SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2112.08440 · 2021

Climate-Invariant Machine Learning

arXiv · PDF · Open in the Atlas

Code that ran

4 of 16 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 3 have no contract verdict; 1 failure of our instrument; 12 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/2112.08440.json · badge: /api/ran/2112.08440.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
tbeucler/CBRAIN-CAM canonical (classifier label)
classifier basis not recorded on the link
4 ran · 12 no recorded run
FunctionStatusWhere it lives
log_loss Ran
ran on a synthesized input
tbeucler/CBRAIN-CAM/cbrain/losses.py
code served (permissive licence) · get_code("6eb107b3aee3be04")
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; 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

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

# Create small sample tensors for y_true and y_pred
# Using batch size 2, dimension 4 as specified
batch_size = 2
dim = 4

y_true = tf.constant(np.random.randn(batch_size, dim), dtype=tf.float32)
y_pred = tf.constant(np.random.randn(batch_size, dim), dtype=tf.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, 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 — no output shape is stated: observe_signature.py recorded SymbolicTensor ndim 1, 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
p_derivative Ran
ran on a synthesized input
tbeucler/CBRAIN-CAM/cbrain/equation_discovery.py
code served (permissive licence) · get_code("021131050e412be8")
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

# Create small test arrays with distinct dimensions
# batch_size=2, L=4 (number of pressure levels)
batch_size = 2
L = 4

# var: shape (batch_size, L) - the variable we're taking derivative of
var = np.random.randn(batch_size, L)

# pressure: shape (batch_size, L) - pressure levels (should be monotonic)
# Create monotonically decreasing pressure (typical for atmospheric data)
pressure = np.array([
    [1000, 850, 500, 100],
    [1000, 850, 500, 100]
], dtype=np.float64)

ARGS = {
    "var": var,
    "pressure": pressure
}
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
rmse Ran
ran on a synthesized input
tbeucler/CBRAIN-CAM/cbrain/losses.py
code served (permissive licence) · get_code("889048aa1db91c55")
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 SymbolicTensor. 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 tensors for y_true and y_pred
# Batch size: 2, dimensions: 4
y_true = tf.constant([[1.0, 2.0, 3.0, 4.0], [5.0, 6.0, 7.0, 8.0]], dtype=tf.float32)
y_pred = tf.constant([[1.1, 2.1, 2.9, 4.2], [4.9, 6.1, 7.1, 7.8]], dtype=tf.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, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type SymbolicTensor — a SymbolicTensor 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
total_error Ran
ran; our contract draft was wrong, not the code
tbeucler/CBRAIN-CAM/cbrain/losses.py
code served (permissive licence) · get_code("6319dcf221a42fc3")
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 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 tensorflow as tf
import tensorflow.keras.backend as K
import numpy as np

# Create small test tensors using TensorFlow/Keras backend
# The function takes y_true and y_pred as inputs
# Both should be tensors that support K.mean(), K.square(), and K.sum() operations

# Create y_true: shape (2, 4) - batch of 2 samples, 4 features
y_true_data = np.array([[1.0, 2.0, 3.0, 4.0],
                         [5.0, 6.0, 7.0, 8.0]], dtype=np.float32)
y_true = K.variable(y_true_data)

# Create y_pred: shape (2, 4) - same shape as y_true
y_pred_data = np.array([[1.5, 2.5, 3.5, 4.5],
                         [5.5, 6.5, 7.5, 8.5]], dtype=np.float32)
y_pred = K.variable(y_pred_data)

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, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EagerTensor, 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
act_layer Not yet run
unverified (harvested, no recorded run)
tbeucler/CBRAIN-CAM/cbrain/models.py
code served (permissive licence) · get_code("17416d46279a0a1c")
How it ran

no invocation recorded.

compute_bp Not yet run
unverified (harvested, no recorded run)
tbeucler/CBRAIN-CAM/cbrain/Test01_preprocess_aqua.py
code served (permissive licence) · get_code("8205d6f9a8646635")
How it ran

no invocation recorded.

compute_c Not yet run
unverified (harvested, no recorded run)
tbeucler/CBRAIN-CAM/cbrain/Test01_preprocess_aqua.py
code served (permissive licence) · get_code("2fda44420d3715eb")
How it ran

no invocation recorded.

compute_dP_tilde Not yet run
unverified (harvested, no recorded run)
tbeucler/CBRAIN-CAM/cbrain/layers.py
code served (permissive licence) · get_code("4c4e6f7352c04cea")
How it ran

no invocation recorded.

crop_ds Not yet run
unverified (harvested, no recorded run)
tbeucler/CBRAIN-CAM/cbrain/Test01_preprocess_aqua.py
code served (permissive licence) · get_code("1762bd0384d00b0a")
How it ran

no invocation recorded.

eice Not yet run
unverified (harvested, no recorded run)
tbeucler/CBRAIN-CAM/cbrain/climate_invariant.py
code served (permissive licence) · get_code("1a139da4f0ec4a88")
How it ran

no invocation recorded.

eliq Not yet run
unverified (harvested, no recorded run)
tbeucler/CBRAIN-CAM/cbrain/climate_invariant.py
code served (permissive licence) · get_code("095238f125e22ca6")
How it ran

no invocation recorded.

esat Not yet run
unverified (harvested, no recorded run)
tbeucler/CBRAIN-CAM/cbrain/climate_invariant.py
code served (permissive licence) · get_code("13759c9dcff1a945")
How it ran

no invocation recorded.

fc_model Not yet run
unverified (harvested, no recorded run)
tbeucler/CBRAIN-CAM/cbrain/models.py
code served (permissive licence) · get_code("019d6321d00e180b")
How it ran

no invocation recorded.

get_jacobian Not yet run
unverified (harvested, no recorded run)
tbeucler/CBRAIN-CAM/cbrain/model_diagnostics.py
code served (permissive licence) · get_code("12ddccc3bed4b967")
How it ran

no invocation recorded.

get_mean_sounding Not yet run
unverified (harvested, no recorded run)
tbeucler/CBRAIN-CAM/cbrain/model_diagnostics.py
code served (permissive licence) · get_code("d5d4134a4e35adb3")
How it ran

no invocation recorded.

p_der_spline Not yet run
unverified (harvested, no recorded run)
tbeucler/CBRAIN-CAM/cbrain/equation_discovery.py
code served (permissive licence) · get_code("7d126dea72401b86")
How it ran

no invocation recorded.

Also present in other papers' code

Of 16 samples lifted from this paper's linked repositories, 16 carry a function identity and 0 do not. Of the 16 distinct functions with an identity, 11 are also present in other papers whose code we hold (11 with at least one paper dated earlier, 0 the same month, 5 later) and 5 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
act_layer 3 other papers
2 dated earlier, 1 later
1806.04731 (2018-06)
Deep learning to represent sub-grid processes in climate models
1909.00912 (2019-09)
Enforcing Analytic Constraints in Neural-Networks Emulating Physical Systems
2402.03079 (2024-02)
Simulating Atmospheric Processes in Earth System Models and Quantifying Uncertainties with Deep Learning Multi-Member and Stochastic Parameterizations
fc_model 3 other papers
2 dated earlier, 1 later
1806.04731 (2018-06)
Deep learning to represent sub-grid processes in climate models
1909.00912 (2019-09)
Enforcing Analytic Constraints in Neural-Networks Emulating Physical Systems
2402.03079 (2024-02)
Simulating Atmospheric Processes in Earth System Models and Quantifying Uncertainties with Deep Learning Multi-Member and Stochastic Parameterizations
get_jacobian 3 other papers
2 dated earlier, 1 later
1806.04731 (2018-06)
Deep learning to represent sub-grid processes in climate models
1909.00912 (2019-09)
Enforcing Analytic Constraints in Neural-Networks Emulating Physical Systems
2402.03079 (2024-02)
Simulating Atmospheric Processes in Earth System Models and Quantifying Uncertainties with Deep Learning Multi-Member and Stochastic Parameterizations
get_mean_sounding 3 other papers
2 dated earlier, 1 later
1806.04731 (2018-06)
Deep learning to represent sub-grid processes in climate models
1909.00912 (2019-09)
Enforcing Analytic Constraints in Neural-Networks Emulating Physical Systems
2402.03079 (2024-02)
Simulating Atmospheric Processes in Earth System Models and Quantifying Uncertainties with Deep Learning Multi-Member and Stochastic Parameterizations
total_error 3 other papers
2 dated earlier, 1 later
1806.04731 (2018-06)
Deep learning to represent sub-grid processes in climate models
1909.00912 (2019-09)
Enforcing Analytic Constraints in Neural-Networks Emulating Physical Systems
2402.03079 (2024-02)
Simulating Atmospheric Processes in Earth System Models and Quantifying Uncertainties with Deep Learning Multi-Member and Stochastic Parameterizations
compute_bp 1 other paper
1 dated earlier
1806.04731 (2018-06)
Deep learning to represent sub-grid processes in climate models
compute_c 1 other paper
1 dated earlier
1806.04731 (2018-06)
Deep learning to represent sub-grid processes in climate models
compute_dP_tilde 1 other paper
1 dated earlier
1806.04731 (2018-06)
Deep learning to represent sub-grid processes in climate models
crop_ds 1 other paper
1 dated earlier
1806.04731 (2018-06)
Deep learning to represent sub-grid processes in climate models
log_loss 1 other paper
1 dated earlier
1806.04731 (2018-06)
Deep learning to represent sub-grid processes in climate models
rmse 1 other paper
1 dated earlier
1806.04731 (2018-06)
Deep learning to represent sub-grid processes in climate models

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

Projecting climate change is a generalization problem: we extrapolate the recent past using physical models across past, present, and future climates. Current climate models require representations of processes that occur at scales smaller than model grid size, which have been the main source of model projection uncertainty. Recent machine learning (ML) algorithms hold promise to improve such process representations, but tend to extrapolate poorly to climate regimes they were not trained on. To get the best of the physical and statistical worlds, we propose a new framework - termed "climate-invariant" ML - incorporating knowledge of climate processes into ML algorithms, and show that it can maintain high offline accuracy across a wide range of climate conditions and configurations in three distinct atmospheric models. Our results suggest that explicitly incorporating physical knowledge into data-driven models of Earth system processes can improve their consistency, data efficiency, and generalizability across climate regimes.

For agents

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

get_harvested_code_for_paper("2112.08440")
get_code_for_paper("2112.08440")
have("2112.08440")

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

curl https://syntology.ai/api/ran/2112.08440.json

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

[![Syntology run record](https://syntology.ai/api/ran/2112.08440.svg)](https://syntology.ai/paper/2112.08440)

Connect an agent — have() is free.