SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2301.04944 · CVPR · 2023

ViTs for SITS: Vision Transformers for Satellite Image Time Series

Stefanos Zafeiriou, Michail Tarasiou, Erik Chavez

arXiv · PDF · Open in the Atlas

Code that ran

11 of 11 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (3 of them constructed an object rather than computing a result; 2 honoured the contract we drafted, 0 violated it and 8 have no contract verdict; 1 failure of our instrument; 0 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/2301.04944.json · badge: /api/ran/2301.04944.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
VSainteuf/pastis-benchmark canonical (classifier label)
classifier basis not recorded on the link
8 ran · 0 no recorded run
michaeltrs/DeepSatModels community
Papers with Code archive
3 ran · 0 no recorded run
FunctionStatusWhere it lives
Attention Ran
ran on a synthesized input · constructed an object; no result was computed
michaeltrs/DeepSatModels/models/TSViT/TSViTcls.py
code served (permissive licence) · get_code("7ff8cc1149fcdb55")
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 an instance of its own class, Attention: the run constructed the object and did not call it, so no result was computed. 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
from torch import nn

torch.manual_seed(0)

# Create an instance of the Attention module
attention_module = nn.Module()

# Define the arguments for the Attention class __init__
# dim: input dimension
# heads: number of attention heads
# dim_head: dimension per head
# dropout: dropout rate

ARGS = {
    "dim": 64,
    "heads": 8,
    "dim_head": 8,
    "dropout": 0.1
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
einops, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Attention — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a Attention 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
TSViTcls Ran
ran on a synthesized input · constructed an object; no result was computed
michaeltrs/DeepSatModels/models/TSViT/TSViTcls.py
code served (permissive licence) · get_code("92778a46d60fa611")
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 an instance of its own class, TSViTcls: the run constructed the object and did not call it, so no result was computed. 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
from torch import einsum
import torch.nn.functional as F
from einops import rearrange, repeat
from einops.layers.torch import Rearrange

torch.manual_seed(0)

model_config = {
    'img_res': 16,
    'patch_size': 4,
    'num_classes': 2,
    'max_seq_len': 3,
    'dim': 8,
    'depth': 2,
    'temporal_depth': 2,
    'spatial_depth': 2,
    'heads': 2,
    'dim_head': 4,
    'dropout': 0.1,
    'emb_dropout': 0.1,
    'pool': 'cls',
    'scale_dim': 2,
    'num_channels': 4,
}

ARGS = {
    'model_config': model_config,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
einops, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type TSViTcls — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a TSViTcls 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
Transformer Ran
ran on a synthesized input · constructed an object; no result was computed
michaeltrs/DeepSatModels/models/TSViT/TSViTcls.py
code served (permissive licence) · get_code("3c8abb5d4b170a1d")
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 an instance of its own class, Transformer: the run constructed the object and did not call it, so no result was computed. 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)

ARGS = {
    "dim": 4,
    "depth": 2,
    "heads": 2,
    "dim_head": 2,
    "mlp_dim": 8,
    "dropout": 0.1,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
einops, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Transformer — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a Transformer 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
bbox Ran
ran on a synthesized input
VSainteuf/pastis-benchmark/code/centerness_ground_truth.py
code served (permissive licence) · get_code("0f29b0188aa76fe2")
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 int; 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 a small binary mask image
# The function expects a 2D numpy array (binary mask)
# We'll create a 5x5 binary mask with some True values
img = np.array([
    [0, 0, 0, 0, 0],
    [0, 1, 1, 0, 0],
    [0, 1, 1, 0, 0],
    [0, 0, 0, 0, 0],
    [0, 0, 0, 0, 0]
], dtype=bool)

ARGS = {
    'img': img
}
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 int — 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
pad_collate Ran
ran on a synthesized input
VSainteuf/pastis-benchmark/code/collate.py
code served (permissive licence) · get_code("ad13585f01796e87")
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 3, dtype family float, shape (2, 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 torch
import collections

torch.manual_seed(0)

# Create a batch of tensors with different sizes along dimension 0
# This tests the padding logic in pad_collate
tensor1 = torch.randn(2, 4)  # shape (2, 4)
tensor2 = torch.randn(3, 4)  # shape (3, 4) - different first dimension
batch = [tensor1, tensor2]

ARGS = {
    'batch': batch,
    'pad_value': 0
}
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 3, dtype family float, shape (2, 3, 4) — shape (2, 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
pad_tensor Ran
ran; our contract draft was wrong, not the code
VSainteuf/pastis-benchmark/code/collate.py
code served (permissive licence) · get_code("314a47dca812f4f4")
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 2, 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
import torch.nn.functional as F

torch.manual_seed(0)

# Create a small tensor to pad
# Using shape (2, 4) - batch size 2, feature dim 4
x = torch.randn(2, 4)

# Target length to pad to
l = 5

# Pad value (default is 0, but we can specify it)
pad_value = 0

ARGS = {
    "x": x,
    "l": l,
    "pad_value": pad_value
}
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 2, dtype family float — no output shape is stated: observe_signature.py could not observe it (read:AttributeError)
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
prepare_dates Ran
ran, honoured the contract we drafted
VSainteuf/pastis-benchmark/code/dataloader.py
code served (permissive licence) · get_code("776464eb827fd3e0")
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 1, dtype family int, shape (2,) 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.

from datetime import datetime
import pandas as pd

# Create sample date dictionary
# Keys can be anything, values should be integers in YYYYMMDD format
date_dict = {
    'date1': 20230115,
    'date2': 20231225
}

# Create a reference date
reference_date = datetime(2023, 1, 1)

ARGS = {
    'date_dict': date_dict,
    'reference_date': reference_date
}
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 ndarray, ndim 1, dtype family int, shape (2,) — shape (2,) 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
prepare_dates Ran
ran, honoured the contract we drafted
VSainteuf/pastis-benchmark/code/dataloader_pixelset.py
code served (permissive licence) · get_code("9a101315e3da556d")
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 1, dtype family int, shape (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.

from datetime import datetime
import pandas as pd

# Construct valid arguments for prepare_dates function
date_dict = {
    'obs_1': 20230115,
    'obs_2': 20231225,
    'obs_3': 20220630
}

reference_date = datetime(2023, 1, 1)

ARGS = {
    'date_dict': date_dict,
    'reference_date': reference_date
}
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 ndarray, ndim 1, dtype family int, shape (3,) — shape (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
repeat_pixel_and_norm Ran
ran on a synthesized input
VSainteuf/pastis-benchmark/code/dataloader_pixelset.py
code served (permissive licence) · get_code("9fbe1d414824dd88")
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. 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

torch.manual_seed(0)

# Case 1: pixels with fewer elements than n_pixel, with normalization
pixels_small = torch.randn(2, 4, 3)  # batch=2, channels=4, pixels=3
n_pixel = 8
norm = (torch.randn(4), torch.ones(4) * 0.5)  # mean and std per channel

ARGS = {
    "pixels": pixels_small,
    "n_pixel": n_pixel,
    "norm": norm,
}
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 tuple — a tuple 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
sample_pixels Ran
ran on a synthesized input
VSainteuf/pastis-benchmark/code/dataloader_pixelset.py
code served (permissive licence) · get_code("97b25d6ee2383bc4")
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 3, dtype family float, shape (4, 3, 5) 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 a small sample pixels array
# Shape: (height, width, num_pixels)
# Using distinct dimensions: 4 x 3 x 10
pixels = np.random.rand(4, 3, 10).astype(np.float32)

# Define n_pixel to be less than the number of pixels in the last dimension
# This will trigger the sampling branch
n_pixel = 5

ARGS = {
    "pixels": pixels,
    "n_pixel": n_pixel,
}
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 3, dtype family float, shape (4, 3, 5) — shape (4, 3, 5) 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
splat Ran
ran on a synthesized input
VSainteuf/pastis-benchmark/code/centerness_ground_truth.py
code served (permissive licence) · get_code("34c4d665e9379f63")
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 (16, 24) 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

ARGS = {
    "shape": (32, 48),
    "center": (16, 24),
    "dims": (8, 12),
    "stride": 2,
    "r": 3,
    "alpha": 0.7,
    "homo": False,
}
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 (16, 24) — shape (16, 24) 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

Also present in other papers' code

Of 11 samples lifted from this paper's linked repositories, 11 carry a function identity and 0 do not. Of the 10 distinct functions with an identity, 4 are also present in other papers whose code we hold (2 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
Attention
brick of this paper's assembly: helper
6 other papers
4 dated earlier, 2 later · 2 only as an assembly brick
2103.11816 (2021-03)
Incorporating Convolution Designs into Visual Transformers
2103.14899 (2021-03)
CrossViT: Cross-Attention Multi-Scale Vision Transformer for Image Classification
2103.15691 (2021-03)
ViViT: A Video Vision Transformer
and 3 more not named here
Transformer
brick of this paper's assembly: model
2 other papers
2 dated earlier
2103.14899 (2021-03)
CrossViT: Cross-Attention Multi-Scale Vision Transformer for Image Classification
2103.15691 (2021-03)
ViViT: A Video Vision Transformer
pad_tensor 2 other papers
2 later
2305.13277 (2023-05)
U-TILISE: A Sequence-to-sequence Model for Cloud Removal in Optical Satellite Time Series
2310.13336 (2023-10)
FLAIR: a Country-Scale Land Cover Semantic Segmentation Dataset From Multi-Source Optical Imagery
prepare_dates 1 other paper
1 later
2405.19055 (2024-05)
FUSU: A Multi-temporal-source Land Use Change Segmentation Dataset for Fine-grained Urban Semantic Understanding

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

In this paper we introduce the Temporo-Spatial Vision Transformer (TSViT), a fully-attentional model for general Satellite Image Time Series (SITS) processing based on the Vision Transformer (ViT). TSViT splits a SITS record into non-overlapping patches in space and time which are tokenized and subsequently processed by a factorized temporo-spatial encoder. We argue, that in contrast to natural images, a temporal-then-spatial factorization is more intuitive for SITS processing and present experimental evidence for this claim. Additionally, we enhance the model's discriminative power by introducing two novel mechanisms for acquisition-time-specific temporal positional encodings and multiple learnable class tokens. The effect of all novel design choices is evaluated through an extensive ablation study. Our proposed architecture achieves stateof-the-art performance, surpassing previous approaches by a significant margin in three publicly available SITS semantic segmentation and classification datasets. All model, training and evaluation codes can be found at https://github.com/michaeltrs/DeepSatModels.

For agents

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

get_harvested_code_for_paper("2301.04944")
get_code_for_paper("2301.04944")
have("2301.04944")

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

curl https://syntology.ai/api/ran/2301.04944.json

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

[![Syntology run record](https://syntology.ai/api/ran/2301.04944.svg)](https://syntology.ai/paper/2301.04944)

Connect an agent — have() is free.