SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2303.04743 · 2023

Vector Quantized Time Series Generation with a Bidirectional Prior Model

arXiv · PDF · Open in the Atlas

Code that ran

16 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, 2 violated it and 13 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/2303.04743.json · badge: /api/ran/2303.04743.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
ml4its/timevqvae canonical (classifier label)
classifier basis not recorded on the link
9 ran · 0 no recorded run
danelee2601/supervised-fcn canonical (classifier label)
classifier basis not recorded on the link
7 ran · 0 no recorded run
FunctionStatusWhere it lives
FeedForward Ran
ran on a synthesized input
ml4its/timevqvae/src/timevqvae/models/bidirectional_transformer.py
code served (permissive licence) · get_code("38b01d4443dbe5d7")
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 Sequential. 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.nn.utils import weight_norm

torch.manual_seed(0)

ARGS = {
    "dim": 64,
    "mult": 4,
}
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 Sequential — a Sequential output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Sequential output the recorded ndim 1 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
calculate_fid Ran
ran on a synthesized input
danelee2601/supervised-fcn/supervised_FCN/example_compute_FID.py
code served (permissive licence) · get_code("2ddb5d4fd13cf25d")
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 float64, ndim 0, 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 torch

torch.manual_seed(0)
np.random.seed(0)

# z1 and z2 should be 2D arrays (batch_size, feature_dim)
# The function calls:
# - z.mean(axis=0) -> needs axis parameter
# - cov(z, rowvar=False) -> needs 2D array with samples as rows
# - matrix operations like .dot(), .sum()
# - trace() on the result

z1 = np.random.randn(2, 4).astype(np.float32)
z2 = np.random.randn(2, 4).astype(np.float32)

ARGS = {
    "z1": z1,
    "z2": z2,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, scipy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, 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
calculate_inception_score Ran
ran on a synthesized input
danelee2601/supervised-fcn/supervised_FCN/example_compute_IS.py
code served (permissive licence) · get_code("01a41e038c0e0ec8")
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 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

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

# P_yx should be a 2D numpy array of shape (batch_size, num_classes)
# representing probability distributions over classes for each sample
# The function divides this into n_split parts and computes KL divergence
batch_size = 20
num_classes = 4
P_yx = np.random.dirichlet(np.ones(num_classes), size=batch_size)

ARGS = {
    "P_yx": P_yx,
    "n_split": 10,
    "shuffle": True,
    "eps": 1e-16,
}
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 float — 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
calculate_padding Ran
ran on a synthesized input
ml4its/timevqvae/src/timevqvae/models/bidirectional_transformer.py
code served (permissive licence) · get_code("4125059c0fec4bb8")
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 math

ARGS = {
    "kernel_size": 3,
    "stride": 1,
    "dilation": 1,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — 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
compute_avg_outs Ran
ran on a synthesized input
danelee2601/supervised-fcn/supervised_FCN/experiments/exp_base.py
code served (permissive licence) · get_code("b720260f0238b6b5")
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 dict. 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)

# The function expects outs to be a list of dicts where each dict contains
# numeric values (floats, ints, or tensors) that can be summed and divided.
# It computes the mean across all dicts in the list.

outs = [
    {
        "loss": 0.5,
        "accuracy": 0.85,
        "metric_1": torch.tensor(1.2),
    },
    {
        "loss": 0.3,
        "accuracy": 0.90,
        "metric_1": torch.tensor(1.8),
    },
    {
        "loss": 0.4,
        "accuracy": 0.88,
        "metric_1": torch.tensor(1.5),
    },
]

ARGS = {
    "outs": outs,
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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_same_padding Ran
ran on a synthesized input
danelee2601/supervised-fcn/supervised_FCN/models/fcn.py
code served (permissive licence) · get_code("e1a3c4d15be8ec0d")
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, 5, 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 torch.nn.functional as F

torch.manual_seed(0)

# Construct small tensors for conv1d_same_padding
# input: (batch=2, channels=3, length=4)
input = torch.randn(2, 3, 4)

# weight: (out_channels=5, in_channels//groups=3, kernel_size=3)
weight = torch.randn(5, 3, 3)

# bias: (out_channels=5,)
bias = torch.randn(5)

# stride: tuple with one element
stride = (1,)

# dilation: tuple with one element
dilation = (1,)

# groups: number of groups for grouped convolution
groups = 1

ARGS = {
    "input": input,
    "weight": weight,
    "bias": bias,
    "stride": stride,
    "dilation": dilation,
    "groups": groups,
}
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, 5, 4) — shape (2, 5, 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
default Ran
ran, violated the contract we drafted
ml4its/timevqvae/src/timevqvae/vector_quantization/vq.py
pointer only (licence: MIT); the stored copy get_code serves, from ThomasMrY/VCT, is pointer only (licence: NONE) · get_code("60fff7c3c400d7ff")
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 ThomasMrY/VCT, 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.

exists Ran
ran, violated the contract we drafted
ml4its/timevqvae/src/timevqvae/models/bidirectional_transformer.py
pointer only (licence: MIT); the stored copy get_code serves, from ThomasMrY/VCT, is pointer only (licence: NONE) · get_code("aa5486a3650902d8")
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 ThomasMrY/VCT, 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_data Ran
ran on a synthesized input
danelee2601/supervised-fcn/supervised_FCN/preprocessing/preprocess_uea.py
code served (permissive licence) · get_code("7509c254e5b132f7")
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 numpy as np
import torch

# Create mock data that matches what extract_data expects
# The function iterates over data, expecting (t_data, t_label) tuples
# t_data should be iterable of objects with .tolist() method (torch tensors or numpy arrays)
# t_label should have .decode('utf-8') method (bytes)

torch.manual_seed(0)

# Create sample tensors and labels
# t_data will be a list of tensors, t_label will be bytes
sample_tensor_1 = torch.randn(4, 3)  # 4 samples, 3 features
sample_tensor_2 = torch.randn(4, 3)
sample_label_1 = b'label_one'

sample_tensor_3 = torch.randn(4, 3)
sample_tensor_4 = torch.randn(4, 3)
sample_label_2 = b'label_two'

# Construct data as list of tuples (data, label)
# where data is a list/tuple of tensors and label is bytes
data = [
    ([sample_tensor_1, sample_tensor_2], sample_label_1),
    ([sample_tensor_3, sample_tensor_4], sample_label_2)
]

ARGS = {
    'data': data
}
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
get_log_items_epoch Ran
ran on a synthesized input
danelee2601/supervised-fcn/supervised_FCN/experiments/exp_base.py
code served (permissive licence) · get_code("b8603cb8e3f0fe35")
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 dict. 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)

# Analyzing the function:
# - kind: str - used to construct dictionary keys
# - current_epoch: int - used as a value in the log_items dict
# - mean_outs: dict - iterated over to create log_items_

ARGS = {
    'kind': 'train',
    'current_epoch': 5,
    'mean_outs': {
        'loss': 0.42,
        'accuracy': 0.95,
        'f1_score': 0.88
    }
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
get_log_items_global_step Ran
ran on a synthesized input
danelee2601/supervised-fcn/supervised_FCN/experiments/exp_base.py
code served (permissive licence) · get_code("4dc7e9755b9f67ef")
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 dict. 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

torch.manual_seed(0)

ARGS = {
    'kind': 'train',
    'global_step': 42,
    'out': {
        'loss': 0.5,
        'accuracy': 0.95,
        'learning_rate': 0.001
    }
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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
l2norm Ran
ran; our contract draft was wrong, not the code
ml4its/timevqvae/src/timevqvae/vector_quantization/vq.py
pointer only (licence: MIT); the stored copy get_code serves, from jy0205/LaVIT, is pointer only (licence: NOASSERTION) · get_code("eca6cdf05972a95e")
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 jy0205/LaVIT, and GitHub could not classify the licence of that repository (NOASSERTION). 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.

quantize Ran
ran on a synthesized input
ml4its/timevqvae/src/timevqvae/utils/nn.py
code served (permissive licence) · get_code("e323b33b21b740e9")
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 torch.nn as nn
from typing import Union

torch.manual_seed(0)

class MockVQModel(nn.Module):
    """Minimal VQ model that returns required outputs"""
    def __init__(self):
        super().__init__()
        self.embedding = nn.Embedding(512, 4)
    
    def forward(self, z, svq_temp: Union[float, None] = None):
        # z shape: (batch, length, channels) or (batch, length, channels)
        batch_size = z.shape[0]
        seq_len = z.shape[1]
        
        # Return quantized (same shape as input), indices, vq_loss, perplexity
        z_q = z.clone()  # quantized output
        indices = torch.randint(0, 512, (batch_size, seq_len))
        vq_loss = torch.tensor(0.1)
        perplexity = torch.tensor(100.0)
        
        return z_q, indices, vq_loss, perplexity

# Test with 2D input (batch, channels, height, width)
z_2d = torch.randn(2, 4, 3, 5)  # batch=2, channels=4, height=3, width=5
vq_model = MockVQModel()

ARGS = {
    "z": z_2d,
    "vq_model": vq_model,
    "transpose_channel_length_axes": False,
    "svq_temp": None
}
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 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
time_to_timefreq Ran
ran on a synthesized input
ml4its/timevqvae/src/timevqvae/utils/signal.py
code served (permissive licence) · get_code("cf92fded0890d4f3")
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 4, dtype family float, shape (2, 8, 33, 33) 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
from einops import rearrange, repeat

torch.manual_seed(0)

# Construct arguments for time_to_timefreq
# x: (B, C, L) - time domain signal
# n_fft: int - FFT size
# C: int - number of channels
# norm: bool - normalization flag

B = 2  # batch size
C = 4  # number of channels
L = 512  # signal length
n_fft = 64  # FFT window size

# Create a small time-domain signal tensor of shape (B, C, L)
x = torch.randn(B, C, L)

ARGS = {
    'x': x,
    'n_fft': n_fft,
    'C': C,
    'norm': 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 Tensor, ndim 4, dtype family float, shape (2, 8, 33, 33) — shape (2, 8, 33, 33) 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
timefreq_to_time Ran
ran on a synthesized input
ml4its/timevqvae/src/timevqvae/utils/signal.py
code served (permissive licence) · get_code("31a6463bd53acfb2")
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, 16) 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
from einops import rearrange

torch.manual_seed(0)

# The function expects:
# x: a tensor that will be rearranged as 'b (c z) n t -> (b c) n t z'
#    This means x has shape (b, c*z, n, t)
#    After rearrange: (b*c, n, t, z)
#    After view_as_complex: (b*c, n, t) complex tensor
#    After istft: (b*c, l) time-domain signal
#    After final rearrange: (b, c, l)

# Let's work backwards:
# - batch size b=2
# - channels C=3
# - n_fft=16, so n must be n_fft//2 + 1 = 9
# - t can be arbitrary, say 5
# - z must be 2 (for complex representation: real and imag)
# - l (output length) will be determined by istft

b = 2
C = 3
n = 9  # n_fft // 2 + 1
t = 5
z = 2

# x shape before rearrange: (b, c*z, n, t) = (2, 6, 9, 5)
x = torch.randn(b, C * z, n, t)

n_fft = 16
norm = True

ARGS = {
    "x": x,
    "n_fft": n_fft,
    "C": C,
    "norm": norm,
}
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 Tensor, ndim 3, dtype family float, shape (2, 3, 16) — shape (2, 3, 16) 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
zero_pad_high_freq Ran
ran on a synthesized input
ml4its/timevqvae/src/timevqvae/utils/signal.py
code served (permissive licence) · get_code("be9de64905dae286")
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 4, dtype family float, shape (2, 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 torch
from einops import rearrange, repeat

torch.manual_seed(0)

# Create a small 4D tensor (B, C, H, W)
# Using distinct sizes: batch=2, channels=4, height=3, width=5
xf = torch.randn(2, 4, 3, 5)

ARGS = {
    "xf": xf,
    "copy": False,
}
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 Tensor, ndim 4, dtype family float, shape (2, 4, 3, 5) — shape (2, 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

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, 4 are also present in other papers whose code we hold (4 with at least one paper dated earlier, 3 the same month, 3 later) and 12 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
exists 486 other papers
150 dated earlier, 15 the same month, 315 later, 4 same year, order not determinable, 2 undated · 78 only as an assembly brick
1603.08155 (2016-03)
Perceptual Losses for Real-Time Style Transfer and Super-Resolution
1611.07004 (2016-11)
Image-to-Image Translation with Conditional Adversarial Networks
1804.04637 (2018-04)
EMBER: An Open Dataset for Training Static PE Malware Machine Learning Models
and 483 more not named here
default 163 other papers
75 dated earlier, 4 the same month, 82 later, 1 same year, order not determinable, 1 undated · 27 only as an assembly brick
1603.08155 (2016-03)
Perceptual Losses for Real-Time Style Transfer and Super-Resolution
1810.00825 (2018-10)
Set Transformer: A Framework for Attention-based Permutation-Invariant Neural Networks
1907.01470 (2019-07)
Augmenting Self-attention with Persistent Memory
and 160 more not named here
l2norm 24 other papers
1 dated earlier, 1 the same month, 22 later · 6 only as an assembly brick
2212.09058 (2022-12)
BEATs: Audio Pre-Training with Acoustic Tokenizers
2303.04341 (2023-03)
Neural Vector Fields: Implicit Representation by Explicit Learning
2305.02541 (2023-05)
Catch Missing Details: Image Reconstruction with Frequency Augmented Variational Autoencoder
and 21 more not named here
conv1d_same_padding 1 other paper
1 dated earlier
2211.05244 (2022-11)
Deep Learning for Time Series Anomaly Detection: A Survey

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

Time series generation (TSG) studies have mainly focused on the use of Generative Adversarial Networks (GANs) combined with recurrent neural network (RNN) variants. However, the fundamental limitations and challenges of training GANs still remain. In addition, the RNN-family typically has difficulties with temporal consistency between distant timesteps. Motivated by the successes in the image generation (IMG) domain, we propose TimeVQVAE, the first work, to our knowledge, that uses vector quantization (VQ) techniques to address the TSG problem. Moreover, the priors of the discrete latent spaces are learned with bidirectional transformer models that can better capture global temporal consistency. We also propose VQ modeling in a time-frequency domain, separated into low-frequency (LF) and high-frequency (HF). This allows us to retain important characteristics of the time series and, in turn, generate new synthetic signals that are of better quality, with sharper changes in modularity, than its competing TSG methods. Our experimental evaluation is conducted on all datasets from the UCR archive, using well-established metrics in the IMG literature, such as Fréchet inception distance and inception scores. Our implementation on GitHub: \url{https://github.com/ML4ITS/TimeVQVAE}.

For agents

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

get_harvested_code_for_paper("2303.04743")
get_code_for_paper("2303.04743")
have("2303.04743")

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

curl https://syntology.ai/api/ran/2303.04743.json

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

[![Syntology run record](https://syntology.ai/api/ran/2303.04743.svg)](https://syntology.ai/paper/2303.04743)

Connect an agent — have() is free.