SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1906.04397 · 2019

Probabilistic Forecasting with Temporal Convolutional Neural Network

arXiv · PDF · Open in the Atlas

Code that ran

7 of 8 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 7 have no contract verdict; 0 failures of our instrument; 1 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/1906.04397.json · badge: /api/ran/1906.04397.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
flaviagiammarino/deep-tcn-tensorflow community (classifier label)
classifier basis not recorded on the link
7 ran · 0 no recorded run
vpozdnyakov/deeptcn community (classifier label)
classifier basis not recorded on the link
0 ran · 1 no recorded run
FunctionStatusWhere it lives
decoder Ran
ran on a synthesized input
flaviagiammarino/deep-tcn-tensorflow/deep_tcn_tensorflow/modules.py
code served (permissive licence) · get_code("974558a5f8f90672")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned EagerTensor, ndim 3, dtype family float, shape (2, 4, 6) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import tensorflow as tf
import numpy as np

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

# Define dimensions
n_samples = 2
n_forecast = 4
n_features = 3
filters = 5
units = 6

# Create decoder_input tensor: (n_samples, n_forecast, n_features)
decoder_input = tf.constant(
    np.random.randn(n_samples, n_forecast, n_features).astype(np.float32)
)

# Create encoder_output tensor: (n_samples, n_forecast, filters)
encoder_output = tf.constant(
    np.random.randn(n_samples, n_forecast, filters).astype(np.float32)
)

# Define units as integer
units_value = units

ARGS = {
    "decoder_input": decoder_input,
    "encoder_output": encoder_output,
    "units": units_value,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EagerTensor, ndim 3, dtype family float, shape (2, 4, 6) — shape (2, 4, 6) 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
encoder Ran
ran on a synthesized input
flaviagiammarino/deep-tcn-tensorflow/deep_tcn_tensorflow/modules.py
code served (permissive licence) · get_code("cccb14c9e1c596b0")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned EagerTensor, ndim 3, dtype family float, shape (2, 4, 8) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import tensorflow as tf
import numpy as np

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

# Create a small encoder input tensor
# Shape: (n_samples=2, n_lookback=4, n_features+n_targets=3)
encoder_input = tf.constant(np.random.randn(2, 4, 3).astype(np.float32))

# Define parameters
filters = 8
kernel_size = 3
dilation_rate = 1

ARGS = {
    "encoder_input": encoder_input,
    "filters": filters,
    "kernel_size": kernel_size,
    "dilation_rate": dilation_rate,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EagerTensor, ndim 3, dtype family float, shape (2, 4, 8) — shape (2, 4, 8) 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_training_sequences Ran
ran on a synthesized input
flaviagiammarino/deep-tcn-tensorflow/deep_tcn_tensorflow/utils.py
code served (permissive licence) · get_code("108fb147d9608231")
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

np.random.seed(0)

# Create a small time series with shape (n_samples, n_targets)
n_samples = 20
n_targets = 3
n_lookback = 5
n_forecast = 4

# Generate synthetic time series data
y = np.random.randn(n_samples, n_targets).astype(np.float32)

ARGS = {
    "y": y,
    "n_samples": n_samples,
    "n_targets": n_targets,
    "n_lookback": n_lookback,
    "n_forecast": n_forecast,
}
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 — 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_training_sequences_with_covariates Ran
ran on a synthesized input
flaviagiammarino/deep-tcn-tensorflow/deep_tcn_tensorflow/utils.py
code served (permissive licence) · get_code("f227531ddca0f834")
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

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

# Define dimensions
n_samples = 20
n_targets = 2
n_features = 3
n_lookback = 5
n_forecast = 4

# Create small sample time series data
y = np.random.randn(n_samples, n_targets).astype(np.float32)
x = np.random.randn(n_samples, n_features).astype(np.float32)

ARGS = {
    'y': y,
    'x': x,
    'n_samples': n_samples,
    'n_targets': n_targets,
    'n_features': n_features,
    'n_lookback': n_lookback,
    'n_forecast': n_forecast,
}
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 — 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
nonparametric_loss Ran
ran on a synthesized input
flaviagiammarino/deep-tcn-tensorflow/deep_tcn_tensorflow/losses.py
code served (permissive licence) · get_code("92c170954aae8d5e")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned EagerTensor, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import tensorflow as tf

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

# Define small tensor dimensions
n_samples = 2
n_forecast = 3
n_targets = 4
n_quantiles = 5

# Create y_true: shape (n_samples, n_forecast, n_targets)
y_true = tf.constant([
    [[1.0, 2.0, 3.0, 4.0],
     [1.5, 2.5, 3.5, 4.5],
     [2.0, 3.0, 4.0, 5.0]],
    [[1.1, 2.1, 3.1, 4.1],
     [1.6, 2.6, 3.6, 4.6],
     [2.1, 3.1, 4.1, 5.1]]
], dtype=tf.float32)

# Create y_pred: shape (n_samples, n_forecast, n_targets, n_quantiles)
y_pred = tf.constant([[
    [[0.9, 1.0, 1.1, 1.2, 1.3],
     [1.9, 2.0, 2.1, 2.2, 2.3],
     [2.9, 3.0, 3.1, 3.2, 3.3],
     [3.9, 4.0, 4.1, 4.2, 4.3]],
    [[1.4, 1.5, 1.6, 1.7, 1.8],
     [2.4, 2.5, 2.6, 2.7, 2.8],
     [3.4, 3.5, 3.6, 3.7, 3.8],
     [4.4, 4.5, 4.6, 4.7, 4.8]],
    [[1.9, 2.0, 2.1, 2.2, 2.3],
     [2.9, 3.0, 3.1, 3.2, 3.3],
     [3.9, 4.0, 4.1, 4.2, 4.3],
     [4.9, 5.0, 5.1, 5.2, 5.3]]
], [
    [[0.95, 1.05, 1.15, 1.25, 1.35],
     [1.95, 2.05, 2.15, 2.25, 2.35],
     [2.95, 3.05, 3.15, 3.25, 3.35],
     [3.95, 4.05, 4.15, 4.25, 4.35]],
    [[1.45, 1.55, 1.65, 1.75, 1.85],
     [2.45, 2.55, 2.65, 2.75, 2.85],
     [3.45, 3.55, 3.65, 3.75, 3.85],
     [4.45, 4.55, 4.65, 4.75, 4.85]],
    [[1.95, 2.05, 2.15, 2.25, 2.35],
     [2.95, 3.05, 3.15, 3.25, 3.35],
     [3.95, 4.05, 4.15, 4.25, 4.35],
     [4.95, 5.05, 5.15, 5.25, 5.35]]
]], dtype=tf.float32)

# Create q: shape (n_quantiles,)
q = tf.constant([0.1, 0.25, 0.5, 0.75, 0.9], dtype=tf.float32)

ARGS = {
    "y_true": y_true,
    "y_pred": y_pred,
    "q": q
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EagerTensor, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
parametric_loss Ran
ran on a synthesized input
flaviagiammarino/deep-tcn-tensorflow/deep_tcn_tensorflow/losses.py
code served (permissive licence) · get_code("59397137520e3ba5")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned EagerTensor, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import tensorflow as tf
import numpy as np

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

# Create small sample data
n_samples = 2
n_forecast = 3
n_targets = 4

# y_true: shape (n_samples, n_forecast, n_targets)
y_true = tf.constant(
    np.random.randn(n_samples, n_forecast, n_targets).astype(np.float32)
)

# params: shape (n_samples, n_forecast, n_targets, 2)
# Last dimension contains [mu, sigma]
# sigma should be positive for valid loss computation
params_data = np.random.randn(n_samples, n_forecast, n_targets, 2).astype(np.float32)
# Ensure sigma (second parameter) is positive
params_data[:, :, :, 1] = np.abs(params_data[:, :, :, 1]) + 0.1
params = tf.constant(params_data)

ARGS = {
    "y_true": y_true,
    "params": params,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EagerTensor, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
soft_relu Ran
ran on a synthesized input
flaviagiammarino/deep-tcn-tensorflow/deep_tcn_tensorflow/model.py
code served (permissive licence) · get_code("97d25536c8f951d0")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned EagerTensor, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import tensorflow as tf
import numpy as np

# Create a small test input tensor
# Using a small shape (batch_size=2, features=4) as per requirements
x = tf.constant([[1.0, 2.0, 3.0, 4.0],
                 [-1.0, -2.0, -3.0, -4.0]], dtype=tf.float32)

ARGS = {
    'x': x
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type EagerTensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
quantile_loss Not yet run
unverified (harvested, no recorded run)
vpozdnyakov/deeptcn/deeptcn/quantile_deeptcn.py
code served (permissive licence) · get_code("6462bf5b424fbb52")
How it ran

no invocation recorded.

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 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 8 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 2026-11-22.

Repositories linked to this paper

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

Abstract

We present a probabilistic forecasting framework based on convolutional neural network for multiple related time series forecasting. The framework can be applied to estimate probability density under both parametric and non-parametric settings. More specifically, stacked residual blocks based on dilated causal convolutional nets are constructed to capture the temporal dependencies of the series. Combined with representation learning, our approach is able to learn complex patterns such as seasonality, holiday effects within and across series, and to leverage those patterns for more accurate forecasts, especially when historical data is sparse or unavailable. Extensive empirical studies are performed on several real-world datasets, including datasets from JD.com, China's largest online retailer. The results show that our framework outperforms other state-of-the-art methods in both accuracy and efficiency.

For agents

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

get_harvested_code_for_paper("1906.04397")
get_code_for_paper("1906.04397")
have("1906.04397")

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

curl https://syntology.ai/api/ran/1906.04397.json

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

[![Syntology run record](https://syntology.ai/api/ran/1906.04397.svg)](https://syntology.ai/paper/1906.04397)

Connect an agent — have() is free.