SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2109.15047 · NeurIPS · 2021

Deep Contextual Video Compression

Bin Li, Jiahao Li, Yan Lu

arXiv · PDF · Open in the Atlas

Code that ran

13 of 21 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; 1 honoured the contract we drafted, 0 violated it and 6 have no contract verdict; 6 failures of our instrument; 8 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/2109.15047.json · badge: /api/ran/2109.15047.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
DeepMC-DCVC/DCVC marked official in Papers with Code
Papers with Code archive
13 ran · 8 no recorded run
FunctionStatusWhere it lives
Bitparm Ran
ran on a synthesized input
DeepMC-DCVC/DCVC/src/models/DCVC_net.py
code served (permissive licence) · get_code("c8ca4619802af671")
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 4, dtype family float, shape (2, 4, 3, 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.

import torch
import torch.nn as nn
import torch.nn.functional as F

class Bitparm(nn.Module):

    def __init__(self, channel, final=False):
        super(Bitparm, self).__init__()
        self.final = final
        self.h = nn.Parameter(torch.nn.init.normal_(torch.empty(channel).view(1, -1, 1, 1), 0, 0.01))
        self.b = nn.Parameter(torch.nn.init.normal_(torch.empty(channel).view(1, -1, 1, 1), 0, 0.01))
        if not final:
            self.a = nn.Parameter(torch.nn.init.normal_(torch.empty(channel).view(1, -1, 1, 1), 0, 0.01))
        else:
            self.a = None

    def forward(self, x):
        if self.final:
            return torch.sigmoid(x * F.softplus(self.h) + self.b)
        else:
            x = x * F.softplus(self.h) + self.b
            return x + torch.tanh(x) * torch.tanh(self.a)


INIT_ARGS = {
    'channel': 4,
    'final': False
}

ARGS = {
    'x': torch.randn(2, 4, 3, 3)
}

if __name__ == '__main__':
    model = Bitparm(**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 4, dtype family float, shape (2, 4, 3, 3) — shape (2, 4, 3, 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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
GDN Ran
ran on a synthesized input
DeepMC-DCVC/DCVC/src/models/DCVC_net.py
code served (permissive licence) · get_code("8157b941c3a4b0ee")
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 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
import torch.nn as nn
from torch.autograd import Function

class LowerBound(Function):

    @staticmethod
    def forward(ctx, inputs, bound):
        b = torch.ones_like(inputs) * bound
        ctx.save_for_backward(inputs, b)
        return torch.max(inputs, b)

    @staticmethod
    def backward(ctx, grad_output):
        inputs, b = ctx.saved_tensors
        pass_through_1 = inputs >= b
        pass_through_2 = grad_output < 0
        pass_through = pass_through_1 | pass_through_2
        return (pass_through.type(grad_output.dtype) * grad_output, None)

class GDN(nn.Module):

    def __init__(self, ch, inverse=False, beta_min=1e-06, gamma_init=0.1, reparam_offset=2 ** (-18)):
        super(GDN, self).__init__()
        self.inverse = inverse
        self.beta_min = beta_min
        self.gamma_init = gamma_init
        self.reparam_offset = reparam_offset
        self.build(ch)

    def build(self, ch):
        self.pedestal = self.reparam_offset ** 2
        self.beta_bound = (self.beta_min + self.reparam_offset ** 2) ** 0.5
        self.gamma_bound = self.reparam_offset
        beta = torch.sqrt(torch.ones(ch) + self.pedestal)
        self.beta = nn.Parameter(beta)
        eye = torch.eye(ch)
        g = self.gamma_init * eye
        g = g + self.pedestal
        gamma = torch.sqrt(g)
        self.gamma = nn.Parameter(gamma)
        self.pedestal = self.pedestal

    def forward(self, inputs):
        unfold = False
        if inputs.dim() == 5:
            unfold = True
            bs, ch, d, w, h = inputs.size()
            inputs = inputs.view(bs, ch, d * w, h)
        _, ch, _, _ = inputs.size()
        beta = LowerBound.apply(self.beta, self.beta_bound)
        beta = beta ** 2 - self.pedestal
        gamma = LowerBound.apply(self.gamma, self.gamma_bound)
        gamma = gamma ** 2 - self.pedestal
        gamma = gamma.view(ch, ch, 1, 1)
        norm_ = nn.functional.conv2d(inputs ** 2, gamma, beta)
        norm_ = torch.sqrt(norm_)
        if self.inverse:
            outputs = inputs * norm_
        else:
            outputs = inputs / norm_
        if unfold:
            outputs = outputs.view(bs, ch, d, w, h)
        return outputs

INIT_ARGS = {
    'ch': 4,
    'inverse': False,
    'beta_min': 1e-06,
    'gamma_init': 0.1,
    'reparam_offset': 2 ** (-18)
}

ARGS = {
    'inputs': torch.randn(2, 4, 3, 5)
}

if __name__ == '__main__':
    gdn = GDN(**INIT_ARGS)
    output = gdn(**ARGS)
    print(f"Input shape: {ARGS['inputs'].shape}")
    print(f"Output shape: {output.shape}")
    print("GDN forward pass successful")
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 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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
MEBasic Ran
ran on a synthesized input
DeepMC-DCVC/DCVC/src/models/DCVC_net.py
code served (permissive licence) · get_code("d5f22419c6a9a736")
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 4, dtype family float, shape (2, 2, 4, 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.

import torch
import torch.nn as nn

class MEBasic(nn.Module):

    def __init__(self):
        super(MEBasic, self).__init__()
        self.conv1 = nn.Conv2d(8, 32, 7, 1, padding=3)
        self.relu1 = nn.ReLU()
        self.conv2 = nn.Conv2d(32, 64, 7, 1, padding=3)
        self.relu2 = nn.ReLU()
        self.conv3 = nn.Conv2d(64, 32, 7, 1, padding=3)
        self.relu3 = nn.ReLU()
        self.conv4 = nn.Conv2d(32, 16, 7, 1, padding=3)
        self.relu4 = nn.ReLU()
        self.conv5 = nn.Conv2d(16, 2, 7, 1, padding=3)

    def forward(self, x):
        x = self.relu1(self.conv1(x))
        x = self.relu2(self.conv2(x))
        x = self.relu3(self.conv3(x))
        x = self.relu4(self.conv4(x))
        x = self.conv5(x)
        return x


INIT_ARGS = {}

ARGS = {
    'x': torch.randn(2, 8, 4, 3)
}

if __name__ == '__main__':
    model = MEBasic(**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 4, dtype family float, shape (2, 2, 4, 3) — shape (2, 2, 4, 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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
MaskedConv2d Ran
ran on a synthesized input
DeepMC-DCVC/DCVC/src/models/DCVC_net.py
code served (permissive licence) · get_code("2e84e4d2265d5ef3")
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 4, dtype family float, shape (2, 4, 6, 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 torch
import torch.nn as nn

class MaskedConv2d(nn.Conv2d):
    """Masked 2D convolution implementation, mask future "unseen" pixels.
    Useful for building auto-regressive network components.

    Introduced in `"Conditional Image Generation with PixelCNN Decoders"
    <https://arxiv.org/abs/1606.05328>`_.

    Inherits the same arguments as a `nn.Conv2d`. Use `mask_type='A'` for the
    first layer (which also masks the "current pixel"), `mask_type='B'` for the
    following layers.
    """

    def __init__(self, *args, mask_type='A', **kwargs):
        super().__init__(*args, **kwargs)
        if mask_type not in ('A', 'B'):
            raise ValueError(f'Invalid "mask_type" value "{mask_type}"')
        self.register_buffer('mask', torch.ones_like(self.weight.data))
        _, _, h, w = self.mask.size()
        self.mask[:, :, h // 2, w // 2 + (mask_type == 'B'):] = 0
        self.mask[:, :, h // 2 + 1:] = 0

    def forward(self, x):
        self.weight.data *= self.mask
        return super().forward(x)


INIT_ARGS = {
    'in_channels': 3,
    'out_channels': 4,
    'kernel_size': 3,
    'mask_type': 'A',
}

ARGS = {
    'x': torch.randn(2, 3, 8, 8),
}

if __name__ == '__main__':
    model = MaskedConv2d(**INIT_ARGS)
    output = model(**ARGS)
    print(f"Input shape: {ARGS['x'].shape}")
    print(f"Output shape: {output.shape}")
    print(f"Mask shape: {model.mask.shape}")
    print("MaskedConv2d executed successfully")
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 4, dtype family float, shape (2, 4, 6, 6) — shape (2, 4, 6, 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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
ResBlock Ran
ran on a synthesized input
DeepMC-DCVC/DCVC/src/models/DCVC_net.py
code served (permissive licence) · get_code("3633e85d48e9359c")
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 4, dtype family float, shape (2, 4, 8, 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 torch
import torch.nn as nn

class ResBlock(nn.Module):

    def __init__(self, inputchannel, outputchannel, kernel_size, stride=1):
        super(ResBlock, self).__init__()
        self.relu1 = nn.ReLU()
        self.conv1 = nn.Conv2d(inputchannel, outputchannel, kernel_size, stride, padding=kernel_size // 2)
        torch.nn.init.xavier_uniform_(self.conv1.weight.data)
        torch.nn.init.constant_(self.conv1.bias.data, 0.0)
        self.relu2 = nn.ReLU()
        self.conv2 = nn.Conv2d(outputchannel, outputchannel, kernel_size, stride, padding=kernel_size // 2)
        torch.nn.init.xavier_uniform_(self.conv2.weight.data)
        torch.nn.init.constant_(self.conv2.bias.data, 0.0)
        if inputchannel != outputchannel:
            self.adapt_conv = nn.Conv2d(inputchannel, outputchannel, 1)
            torch.nn.init.xavier_uniform_(self.adapt_conv.weight.data)
            torch.nn.init.constant_(self.adapt_conv.bias.data, 0.0)
        else:
            self.adapt_conv = None

    def forward(self, x):
        x_1 = self.relu1(x)
        firstlayer = self.conv1(x_1)
        firstlayer = self.relu2(firstlayer)
        seclayer = self.conv2(firstlayer)
        if self.adapt_conv is None:
            return x + seclayer
        else:
            return self.adapt_conv(x) + seclayer


INIT_ARGS = {
    'inputchannel': 3,
    'outputchannel': 4,
    'kernel_size': 3,
    'stride': 1
}

ARGS = {
    'x': torch.randn(2, 3, 8, 8)
}

if __name__ == '__main__':
    model = ResBlock(**INIT_ARGS)
    output = model(**ARGS)
    print(f"Output shape: {output.shape}")
    print(f"Expected shape: torch.Size([2, 4, 8, 8])")
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 4, dtype family float, shape (2, 4, 8, 8) — shape (2, 4, 8, 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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
ResBlock_LeakyReLU_0_Point_1 Ran
ran on a synthesized input
DeepMC-DCVC/DCVC/src/models/DCVC_net.py
code served (permissive licence) · get_code("63ea86c225934c81")
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 4, 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

INIT_ARGS = {
    "d_model": 4
}

# Create a small input tensor: batch=2, channels=4, height=3, width=5
input_tensor = torch.randn(2, 4, 3, 5)

ARGS = {
    "x": input_tensor
}
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 4, 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)
bilinearupsacling Ran
ran; our contract draft was wrong, not the code
DeepMC-DCVC/DCVC/src/models/DCVC_net.py
code served (permissive licence) · get_code("3975b6c6c5b10b19")
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 Tensor, ndim 4, dtype family float, shape (2, 4, 6, 10) 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)

# Create a small input tensor with shape (batch, channels, height, width)
# Using distinct dimensions: batch=2, channels=4, height=3, width=5
inputfeature = torch.randn(2, 4, 3, 5)

ARGS = {
    "inputfeature": inputfeature
}
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 Tensor, ndim 4, dtype family float, shape (2, 4, 6, 10) — shape (2, 4, 6, 10) 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
decode_p Ran
ran; our contract draft was wrong, not the code
DeepMC-DCVC/DCVC/src/models/DCVC_net.py
code served (permissive licence) · get_code("63d3e8e4373913f9")
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 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 struct
from pathlib import Path
import tempfile
import os

# Create a temporary binary file with valid structure
temp_dir = tempfile.mkdtemp()
temp_file = os.path.join(temp_dir, "test.bin")

# Write a valid binary file matching the expected format
with open(temp_file, 'wb') as f:
    # Write 6 unsigned ints as header
    # height=10, width=20, mv_y_len=5, mv_z_len=7, y_len=3, z_len=4
    header = struct.pack('>6I', 10, 20, 5, 7, 3, 4)
    f.write(header)
    
    # Write the string data
    f.write(struct.pack('>5s', b'hello'))  # mv_y_string (5 bytes)
    f.write(struct.pack('>7s', b'world!!'))  # mv_z_string (7 bytes)
    f.write(struct.pack('>3s', b'foo'))  # y_string (3 bytes)
    f.write(struct.pack('>4s', b'barx'))  # z_string (4 bytes)

ARGS = {
    'inputpath': temp_file
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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 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
flow_warp Ran
ran; our contract draft was wrong, not the code
DeepMC-DCVC/DCVC/src/models/DCVC_net.py
code served (permissive licence) · get_code("e7a1c891e1de7228")
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, 3, 4, 5) as a later run recorded it. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is 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)

# Create a small image tensor: (batch=2, channels=3, height=4, width=5)
im = torch.randn(2, 3, 4, 5)

# Create a flow tensor: (batch=2, channels=2, height=4, width=5)
# Channel 0 is horizontal flow, channel 1 is vertical flow
flow = torch.randn(2, 2, 4, 5) * 0.1

ARGS = {
    'im': im,
    'flow': flow
}
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 4, dtype family float, shape (2, 3, 4, 5) — shape (2, 3, 4, 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
read_bytes Ran
ran; our contract draft was wrong, not the code
DeepMC-DCVC/DCVC/src/models/DCVC_net.py
code served (permissive licence) · get_code("bd13dca3e1023087")
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 bytes. 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 struct
import io

# Create a BytesIO object to act as a file descriptor
# We'll write some test data that can be read back
test_data = struct.pack('>5s', b'hello')
fd = io.BytesIO(test_data)

ARGS = {
    'fd': fd,
    'n': 5,
    'fmt': '>{:d}s'
}
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 bytes — a bytes output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a bytes 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
read_uints Ran
ran, honoured the contract we drafted
DeepMC-DCVC/DCVC/src/models/DCVC_net.py
code served (permissive licence) · get_code("ae94682e8b678a9e")
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 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 struct
import io

# Create a BytesIO object to simulate a file descriptor
# We'll write some unsigned integers to it
data = struct.pack('>4I', 10, 20, 30, 40)  # 4 unsigned ints in big-endian format
fd = io.BytesIO(data)

ARGS = {
    'fd': fd,
    'n': 4,
    'fmt': '>{:d}I'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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 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
subpel_conv3x3 Ran
ran; our contract draft was wrong, not the code
DeepMC-DCVC/DCVC/src/models/DCVC_net.py
code served (permissive licence) · get_code("4835059d9e92906b")
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 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

torch.manual_seed(0)

ARGS = {
    "in_ch": 3,
    "out_ch": 64,
    "r": 2
}
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 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
torch_warp Ran
ran; our contract draft was wrong, not the code
DeepMC-DCVC/DCVC/src/models/DCVC_net.py
code served (permissive licence) · get_code("675fd1b09a214c40")
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, 3, 4, 5) as a later run recorded it. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is 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)

# Create a small batch of input images: (batch=2, channels=3, height=4, width=5)
tensorInput = torch.randn(2, 3, 4, 5)

# Create a small batch of flow fields: (batch=2, channels=2, height=4, width=5)
# Channel 0 is horizontal flow, channel 1 is vertical flow
tensorFlow = torch.randn(2, 2, 4, 5)

ARGS = {
    'tensorInput': tensorInput,
    'tensorFlow': tensorFlow
}
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 4, dtype family float, shape (2, 3, 4, 5) — shape (2, 3, 4, 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
BitEstimator Not yet run
unverified (harvested, no recorded run)
DeepMC-DCVC/DCVC/src/models/DCVC_net.py
code served (permissive licence) · get_code("2dec82dbb841c515")
How it ran

no invocation recorded.

DCVC_net Not yet run
unverified (harvested, no recorded run)
DeepMC-DCVC/DCVC/src/models/DCVC_net.py
code served (permissive licence) · get_code("2c21264b0af480b8")
How it ran

no invocation recorded.

EntropyCoder Not yet run
unverified (harvested, no recorded run)
DeepMC-DCVC/DCVC/src/models/DCVC_net.py
code served (permissive licence) · get_code("9de28791581c6ec6")
How it ran

no invocation recorded.

GaussianEncoder Not yet run
unverified (harvested, no recorded run)
DeepMC-DCVC/DCVC/src/models/DCVC_net.py
code served (permissive licence) · get_code("88c850d81bb8e114")
How it ran

no invocation recorded.

ME_Spynet Not yet run
unverified (harvested, no recorded run)
DeepMC-DCVC/DCVC/src/models/DCVC_net.py
code served (permissive licence) · get_code("739517aad1af8b63")
How it ran

no invocation recorded.

encode_p Not yet run
unverified (harvested, no recorded run)
DeepMC-DCVC/DCVC/src/models/DCVC_net.py
code served (permissive licence) · get_code("3c6ecd5e0c51b5d8")
How it ran

no invocation recorded.

write_bytes Not yet run
unverified (harvested, no recorded run)
DeepMC-DCVC/DCVC/src/models/DCVC_net.py
code served (permissive licence) · get_code("02d0069a339c80a7")
How it ran

no invocation recorded.

write_uints Not yet run
unverified (harvested, no recorded run)
DeepMC-DCVC/DCVC/src/models/DCVC_net.py
code served (permissive licence) · get_code("f2a8983488d1cd94")
How it ran

no invocation recorded.

Also present in other papers' code

Of 21 samples lifted from this paper's linked repositories, 21 carry a function identity and 0 do not. Of the 21 distinct functions with an identity, 6 are also present in other papers whose code we hold (4 with at least one paper dated earlier, 0 the same month, 5 later) and 15 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
subpel_conv3x3
brick of this paper's assembly: helper
3 other papers
1 dated earlier, 1 later, 1 same year, order not determinable
Deng_Deep_Homography_for_Efficient_Stereo_Image_Compression_CVPR_2021_paper (2021 (year only))
Deep Homography for Efficient Stereo Image Compression
2108.03690 (2021-08)
Enhanced Invertible Encoding for Learned Image Compression
2402.17414 (2024-02)
Neural Video Compression with Feature Modulation
read_bytes
brick of this paper's assembly: helper
2 other papers
1 dated earlier, 1 later · 1 only as an assembly brick
2106.10800 (2021-06)
Lossy Compression for Lossless Prediction
2302.05071 (2023-02)
EVC: Towards Real-Time Neural Image Compression with Mask Decay
read_uints
brick of this paper's assembly: helper
2 other papers
1 dated earlier, 1 later · 1 only as an assembly brick
2106.10800 (2021-06)
Lossy Compression for Lossless Prediction
2302.05071 (2023-02)
EVC: Towards Real-Time Neural Image Compression with Mask Decay
Bitparm
brick of this paper's assembly: helper
1 other paper
1 dated earlier
2105.09600 (2021-05)
FVC: A New Framework towards Deep Video Compression in Feature Space
write_bytes
brick of this paper's assembly: helper
1 other paper
1 later · 1 only as an assembly brick
2302.05071 (2023-02)
EVC: Towards Real-Time Neural Image Compression with Mask Decay
write_uints
brick of this paper's assembly: helper
1 other paper
1 later · 1 only as an assembly brick
2302.05071 (2023-02)
EVC: Towards Real-Time Neural Image Compression with Mask Decay

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

Most of the existing neural video compression methods adopt the predictive coding framework, which first generates the predicted frame and then encodes its residue with the current frame. However, as for compression ratio, predictive coding is only a sub-optimal solution as it uses simple subtraction operation to remove the redundancy across frames. In this paper, we propose a deep contextual video compression framework to enable a paradigm shift from predictive coding to conditional coding. In particular, we try to answer the following questions: how to define, use, and learn condition under a deep video compression framework. To tap the potential of conditional coding, we propose using feature domain context as condition. This enables us to leverage the high dimension context to carry rich information to both the encoder and the decoder, which helps reconstruct the high-frequency contents for higher video quality. Our framework is also extensible, in which the condition can be flexibly designed. Experiments show that our method can significantly outperform the previous state-of-the-art (SOTA) deep video compression methods. When compared with x265 using veryslow preset, we can achieve 26.0% bitrate saving for 1080P standard test videos.

For agents

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

get_harvested_code_for_paper("2109.15047")
get_code_for_paper("2109.15047")
have("2109.15047")

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

curl https://syntology.ai/api/ran/2109.15047.json

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

[![Syntology run record](https://syntology.ai/api/ran/2109.15047.svg)](https://syntology.ai/paper/2109.15047)

Connect an agent — have() is free.