SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2203.14814 · 2022

Using Probabilistic Machine Learning to Better Model Temporal Patterns in Parameterizations: a case study with the Lorenz 96 model

arXiv · PDF · Open in the Atlas

Code that ran

5 of 9 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 5 have no contract verdict; 0 failures of our instrument; 4 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2203.14814.json · badge: /api/ran/2203.14814.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
raghul-parthipan/l96_rnn canonical (classifier label)
classifier basis not recorded on the link
5 ran · 4 no recorded run
FunctionStatusWhere it lives
U Ran
ran on a synthesized input
raghul-parthipan/l96_rnn/saved_models/gan/importance_sampler_for_gan_results.py
code served (permissive licence) · get_code("db6e1a26d5d8cce7")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 2, dtype family float, shape (2, 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

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

# Create small test arrays with distinct dimensions
# Xt: current state, shape (batch=2, dim=4)
Xt = np.random.randn(2, 4).astype(np.float32)

# Xt_1: previous/target state, shape (batch=2, dim=4)
Xt_1 = np.random.randn(2, 4).astype(np.float32)

# F: forcing term, shape (batch=2, dim=4)
F = np.random.randn(2, 4).astype(np.float32)

# dt: time step (default 0.005)
dt = 0.005

ARGS = {
    "Xt": Xt,
    "Xt_1": Xt_1,
    "F": F,
    "dt": dt
}
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 ndarray, 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
create_batch Ran
ran on a synthesized input
raghul-parthipan/l96_rnn/saved_models/rnn/models_for_full_dataset/helper.py
code served (permissive licence) · get_code("1d1a44a5a1e1e848")
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. 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

# Create sample input data - a list of numpy arrays
# The function indexes into the first dimension, so we need arrays with batch dimension
array1 = np.random.randn(10, 4)  # 10 samples, 4 features
array2 = np.random.randn(10, 3)  # 10 samples, 3 features
array3 = np.random.randn(10, 5)  # 10 samples, 5 dimensions

input_list = [array1, array2, array3]

ARGS = {
    "input_list": input_list,
    "batch_s": 32
}
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 list — a list 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
list_average Ran
ran on a synthesized input
raghul-parthipan/l96_rnn/saved_models/rnn/models_for_full_dataset/helper.py
code served (permissive licence) · get_code("2e72cd910d67557d")
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

torch.manual_seed(0)

# The function list_average takes a list of loss values and computes their average
# It calls sum() on the list and len() on the list
# Loss values are typically scalars (floats or 0-d tensors)

list_of_loss = [
    torch.tensor(0.5),
    torch.tensor(0.3),
    torch.tensor(0.7),
    torch.tensor(0.4)
]

ARGS = {
    "list_of_loss": list_of_loss
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 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
progress_bar Ran
ran on a synthesized input
raghul-parthipan/l96_rnn/saved_models/rnn/models_for_full_dataset/helper.py
code served (permissive licence) · get_code("109f314f67aceb55")
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
import numpy as np

ARGS = {
    "iteration": 5,
    "total": 20,
    "size": 30
}
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 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
u_deriver Ran
ran on a synthesized input
raghul-parthipan/l96_rnn/saved_models/gan/gan_training.py
code served (permissive licence) · get_code("cc4328be7b9a6090")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 2, dtype family float, shape (2, 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

# Construct arguments for u_deriver
# The function expects:
# - x_train: a 2D numpy array with shape (time_steps, spatial_dims)
# - F: a scalar or array that gets subtracted
# - save_time_step: a scalar time step value

# Create x_train with shape (3, 4) - 3 time steps, 4 spatial dimensions
# This allows x_train[:-1, :] to have shape (2, 4) and x_train[1:, :] to have shape (2, 4)
np.random.seed(0)
x_train = np.random.randn(3, 4).astype(np.float32)

# F is a scalar or array that gets subtracted from the rolled arrays
F = 0.5

# save_time_step is a scalar used for time stepping
save_time_step = 0.1

ARGS = {
    "x_train": x_train,
    "F": F,
    "save_time_step": save_time_step,
}
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 ndarray, 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
simulate_gan_param Not yet run
unverified (harvested, no recorded run)
raghul-parthipan/l96_rnn/saved_models/gan/generate_data.py
code served (permissive licence) · get_code("cf7449e77170f1b0")
How it ran

no invocation recorded.

simulate_gan_param_for_weather Not yet run
unverified (harvested, no recorded run)
raghul-parthipan/l96_rnn/saved_models/gan/generate_weather_data.py
code served (permissive licence) · get_code("ec97c4e3e0e0ca8e")
How it ran

no invocation recorded.

simulate_polynomial_param Not yet run
unverified (harvested, no recorded run)
raghul-parthipan/l96_rnn/saved_models/polynomial/generate_data.py
code served (permissive licence) · get_code("4b85b21065f4f51a")
How it ran

no invocation recorded.

simulate_polynomial_param_weather Not yet run
unverified (harvested, no recorded run)
raghul-parthipan/l96_rnn/saved_models/polynomial/generate_weather_data.py
code served (permissive licence) · get_code("2e3042c573a8de99")
How it ran

no invocation recorded.

Also present in other papers' code

Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 distinct functions with an identity, 3 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 3 later) and 6 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
create_batch 1 other paper
1 later
2210.04001 (2022-10)
Don't Waste Data: Transfer Learning to Leverage All Data for Machine-Learnt Climate Model Emulation
list_average 1 other paper
1 later
2210.04001 (2022-10)
Don't Waste Data: Transfer Learning to Leverage All Data for Machine-Learnt Climate Model Emulation
progress_bar 1 other paper
1 later
2210.04001 (2022-10)
Don't Waste Data: Transfer Learning to Leverage All Data for Machine-Learnt Climate Model Emulation

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

The modelling of small-scale processes is a major source of error in climate models, hindering the accuracy of low-cost models which must approximate such processes through parameterization. Red noise is essential to many operational parameterization schemes, helping model temporal correlations. We show how to build on the successes of red noise by combining the known benefits of stochasticity with machine learning. This is done using a physically-informed recurrent neural network within a probabilistic framework. Our model is competitive and often superior to both a bespoke baseline and an existing probabilistic machine learning approach (GAN) when applied to the Lorenz 96 atmospheric simulation. This is due to its superior ability to model temporal patterns compared to standard first-order autoregressive schemes. It also generalises to unseen scenarios. We evaluate across a number of metrics from the literature, and also discuss the benefits of using the probabilistic metric of hold-out likelihood.

For agents

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

get_harvested_code_for_paper("2203.14814")
get_code_for_paper("2203.14814")
have("2203.14814")

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

curl https://syntology.ai/api/ran/2203.14814.json

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

[![Syntology run record](https://syntology.ai/api/ran/2203.14814.svg)](https://syntology.ai/paper/2203.14814)

Connect an agent — have() is free.