SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2008.04679 · 2020

ClimAlign: Unsupervised statistical downscaling of climate variables via normalizing flows

arXiv · PDF · Open in the Atlas

Code that ran

3 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 3 have no contract verdict; 0 failures of our instrument; 6 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/2008.04679.json · badge: /api/ran/2008.04679.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
bgroenks96/generative-downscaling canonical (classifier label)
classifier basis: census
3 ran · 6 no recorded run
FunctionStatusWhere it lives
prepare_downscaling_data Ran
ran on a synthesized input
bgroenks96/generative-downscaling/experiments/common.py
code served (permissive licence) · get_code("c6083e0961aa2a06")
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 _BatchDataset. 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 small tf.data.Dataset objects for data_lo and data_hi
# Each dataset should yield tensors of shape (4,) for simplicity
np.random.seed(0)

# Create data_lo dataset: 10 samples of shape (4,)
data_lo_array = np.random.randn(10, 4).astype(np.float32)
data_lo = tf.data.Dataset.from_tensor_slices(data_lo_array)

# Create data_hi dataset: 10 samples of shape (3,) - distinct dimension to avoid ambiguity
data_hi_array = np.random.randn(10, 3).astype(np.float32)
data_hi = tf.data.Dataset.from_tensor_slices(data_hi_array)

ARGS = {
    'data_lo': data_lo,
    'data_hi': data_hi,
    'batch_size': 2,
    'buffer_size': 10,
    'mode': 'supervised'
}
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 _BatchDataset — a _BatchDataset output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a _BatchDataset 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
split_path Ran
ran on a synthesized input
bgroenks96/generative-downscaling/gcsfs_map.py
code served (permissive licence) · get_code("452d13ab3040e70a")
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 torch

# The function split_path takes a single parameter: path
# It expects a string that may represent a GCS (Google Cloud Storage) path
# The function normalizes it by:
# 1. Removing 'gs://' prefix if present
# 2. Stripping trailing and leading slashes
# 3. Splitting on the first '/' to separate bucket from key

# Test with a typical GCS path
ARGS = {
    'path': 'gs://my-bucket/path/to/object'
}
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 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
upsample Ran
ran on a synthesized input
bgroenks96/generative-downscaling/experiments/common.py
code served (permissive licence) · get_code("9050978bc25385dd")
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 tensorflow as tf
import numpy as np

# Construct arguments for the upsample function
new_wt = 8  # new width
new_ht = 6  # new height
method = "bilinear"  # resizing method supported by tf.image.resize
scale_factor = 2.0  # scale factor for division

ARGS = {
    "new_wt": new_wt,
    "new_ht": new_ht,
    "method": method,
    "scale_factor": scale_factor,
}
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 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
create_srcnn Not yet run
unverified (harvested, no recorded run)
bgroenks96/generative-downscaling/baselines/dscnn.py
code served (permissive licence) · get_code("070adb2f85bca8a0")
How it ran

no invocation recorded.

create_vdsrcnn Not yet run
unverified (harvested, no recorded run)
bgroenks96/generative-downscaling/baselines/dscnn.py
code served (permissive licence) · get_code("7ac1667a359ea65a")
How it ran

no invocation recorded.

mixed_spatial_mse Not yet run
unverified (harvested, no recorded run)
bgroenks96/generative-downscaling/baselines/dscnn.py
code served (permissive licence) · get_code("be220899470ede88")
How it ran

no invocation recorded.

pacific_nw Not yet run
unverified (harvested, no recorded run)
bgroenks96/generative-downscaling/regions.py
code served (permissive licence) · get_code("cb52b4b309a79beb")
How it ran

no invocation recorded.

pacific_nw_inland Not yet run
unverified (harvested, no recorded run)
bgroenks96/generative-downscaling/regions.py
code served (permissive licence) · get_code("1478550fc98baa74")
How it ran

no invocation recorded.

southeast_us Not yet run
unverified (harvested, no recorded run)
bgroenks96/generative-downscaling/regions.py
code served (permissive licence) · get_code("7a5792cb04dd03e0")
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, 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 9 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 2026-11-21.

Repositories linked to this paper

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

Abstract

Downscaling is a landmark task in climate science and meteorology in which the goal is to use coarse scale, spatio-temporal data to infer values at finer scales. Statistical downscaling aims to approximate this task using statistical patterns gleaned from an existing dataset of downscaled values, often obtained from observations or physical models. In this work, we investigate the application of deep latent variable learning to the task of statistical downscaling. We present ClimAlign, a novel method for unsupervised, generative downscaling using adaptations of recent work in normalizing flows for variational inference. We evaluate the viability of our method using several different metrics on two datasets consisting of daily temperature and precipitation values gridded at low (1 degree latitude/longitude) and high (1/4 and 1/8 degree) resolutions. We show that our method achieves comparable predictive performance to existing supervised statistical downscaling methods while simultaneously allowing for both conditional and unconditional sampling from the joint distribution over high and low resolution spatial fields. We provide publicly accessible implementations of our method, as well as the baselines used for comparison, on GitHub.

For agents

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

get_harvested_code_for_paper("2008.04679")
get_code_for_paper("2008.04679")
have("2008.04679")

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

curl https://syntology.ai/api/ran/2008.04679.json

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

[![Syntology run record](https://syntology.ai/api/ran/2008.04679.svg)](https://syntology.ai/paper/2008.04679)

Connect an agent — have() is free.