd.DataFrame(data_values, index=dates)

ARGS = {
    'data': data
}
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 (10, 3) — shape (10, 3) 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
getTimestamp Ran
ran on a synthesized input
deepkashiwa20/MegaCRN/model/utils.py
code served (permissive licence) · get_code("e7d2c137521db22e")
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, 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 numpy as np
import pandas as pd

# Create a small DataFrame with datetime index
# The function expects data.index.values to be datetime64 objects
dates = pd.date_range('2024-01-01', periods=3, freq='D')
# Add intraday times by creating timestamps with time components
timestamps = pd.DatetimeIndex([
    '2024-01-01 06:00:00',
    '2024-01-01 12:30:00',
    '2024-01-02 08:15:00',
])

# Create data with 3 samples and 4 nodes
data_values = np.random.randn(3, 4)
data = pd.DataFrame(data_values, index=timestamps)

ARGS = {
    'data': data
}
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, 4) — shape (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
load_pickle Ran
ran; our contract draft was wrong, not the code
deepkashiwa20/MegaCRN/model_EXPYTKY/utils.py
pointer only (licence: MIT); the stored copy get_code serves, from coco0106/MO-STEP, is pointer only (licence: NONE) · get_code("cd17080a788963f3")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from coco0106/MO-STEP, and GitHub detected no licence in that repository 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.

masked_mae Ran
ran on a synthesized input
deepkashiwa20/MegaCRN/model_EXPYTKY/utils.py
code served (permissive licence) · get_code("90dd312dfe854032")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Tensor, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)

# Create small tensors for predictions and labels
batch_size = 2
num_features = 4

preds = torch.randn(batch_size, num_features)
labels = torch.randn(batch_size, num_features)

# Define ARGS dict with all parameters
ARGS = {
    "preds": preds,
    "labels": labels,
    "null_val": 0.0
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
masked_mae_loss Ran
ran; our contract draft was wrong, not the code
deepkashiwa20/MegaCRN/model/GTS.py
pointer only (licence: MIT); the stored copy get_code serves, from lingbai-kong/causalformer, is pointer only (licence: GPL-3.0) · get_code("2a82a90d91ced4b6")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from lingbai-kong/causalformer, and GPL-3.0, as GitHub detects it on that repository, 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. A run is recorded.

masked_mape_loss Ran
ran; our contract draft was wrong, not the code
deepkashiwa20/MegaCRN/model/GTS.py
pointer only (licence: MIT); the stored copy get_code serves, from beautyonce/sgode, is pointer only (licence: NONE) · get_code("0bf4404d22e1c154")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from beautyonce/sgode, and GitHub detected no licence in that repository 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.

masked_rmse_loss Ran
ran; our contract draft was wrong, not the code
deepkashiwa20/MegaCRN/model/GTS.py
pointer only (licence: MIT); the stored copy get_code serves, from beautyonce/sgode, is pointer only (licence: NONE) · get_code("9daca4e67045f36c")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from beautyonce/sgode, and GitHub detected no licence in that repository 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.

evaluateModel Not yet run
unverified (harvested, no recorded run)
deepkashiwa20/MegaCRN/model_EXPYTKY/traintest_MegaCRN.py
code served (permissive licence) · get_code("4625fd603dd1890c")
How it ran

no invocation recorded.

generate_graph_seq2seq_io_data Not yet run
unverified (harvested, no recorded run)
deepkashiwa20/MegaCRN/generate_training_data.py
code served (permissive licence) · get_code("fde1716c2510a3a6")
How it ran

no invocation recorded.

getDayTimestamp_ Not yet run
unverified (harvested, no recorded run)
deepkashiwa20/MegaCRN/model/utils.py
code served (permissive licence) · get_code("402cc696f39ab96a")
How it ran

no invocation recorded.

prepare_x_y Not yet run
unverified (harvested, no recorded run)
deepkashiwa20/MegaCRN/model/traintest_MegaCRN.py
code served (permissive licence) · get_code("8f6da201324530dd")
How it ran

no invocation recorded.

Also present in other papers' code

Of 17 samples lifted from this paper's linked repositories, 17 carry a function identity and 0 do not. Of the 16 distinct functions with an identity, 8 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 4 later) and 8 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
load_pickle 7 other papers
7 later
2401.03138 (2024-01)
TelTrans: Applying Multi-Type Telecom Data to Transportation Evaluation and Prediction via Multifaceted Graph Modeling
2402.00397 (2024-02)
Multi-scale Traffic Pattern Bank for Cross-city Few-shot Traffic Forecasting
2402.02005 (2024-02)
Topology-Informed Graph Transformer
and 4 more not named here
masked_mae_loss 5 other papers
2 later, 3 same year, order not determinable
2304.14343 (2023-04)
LibCity: A Unified Library Towards Efficient and Comprehensive Urban Spatial-Temporal Prediction
2310.06119 (2023-10)
Exploring Progress in Multivariate Time Series Forecasting: Comprehensive Benchmarking and Heterogeneity Analysis
2312.11198 (2023-12)
Signed Graph Neural Ordinary Differential Equation for Modeling Continuous-time Dynamics
and 2 more not named here
generate_graph_seq2seq_io_data 4 other papers
3 dated earlier, 1 same year, order not determinable
1707.01926 (2017-07)
Diffusion Convolutional Recurrent Neural Network: Data-Driven Traffic Forecasting
2101.06861 (2021-01)
Discrete Graph Structure Learning for Forecasting Multiple Time Series
2211.14701 (2022-11)
Spatio-Temporal Meta-Graph Learning for Traffic Forecasting
and 1 more not named here
RMSE 3 other papers
3 later
2402.02005 (2024-02)
Topology-Informed Graph Transformer
2405.10800 (2024-05)
Heterogeneity-Informed Meta-Parameter Learning for Spatiotemporal Time Series Forecasting
2406.11244 (2024-06)
SpoT-Mamba: Learning Long-Range Dependency on Spatio-Temporal Graphs with Selective State Spaces
MSE 2 other papers
2 later
2402.02005 (2024-02)
Topology-Informed Graph Transformer
2405.10800 (2024-05)
Heterogeneity-Informed Meta-Parameter Learning for Spatiotemporal Time Series Forecasting
masked_mape_loss 1 other paper
1 same year, order not determinable
2312.11198 (2023-12)
Signed Graph Neural Ordinary Differential Equation for Modeling Continuous-time Dynamics
masked_rmse_loss 1 other paper
1 same year, order not determinable
2312.11198 (2023-12)
Signed Graph Neural Ordinary Differential Equation for Modeling Continuous-time Dynamics
prepare_x_y 1 other paper
1 dated earlier
2211.14701 (2022-11)
Spatio-Temporal Meta-Graph Learning for Traffic Forecasting

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

Traffic forecasting as a canonical task of multivariate time series forecasting has been a significant research topic in AI community. To address the spatio-temporal heterogeneity and non-stationarity implied in the traffic stream, in this study, we propose Spatio-Temporal Meta-Graph Learning as a novel Graph Structure Learning mechanism on spatiotemporal data. Specifically, we implement this idea into Meta-Graph Convolutional Recurrent Network (MegaCRN) by plugging the Meta-Graph Learner powered by a Meta-Node Bank into GCRN encoder-decoder. We conduct a comprehensive evaluation on two benchmark datasets (i.e., METR-LA and PEMS-BAY) and a new large-scale traffic speed dataset called EXPY-TKY that covers 1843 expressway road links in Tokyo. Our model outperformed the stateof-the-arts on all three datasets. Besides, through a series of qualitative evaluations, we demonstrate that our model can explicitly disentangle the road links and time slots with different patterns and be robustly adaptive to any anomalous traffic situations. Codes and datasets are available at https: //github.com/deepkashiwa20/MegaCRN.

For agents

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

get_harvested_code_for_paper("aaai_25976")
get_code_for_paper("aaai_25976")
have("aaai_25976")

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

curl https://syntology.ai/api/ran/aaai_25976.json

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

[![Syntology run record](https://syntology.ai/api/ran/aaai_25976.svg)](https://syntology.ai/paper/aaai_25976)

Connect an agent — have() is free.