SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2107.03502 · NeurIPS · 2021

CSDI: Conditional Score-based Diffusion Models for Probabilistic Time Series Imputation

Yang Song, Stefano Ermon, Jiaming Song, Yusuke Tashiro

arXiv · PDF · Open in the Atlas

Code that ran

12 of 19 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (4 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; 2 failures of our instrument; 7 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/2107.03502.json · badge: /api/ran/2107.03502.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
ermongroup/CSDI canonical (classifier label)
classifier basis: census
6 ran · 4 no recorded run
adrianatienza1996/DiffusionModels community
Papers with Code archive
5 ran · 3 no recorded run
airalcorn2/deformer community
Papers with Code archive
1 ran · 0 no recorded run
FunctionStatusWhere it lives
Conv1d Ran
ran on a synthesized input · constructed an object; no result was computed
adrianatienza1996/DiffusionModels/CSDI/model.py
code served (permissive licence) · get_code("5dfc5500207d1921")
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, Conv1d: 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 = {
    'channels_in': 3,
    'channels_out': 16,
    'kernel_size': 3,
    'use_act': True
}
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 Conv1d — 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 Conv1d 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
Conv1d_with_init Ran
ran; our contract draft was wrong, not the code
ermongroup/CSDI/diff_models.py
code served (permissive licence) · get_code("7340b482ffe44836")
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 Conv1d. 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 = {
    "in_channels": 3,
    "out_channels": 16,
    "kernel_size": 5
}
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 Conv1d — a Conv1d output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Conv1d 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
DEformerCSDI Ran
ran on a synthesized input · constructed an object; no result was computed
airalcorn2/deformer/deformer_csdi.py
pointer only (licence: NONE) · get_code("eba03a4da0243c1e")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, airalcorn2/deformer 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.

DiffusionEmbedding Ran
ran on a synthesized input
adrianatienza1996/DiffusionModels/CSDI/model.py
code served (permissive licence) · get_code("e810d3ec37cb56b4")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) 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 as nn
import torch.nn.functional as F

class DiffusionEmbedding(nn.Module):

    def __init__(self, num_steps, embedding_dim=128):
        super().__init__()
        embedding = self._build_embedding(num_steps, embedding_dim / 2)
        self.register_buffer('embedding', embedding)
        self.projection1 = nn.Linear(embedding_dim, embedding_dim)
        self.projection2 = nn.Linear(embedding_dim, embedding_dim)

    def forward(self, diffusion_step):
        x = self.embedding[diffusion_step]
        x = self.projection1(x)
        x = F.silu(x)
        x = self.projection2(x)
        x = F.silu(x)
        return x

    def _build_embedding(self, num_steps, dim):
        steps = torch.arange(num_steps).unsqueeze(1)
        frequencies = 10.0 ** (torch.arange(dim) / (dim - 1) * 4.0).unsqueeze(0)
        table = steps * frequencies
        table = torch.cat([torch.sin(table), torch.cos(table)], dim=1)
        return table


INIT_ARGS = {
    'num_steps': 16,
    'embedding_dim': 8
}

ARGS = {
    'diffusion_step': torch.tensor([0, 3, 7, 15])
}

if __name__ == '__main__':
    model = DiffusionEmbedding(**INIT_ARGS)
    output = model(**ARGS)
    print(f"Output shape: {output.shape}")
    print(f"Output: {output}")
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
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 observed this class with its constructor sizes canonicalised where it could, which the node does not record, so its shape is not stated as this setup block's
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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
FeedForwardNet Ran
ran on a synthesized input · constructed an object; no result was computed
adrianatienza1996/DiffusionModels/CSDI/model.py
code served (permissive licence) · get_code("be39f9c54f305b37")
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, FeedForwardNet: 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)

# Parameters for FeedForwardNet
model_dim = 8
do_prob = 0.1
wide_factor = 2

ARGS = {
    "model_dim": model_dim,
    "do_prob": do_prob,
    "wide_factor": wide_factor,
}
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 FeedForwardNet — 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 FeedForwardNet 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
MultiHead_Attention Ran
ran on a synthesized input
adrianatienza1996/DiffusionModels/CSDI/model.py
code served (permissive licence) · get_code("be774a732d955316")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned Tensor, ndim 3, 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 as nn
from einops.layers.torch import Rearrange

class MultiHead_Attention(nn.Module):

    def __init__(self, model_dim, number_heads, do_prob):
        super(MultiHead_Attention, self).__init__()
        self.number_heads = number_heads
        self.scale_factor = 1 / (model_dim / number_heads) ** 0.5
        self.att_drop_out = nn.Dropout(do_prob)
        self.output_drop_out = nn.Dropout(do_prob)
        self.block_output = nn.Linear(model_dim, model_dim)
        self.split_head = Rearrange('b l (h d) -> b h l d', h=self.number_heads)
        self.split_head_t = Rearrange('b l (h d) -> b h d l', h=self.number_heads)
        self.concat = Rearrange('b h l d -> b l (h d)')
        self.x_to_q = nn.Linear(model_dim, model_dim)
        self.x_to_k = nn.Linear(model_dim, model_dim)
        self.x_to_v = nn.Linear(model_dim, model_dim)

    def forward(self, q, k, v):
        q = self.split_head(self.x_to_q(q))
        k_transpose = self.split_head_t(self.x_to_k(k))
        v = self.split_head(self.x_to_v(v))
        attention = torch.matmul(q, k_transpose)
        attention = attention * self.scale_factor
        attention = attention.softmax(-1)
        output = self.att_drop_out(attention)
        output = torch.matmul(output, v)
        output = self.concat(output)
        output = self.block_output(output)
        return self.output_drop_out(output)


INIT_ARGS = {
    'model_dim': 4,
    'number_heads': 2,
    'do_prob': 0.1
}

ARGS = {
    'q': torch.randn(2, 3, 4),
    'k': torch.randn(2, 3, 4),
    'v': torch.randn(2, 3, 4)
}

if __name__ == '__main__':
    model = MultiHead_Attention(**INIT_ARGS)
    model.eval()
    output = model(**ARGS)
    print(f"Output shape: {output.shape}")
    print(f"Output: {output}")
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
Imports
einops, 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 — no output shape is stated: observe_signature.py observed this class with its constructor sizes canonicalised where it could, which the node does not record, so its shape is not stated as this setup block's
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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
TransformerLayer Ran
ran on a synthesized input · constructed an object; no result was computed
adrianatienza1996/DiffusionModels/CSDI/model.py
code served (permissive licence) · get_code("4a2a993673dda35a")
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, TransformerLayer: 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 einops.layers.torch import Rearrange

torch.manual_seed(0)

# Define the ARGS dictionary with parameters for TransformerLayer
ARGS = {
    'number_heads': 2,
    'model_dim': 8,
    'do_prob': 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 TransformerLayer — 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 TransformerLayer 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
calc_denominator Ran
ran, honoured the contract we drafted
ermongroup/CSDI/utils.py
pointer only (licence: MIT); the stored copy get_code serves, from xuangu-fang/bayotide, is pointer only (licence: NONE) · get_code("0eabe6591b009ad0")
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 xuangu-fang/bayotide, 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.

extract_hour Ran
ran on a synthesized input
ermongroup/CSDI/dataset_physio.py
code served (permissive licence) · get_code("a8551932723a26a9")
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 int, ndim 0, dtype family int, 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)

def extract_hour(x):
    h, _ = map(int, x.split(':'))
    return h

ARGS = {
    'x': '14:30'
}
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 int, ndim 0, dtype family int, 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
get_torch_trans Ran
ran; our contract draft was wrong, not the code
ermongroup/CSDI/diff_models.py
code served (permissive licence) · get_code("e706793d7d22ae44")
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) and returned TransformerEncoder. 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 = {
    "heads": 8,
    "layers": 1,
    "channels": 64,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type TransformerEncoder — a TransformerEncoder output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a TransformerEncoder 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
parse_data Ran
ran on a synthesized input
ermongroup/CSDI/dataset_physio.py
code served (permissive licence) · get_code("d166bfcd5d4c397e")
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 numpy as np
import pandas as pd

attributes = ['DiasABP', 'HR', 'Na', 'Lactate', 'NIDiasABP', 'PaO2', 'WBC', 'pH', 'Albumin', 'ALT', 'Glucose', 'SaO2', 'Temp', 'AST', 'Bilirubin', 'HCO3', 'BUN', 'RespRate', 'Mg', 'HCT', 'SysABP', 'FiO2', 'K', 'GCS', 'Cholesterol', 'NISysABP', 'TroponinT', 'MAP', 'TroponinI', 'PaCO2', 'Platelets', 'Urine', 'NIMAP', 'Creatinine', 'ALP']

# Create a minimal DataFrame that has the structure expected by parse_data
# The function calls x.set_index('Parameter').to_dict()['Value']
# So we need a DataFrame with 'Parameter' and 'Value' columns

data_dict = {
    'Parameter': ['DiasABP', 'HR', 'Na', 'Lactate', 'pH', 'Glucose'],
    'Value': [75.2, 82.5, 138.0, 1.8, 7.35, 145.0]
}

df = pd.DataFrame(data_dict)

ARGS = {
    'x': df
}
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 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
quantile_loss Ran
ran, honoured the contract we drafted
ermongroup/CSDI/utils.py
pointer only (licence: MIT); the stored copy get_code serves, from xuangu-fang/bayotide, is pointer only (licence: NONE) · get_code("ff7efc68e98e1d55")
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 xuangu-fang/bayotide, 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.

Add_and_Norm Not yet run
unverified (harvested, no recorded run)
adrianatienza1996/DiffusionModels/CSDI/model.py
code served (permissive licence) · get_code("aed37e235474038a")
How it ran

no invocation recorded.

CSDI Not yet run
unverified (harvested, no recorded run)
adrianatienza1996/DiffusionModels/CSDI/model.py
code served (permissive licence) · get_code("641efb52636a5e7c")
How it ran

no invocation recorded.

ResidualBlock Not yet run
unverified (harvested, no recorded run)
adrianatienza1996/DiffusionModels/CSDI/model.py
code served (permissive licence) · get_code("d7da28db98f57a53")
How it ran

no invocation recorded.

calc_quantile_CRPS Not yet run
unverified (harvested, no recorded run)
ermongroup/CSDI/utils.py
code served (permissive licence) · get_code("265f06f01c273de4")
How it ran

no invocation recorded.

get_dataloader Not yet run
unverified (harvested, no recorded run)
ermongroup/CSDI/dataset_forecasting.py
code served (permissive licence) · get_code("4206103b12cb44b7")
How it ran

no invocation recorded.

get_dataloader Not yet run
unverified (harvested, no recorded run)
ermongroup/CSDI/dataset_pm25.py
code served (permissive licence) · get_code("3af76bc0fdd74740")
How it ran

no invocation recorded.

parse_id Not yet run
unverified (harvested, no recorded run)
ermongroup/CSDI/dataset_physio.py
code served (permissive licence) · get_code("5c04f3d80eeca460")
How it ran

no invocation recorded.

Also present in other papers' code

Of 19 samples lifted from this paper's linked repositories, 19 carry a function identity and 0 do not. Of the 19 distinct functions with an identity, 8 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 8 later) and 11 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
Conv1d_with_init 6 other papers
6 later
2202.08516 (2022-02)
SAITS: Self-Attention-based Imputation for Time Series
2210.17128 (2022-10)
Diffusion models for missing value imputation in tabular data
2304.06707 (2023-04)
Toward Reliable Human Pose Forecasting with Uncertainty
and 3 more not named here
get_torch_trans 6 other papers
6 later
2202.08516 (2022-02)
SAITS: Self-Attention-based Imputation for Time Series
2210.17128 (2022-10)
Diffusion models for missing value imputation in tabular data
2304.06707 (2023-04)
Toward Reliable Human Pose Forecasting with Uncertainty
and 3 more not named here
calc_denominator 2 other papers
2 later
2308.14906 (2023-08)
BayOTIDE: Bayesian Online Multivariate Time Series Imputation with Functional Decomposition
2401.03006 (2024-01)
The Rise of Diffusion Models in Time-Series Forecasting
quantile_loss 2 other papers
2 later
2308.14906 (2023-08)
BayOTIDE: Bayesian Online Multivariate Time Series Imputation with Functional Decomposition
2401.03006 (2024-01)
The Rise of Diffusion Models in Time-Series Forecasting
calc_quantile_CRPS 1 other paper
1 later
2401.03006 (2024-01)
The Rise of Diffusion Models in Time-Series Forecasting
extract_hour 1 other paper
1 later
2202.08516 (2022-02)
SAITS: Self-Attention-based Imputation for Time Series
parse_data 1 other paper
1 later
2202.08516 (2022-02)
SAITS: Self-Attention-based Imputation for Time Series
parse_id 1 other paper
1 later
2202.08516 (2022-02)
SAITS: Self-Attention-based Imputation for Time Series

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 imputation of missing values in time series has many applications in healthcare and finance. While autoregressive models are natural candidates for time series imputation, score-based diffusion models have recently outperformed existing counterparts including autoregressive models in many tasks such as image generation and audio synthesis, and would be promising for time series imputation. In this paper, we propose Conditional Score-based Diffusion models for Imputation (CSDI), a novel time series imputation method that utilizes score-based diffusion models conditioned on observed data. Unlike existing score-based approaches, the conditional diffusion model is explicitly trained for imputation and can exploit correlations between observed values. On healthcare and environmental data, CSDI improves by 40-65% over existing probabilistic imputation methods on popular performance metrics. In addition, deterministic imputation by CSDI reduces the error by 5-20% compared to the state-of-the-art deterministic imputation methods. Furthermore, CSDI can also be applied to time series interpolation and probabilistic forecasting, and is competitive with existing baselines. The code is available at https://github.com/ermongroup/CSDI.

For agents

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

get_harvested_code_for_paper("2107.03502")
get_code_for_paper("2107.03502")
have("2107.03502")

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

curl https://syntology.ai/api/ran/2107.03502.json

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

[![Syntology run record](https://syntology.ai/api/ran/2107.03502.svg)](https://syntology.ai/paper/2107.03502)

Connect an agent — have() is free.