SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1608.08710 · 2016

Pruning Filters for Efficient ConvNets

Hao Li, Asim Kadav, Igor Durdanovic, Hanan Samet, Hans Graf

arXiv · PDF · Open in the Atlas

Code that ran

23 of 36 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; 1 honoured the contract we drafted, 1 violated it and 6 have no contract verdict; 15 failures of our instrument; 13 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/1608.08710.json · badge: /api/ran/1608.08710.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
EstherBear/implementation-of-pruning-filters community
Papers with Code archive
5 ran · 3 no recorded run
lehduong/ginp community (classifier label)
classifier basis not recorded on the link
4 ran · 4 no recorded run
siyuan0/pytorch_model_prune community
Papers with Code archive
4 ran · 2 no recorded run
AnishDelft/ModelCompression community
Papers with Code archive
4 ran · 0 no recorded run
marcoancona/TorchPruner community
Papers with Code archive
1 ran · 1 no recorded run
he-y/filter-pruning-geometric-median community
Papers with Code archive
1 ran · 0 no recorded run
guoxiaolu/model_compression community
Papers with Code archive
1 ran · 0 no recorded run
mvpzhangqiu/yolov5prune community
Papers with Code archive
1 ran · 0 no recorded run
midasklr/yolov5prune community
Papers with Code archive
1 ran · 0 no recorded run
prerakmody/CS4180-DL community
Papers with Code archive
1 ran · 0 no recorded run
mingsun-tse/regularization-pruning community
Papers with Code archive
0 ran · 3 no recorded run
FunctionStatusWhere it lives
Mask Ran
ran on a synthesized input · constructed an object; no result was computed
he-y/filter-pruning-geometric-median/VGG_cifar/pruning_cifar_vgg.py
pointer only (licence: NONE) · get_code("05b9fa9dad6ede01")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, he-y/filter-pruning-geometric-median 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.

Pruner Ran
ran on a synthesized input · constructed an object; no result was computed
marcoancona/TorchPruner/torchpruner/pruner/pruner.py
code served (permissive licence) · get_code("5d815bc1bdb41c80")
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, Pruner: 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
import numpy as np

torch.manual_seed(0)

# Build a minimal model that satisfies the Pruner usage
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear1 = nn.Linear(4, 5)
        self.bn1 = nn.BatchNorm1d(5)
        self.linear2 = nn.Linear(5, 3)
    
    def forward(self, x):
        x = self.linear1(x)
        x = self.bn1(x)
        x = self.linear2(x)
        return x

model = SimpleModel()

# Create optimizer
optimizer = torch.optim.SGD(model.parameters(), lr=0.01, momentum=0.9)

# Run a forward pass to initialize optimizer state
x = torch.randn(2, 4)
y = model(x)
loss = y.sum()
loss.backward()
optimizer.step()

device = 'cpu'
input_size = (2, 4)

ARGS = {
    'model': model,
    'input_size': input_size,
    'device': device,
    'optimizer': optimizer
}
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 Pruner — 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 Pruner 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
accuracy Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
lehduong/ginp/imagenet/weight-level/main_finetune.py
code served (permissive licence) · get_code("80752954f2a868b8")
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 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

torch.manual_seed(0)

# Create small tensors for testing
batch_size = 2
num_classes = 4

# output: logits/scores tensor of shape (batch_size, num_classes)
output = torch.randn(batch_size, num_classes)

# target: ground truth labels of shape (batch_size,)
target = torch.randint(0, num_classes, (batch_size,))

# topk: tuple of k values to compute accuracy for
topk = (1,)

ARGS = {
    "output": output,
    "target": target,
    "topk": topk,
}
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 list, ndim 2, 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
arg_nonzero_min Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
AnishDelft/ModelCompression/src/pruning/weightPruning/methods.py
pointer only (licence: NONE) · get_code("7b447ef653b6fd6a")
How it ran

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

channels_index Ran
ran; our contract draft was wrong, not the code
EstherBear/implementation-of-pruning-filters/prune.py
pointer only (licence: NONE) · get_code("152f217a1bf86018")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, EstherBear/implementation-of-pruning-filters 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.

conv3x3 Ran
ran; our contract draft was wrong, not the code
lehduong/ginp/cifar/filter_pruning/models/preresnet.py
pointer only (licence: MIT); the stored copy get_code serves, from lixiaotong97/DSU, is pointer only (licence: NONE) · get_code("fac5364e2f53c6db")
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 lixiaotong97/DSU, 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.

create_l1_norm_mask Ran
ran on a synthesized input
lehduong/ginp/imagenet/l1-norm-pruning/residualprune.py
code served (permissive licence) · get_code("7835366b1d8d0c63")
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 1, dtype family float, shape (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
import numpy as np

torch.manual_seed(0)

# Create a small Conv2d layer with out_channels=8
# The function uses: layer.out_channels, layer.weight.data
layer = nn.Conv2d(in_channels=3, out_channels=8, kernel_size=3, padding=1)

# num_keep should be less than out_channels
num_keep = 5

ARGS = {
    "layer": layer,
    "num_keep": num_keep,
}
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 Tensor, ndim 1, dtype family float, shape (8,) — shape (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
filter_prune Ran
ran; our contract draft was wrong, not the code
AnishDelft/ModelCompression/src/pruning/weightPruning/methods.py
pointer only (licence: NONE) · get_code("2af2848312c7f731")
How it ran

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

get_conv_zero_param Ran
ran on a synthesized input
lehduong/ginp/imagenet/weight-level/main_finetune.py
code served (permissive licence) · get_code("89d6cfa68bf25631")
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 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
import torch.nn as nn

torch.manual_seed(0)

# Build a small model with Conv2d layers, some with zero weights
class SimpleConvModel(nn.Module):
    def __init__(self):
        super().__init__()
        # First conv layer with some zero weights
        self.conv1 = nn.Conv2d(3, 4, kernel_size=3)
        # Second conv layer - we'll manually set some weights to zero
        self.conv2 = nn.Conv2d(4, 5, kernel_size=3)
        
        # Set some weights to zero in conv2
        with torch.no_grad():
            self.conv2.weight.data[0, :, :, :] = 0
            self.conv2.weight.data[1, 0, 0, 0] = 0
    
    def forward(self, x):
        x = self.conv1(x)
        x = self.conv2(x)
        return x

model = SimpleConvModel()

ARGS = {
    "model": model
}
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 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_filtered_idx Ran
ran, violated the contract we drafted
guoxiaolu/model_compression/inner_func.py
pointer only (licence: NONE) · get_code("41598b11a124fdd7")
How it ran

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

get_new_conv Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
EstherBear/implementation-of-pruning-filters/prune.py
pointer only (licence: NONE) · get_code("b2d26cb03ffebe9d")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, EstherBear/implementation-of-pruning-filters 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.

get_new_linear Ran
ran; our contract draft was wrong, not the code
EstherBear/implementation-of-pruning-filters/prune.py
pointer only (licence: NONE) · get_code("e4917253b4ffc6a8")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, EstherBear/implementation-of-pruning-filters 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.

get_new_norm Ran
ran; our contract draft was wrong, not the code
EstherBear/implementation-of-pruning-filters/prune.py
pointer only (licence: NONE) · get_code("6f17e5800058ad1a")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, EstherBear/implementation-of-pruning-filters 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.

layer_eval Ran
ran; our contract draft was wrong, not the code
siyuan0/pytorch_model_prune/prune.py
code served (permissive licence) · get_code("0a177233e25de4aa")
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 float, 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 torch
import torch.nn as nn

torch.manual_seed(0)

# The function expects a layer that:
# - has a view(-1) method (i.e., is a tensor)
# - supports iteration over its flattened elements
# - each element has an .item() method

layer = torch.randn(2, 4)

ARGS = {
    "layer": layer
}
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 float, 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
obtain_bn_mask Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
mvpzhangqiu/yolov5prune/prune_utils.py
pointer only (licence: NONE) · get_code("fbb0ce8d88ebc264")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, mvpzhangqiu/yolov5prune 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.

obtain_bn_mask Ran
ran; our contract draft was wrong, not the code
midasklr/yolov5prune/prune_utils.py
pointer only (licence: NONE) · get_code("a391dad107f2f625")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, midasklr/yolov5prune 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.

prune_one_filter Ran
ran; our contract draft was wrong, not the code
AnishDelft/ModelCompression/src/pruning/weightPruning/methods.py
pointer only (licence: NONE) · get_code("05588a795fc38ffd")
How it ran

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

prune_rate Ran
ran, honoured the contract we drafted
AnishDelft/ModelCompression/src/pruning/weightPruning/methods.py
pointer only (licence: NONE) · get_code("ef13bc5f154a40bc")
How it ran

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

pruned_conv2d Ran
ran on a synthesized input · constructed an object; no result was computed
siyuan0/pytorch_model_prune/prune.py
code served (permissive licence) · get_code("73a3fe3ae770b450")
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, pruned_conv2d: 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)

# Create a simple Conv2d layer to pass to pruned_conv2d
# The pruned_conv2d constructor expects a conv2d object with:
# - in_channels, out_channels, kernel_size, stride, padding, dilation, groups, bias, weight
conv2d = nn.Conv2d(
    in_channels=3,
    out_channels=8,
    kernel_size=3,
    stride=1,
    padding=1,
    dilation=1,
    groups=1,
    bias=False
)

# Initialize weights with some non-zero values
with torch.no_grad():
    conv2d.weight.fill_(0.1)

ARGS = {
    "conv2d": conv2d,
    "cut_off": 0.0
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type pruned_conv2d — 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 pruned_conv2d 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
quick_filter_prune_v2 Ran
ran; our contract draft was wrong, not the code
prerakmody/CS4180-DL/src/pruning/weightPruning/methods.py
pointer only (licence: NONE) · get_code("d02925cbefd3694f")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, prerakmody/CS4180-DL 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.

select_channels Ran
ran; our contract draft was wrong, not the code
EstherBear/implementation-of-pruning-filters/prune.py
pointer only (licence: NONE) · get_code("769b77d02a742c60")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, EstherBear/implementation-of-pruning-filters 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.

unwrap_model Ran
ran; our contract draft was wrong, not the code
siyuan0/pytorch_model_prune/prune.py
code served (permissive licence) · get_code("f21b8549dee8b153")
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 ModuleList. 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)

# Build a model with nested Sequential and ModuleList structures
# The function unwrap_model recursively traverses and flattens them

model = nn.Sequential(
    nn.Linear(4, 3),
    nn.Sequential(
        nn.Linear(3, 2),
        nn.ReLU()
    ),
    nn.ModuleList([
        nn.Linear(2, 5),
        nn.Sequential(
            nn.Linear(5, 3),
            nn.Tanh()
        )
    ]),
    nn.Conv2d(1, 2, 3)
)

ARGS = {
    "model": model
}
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 ModuleList — a ModuleList output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ModuleList 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
zero_padding Ran
ran on a synthesized input · constructed an object; no result was computed
siyuan0/pytorch_model_prune/prune.py
code served (permissive licence) · get_code("be78e984674fdadb")
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, zero_padding: 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)

# Analyze the function:
# - __init__ takes: num_channels (int), keep_channel_idx (int or indexable)
# - forward takes: x (tensor with shape [batch, channels, height, width])
# - The forward creates output of shape [batch, num_channels, height, width]
# - It assigns x to output[:, keep_channel_idx, :, :], so x should have shape [batch, 1, height, width]
#   (or x could be the size of keep_channel_idx if it's a list, but the assignment suggests single channel)

# Create a small input tensor: batch=2, channels=1, height=4, width=3
x = torch.randn(2, 1, 4, 3)

# Parameters for zero_padding
num_channels = 5  # output will have 5 channels
keep_channel_idx = 2  # the input channel will be placed at index 2 in output

ARGS = {
    "num_channels": num_channels,
    "keep_channel_idx": keep_channel_idx,
}
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 zero_padding — 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 zero_padding 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
MetaPruner Not yet run
unverified (harvested, no recorded run)
mingsun-tse/regularization-pruning/pruner/l1_pruner.py
pointer only (licence: NONE) · get_code("841ab10a9e621c3e")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, mingsun-tse/regularization-pruning 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. No invocation recorded.

OptimizerPruner Not yet run
unverified (harvested, no recorded run)
marcoancona/TorchPruner/torchpruner/pruner/pruner.py
code served (permissive licence) · get_code("b5096faa1be476c7")
How it ran

no invocation recorded.

Pruner Not yet run
unverified (harvested, no recorded run)
mingsun-tse/regularization-pruning/pruner/l1_pruner.py
pointer only (licence: NONE) · get_code("eb229809230a2c98")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, mingsun-tse/regularization-pruning 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. No invocation recorded.

_weights_init Not yet run
unverified (harvested, no recorded run)
mingsun-tse/regularization-pruning/pruner/l1_pruner.py
pointer only (licence: NONE) · get_code("8dbc5200982548c5")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, mingsun-tse/regularization-pruning 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. No invocation recorded.

accuracy Not yet run
unverified (harvested, no recorded run)
lehduong/ginp/imagenet/l1-norm-pruning/ensemble_finetune.py
pointer only (licence: MIT); the stored copy get_code serves, from kkonstantinidis/ByzShield, is pointer only (licence: NONE) · get_code("4d3faaaa1c706b0b")
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 kkonstantinidis/ByzShield, 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. No invocation recorded.

count_model_param_flops Not yet run
unverified (harvested, no recorded run)
siyuan0/pytorch_model_prune/prune.py
code served (permissive licence) · get_code("ee1b9c63cd1f32ce")
How it ran

no invocation recorded.

prune_model Not yet run
unverified (harvested, no recorded run)
siyuan0/pytorch_model_prune/prune.py
code served (permissive licence) · get_code("3adf7e9d6eff8e41")
How it ran

no invocation recorded.

prune_net Not yet run
unverified (harvested, no recorded run)
EstherBear/implementation-of-pruning-filters/prune.py
pointer only (licence: NONE) · get_code("85de63fef209d807")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, EstherBear/implementation-of-pruning-filters 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. No invocation recorded.

prune_resnet Not yet run
unverified (harvested, no recorded run)
EstherBear/implementation-of-pruning-filters/prune.py
pointer only (licence: NONE) · get_code("85dea7b7b4ee357b")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, EstherBear/implementation-of-pruning-filters 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. No invocation recorded.

prune_vgg Not yet run
unverified (harvested, no recorded run)
EstherBear/implementation-of-pruning-filters/prune.py
pointer only (licence: NONE) · get_code("fd290760eb0c67b9")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, EstherBear/implementation-of-pruning-filters 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. No invocation recorded.

validate Not yet run
unverified (harvested, no recorded run)
lehduong/ginp/imagenet/l1-norm-pruning/ensemble_finetune.py
code served (permissive licence) · get_code("15487f2b7b36c9cc")
How it ran

no invocation recorded.

validate Not yet run
unverified (harvested, no recorded run)
lehduong/ginp/imagenet/l1-norm-pruning/main_finetune.py
code served (permissive licence) · get_code("1dcc89e293ab6570")
How it ran

no invocation recorded.

validate Not yet run
unverified (harvested, no recorded run)
lehduong/ginp/imagenet/weight-level/main_finetune.py
code served (permissive licence) · get_code("61e4d0bc645ebb64")
How it ran

no invocation recorded.

Also present in other papers' code

Of 36 samples lifted from this paper's linked repositories, 36 carry a function identity and 0 do not. Of the 36 distinct functions with an identity, 9 are also present in other papers whose code we hold (3 with at least one paper dated earlier, 2 the same month, 9 later) and 27 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
conv3x3 686 other papers
12 dated earlier, 1 the same month, 673 later · 31 only as an assembly brick
1406.2661 (2014-06)
Generative Adversarial Networks
1409.0575 (2014-09)
ImageNet Large Scale Visual Recognition Challenge
1412.6572 (2014-12)
Explaining and Harnessing Adversarial Examples
and 683 more not named here
accuracy 80 other papers
4 dated earlier, 76 later
1307.0414 (2013-07)
Challenges in Representation Learning: A report on three machine learning contests
1412.6553 (2014-12)
Speeding-up Convolutional Neural Networks Using Fine-tuned CP-Decomposition
1503.02531 (2015-03)
Distilling the Knowledge in a Neural Network
and 77 more not named here
accuracy 23 other papers
1 dated earlier, 1 the same month, 21 later
1603.07285 (2016-03)
A guide to convolution arithmetic for deep learning
1608.07017 (2016-08)
Ambient Sound Provides Supervision for Visual Learning
1803.09877 (2018-03)
DRACO: Byzantine-resilient Distributed Training via Redundant Gradients
and 20 more not named here
MetaPruner
brick of this paper's assembly: helper
1 other paper
1 later
2012.09243 (2020-12)
Neural Pruning via Growing Regularization
create_l1_norm_mask 1 other paper
1 later
2006.11487 (2020-06)
Paying more attention to snapshots of Iterative Pruning: Improving Model Compression via Ensemble Distillation
get_conv_zero_param 1 other paper
1 later
2006.11487 (2020-06)
Paying more attention to snapshots of Iterative Pruning: Improving Model Compression via Ensemble Distillation
validate 1 other paper
1 later
2006.11487 (2020-06)
Paying more attention to snapshots of Iterative Pruning: Improving Model Compression via Ensemble Distillation
validate 1 other paper
1 later
2006.11487 (2020-06)
Paying more attention to snapshots of Iterative Pruning: Improving Model Compression via Ensemble Distillation
validate 1 other paper
1 later
2006.11487 (2020-06)
Paying more attention to snapshots of Iterative Pruning: Improving Model Compression via Ensemble Distillation

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 success of CNNs in various applications is accompanied by a significant increase in the computation and parameter storage costs. Recent efforts toward reducing these overheads involve pruning and compressing the weights of various layers without hurting original accuracy. However, magnitude-based pruning of weights reduces a significant number of parameters from the fully connected layers and may not adequately reduce the computation costs in the convolutional layers due to irregular sparsity in the pruned networks. We present an acceleration method for CNNs, where we prune filters from CNNs that are identified as having a small effect on the output accuracy. By removing whole filters in the network together with their connecting feature maps, the computation costs are reduced significantly. In contrast to pruning weights, this approach does not result in sparse connectivity patterns. Hence, it does not need the support of sparse convolution libraries and can work with existing efficient BLAS libraries for dense matrix multiplications. We show that even simple filter pruning techniques can reduce inference costs for VGG-16 by up to 34% and ResNet-110 by up to 38% on CIFAR10 while regaining close to the original accuracy by retraining the networks.

For agents

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

get_harvested_code_for_paper("1608.08710")
get_code_for_paper("1608.08710")
have("1608.08710")

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

curl https://syntology.ai/api/ran/1608.08710.json

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

[![Syntology run record](https://syntology.ai/api/ran/1608.08710.svg)](https://syntology.ai/paper/1608.08710)

Connect an agent — have() is free.