SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1909.08174 · 2019

Gate Decorator: Global Filter Pruning Method for Accelerating Deep Convolutional Neural Networks

arXiv · PDF · Open in the Atlas

Code that ran

17 of 18 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 15 have no contract verdict; 1 failure 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/1909.08174.json · badge: /api/ran/1909.08174.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
youzhonghui/gate-decorator-pruning canonical (classifier label)
classifier basis: census
4 ran · 1 no recorded run
softsys4ai/neural-distiller community (classifier label)
classifier basis not recorded on the link
13 ran · 0 no recorded run
FunctionStatusWhere it lives
bound_policy Ran
ran on a synthesized input
softsys4ai/neural-distiller/src/experiments/ternary_train.py
code served (permissive licence) · get_code("fa79dbf3466aa49e")
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

torch.manual_seed(0)

ARGS = {
    "w": 0.5
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type 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
create_meta Ran
ran on a synthesized input
softsys4ai/neural-distiller/src/experiments/multistage_knowledge_distillation_train.py
code served (permissive licence) · get_code("43ad5039f7bea39a")
How it ran

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

import datetime
import torch
import torch.nn as nn

torch.manual_seed(0)

# Define ARGS dict with values for all parameters of create_meta
ARGS = {
    'dataset': 'CIFAR10',
    'teacher_name': 'ResNet50',
    'epochs': 100,
    'temp': 4.0,
    'alpha': 0.5,
    'order': 'sequential'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
create_result Ran
ran on a synthesized input
softsys4ai/neural-distiller/src/experiments/multistage_knowledge_distillation_train.py
code served (permissive licence) · get_code("b054430125f1402e")
How it ran

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

import torch
import datetime

torch.manual_seed(0)

# Analyze the function:
# - netSize: converted to str, no specific requirements
# - temp: converted to str, no specific requirements
# - alpha: converted to str, no specific requirements
# - train_score: indexed with [0] and [1], so it's a tuple or list with at least 2 elements
# - val_score: indexed with [0] and [1], so it's a tuple or list with at least 2 elements

# train_score and val_score appear to be (loss, accuracy) tuples based on usage
train_score = (0.45, 0.92)  # (loss, accuracy)
val_score = (0.52, 0.89)    # (loss, accuracy)

ARGS = {
    'netSize': 64,
    'temp': 0.5,
    'alpha': 0.01,
    'train_score': train_score,
    'val_score': val_score
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
finetune Ran
ran on a synthesized input
youzhonghui/gate-decorator-pruning/prune/utils.py
code served (permissive licence) · get_code("443447da85da3fcc")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
    
    def forward(self, x):
        return self.linear(x)

class SimpleOptimizer:
    def __init__(self, params):
        self.param_groups = [{'params': list(params), 'lr': 0.01, 'momentum': 0.9}]
    
    def zero_grad(self):
        pass
    
    def step(self):
        pass

class SimpleTrainer:
    def __init__(self, model):
        self.model = model
    
    def train(self, pack, iter_hook=None):
        # Simulate training with iter_hook calls
        total_iter = 2
        for i in range(total_iter):
            if iter_hook:
                iter_hook(i, total_iter)
        return {'train_loss': 0.5, 'train_acc': 0.9}
    
    def test(self, pack):
        return {'test_loss': 0.6, 'test_acc': 0.85}

class Pack:
    def __init__(self):
        self.model = SimpleModel()
        self.optimizer = SimpleOptimizer(self.model.parameters())
        self.trainer = SimpleTrainer(self.model)

pack = Pack()
lr_min = 0.001
lr_max = 0.1
T = 2
mute = False

ARGS = {
    'pack': pack,
    'lr_min': lr_min,
    'lr_max': lr_max,
    'T': T,
    'mute': mute,
}
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 1, dtype family float — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
import_config Ran
ran on a synthesized input
softsys4ai/neural-distiller/src/experiments/multistage_knowledge_distillation_train.py
code served (permissive licence) · get_code("8fa48caebcb5f044")
How it ran

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

import json
import tempfile
import os

# Create a temporary JSON config file
temp_dir = tempfile.mkdtemp()
config_file_path = os.path.join(temp_dir, "test_config.json")

# Create a sample configuration
config_data = {
    "model": "bert",
    "learning_rate": 0.001,
    "batch_size": 32,
    "epochs": 10,
    "hidden_dim": 768,
    "layers": [1, 2, 3]
}

# Write the config to the temporary file
with open(config_file_path, 'w') as f:
    json.dump(config_data, f)

ARGS = {
    "config_file_path": config_file_path
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
lr_schedule Ran
ran, honoured the contract we drafted
softsys4ai/neural-distiller/src/models/generate_resnet.py
pointer only (licence: MIT); the stored copy get_code serves, from Jeongyun-Lee-0423/Classification, is pointer only (licence: NONE) · get_code("67e18e3fdd805d03")
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 Jeongyun-Lee-0423/Classification, 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.

resnet20 Ran
ran; our contract draft was wrong, not the code
youzhonghui/gate-decorator-pruning/models/cifar/resnet56.py
code served (permissive licence) · get_code("025a0fa4b9bd03c4")
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 ResNet. 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 = {
    "num_classes": 10
}
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 ResNet — a ResNet output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ResNet 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
resnet32 Ran
ran on a synthesized input
youzhonghui/gate-decorator-pruning/models/cifar/resnet56.py
code served (permissive licence) · get_code("aee5de704735856b")
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 ResNet. 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 = {
    "num_classes": 10
}
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 ResNet — a ResNet output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ResNet 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
resnet44 Ran
ran on a synthesized input
youzhonghui/gate-decorator-pruning/models/cifar/resnet56.py
code served (permissive licence) · get_code("5cdc625b5824a25a")
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 ResNet. 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 = {
    "num_classes": 10
}
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 ResNet — a ResNet output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ResNet 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
save_model Ran
ran on a synthesized input
softsys4ai/neural-distiller/src/experiments/eskd_knowledge_distillation_train.py
code served (permissive licence) · get_code("d016fd58a1a0d7e8")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned tuple, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import os
import tempfile
import torch
import torch.nn as nn

# Create a minimal Keras-like model class with save and save_weights methods
class MinimalModel:
    def __init__(self):
        self.saved_model_path = None
        self.saved_weights_path = None
    
    def save(self, path):
        """Mock save method that writes a dummy file"""
        os.makedirs(os.path.dirname(path), exist_ok=True)
        with open(path, 'w') as f:
            f.write("mock_model_data")
        self.saved_model_path = path
    
    def save_weights(self, path):
        """Mock save_weights method that writes a dummy file"""
        os.makedirs(os.path.dirname(path), exist_ok=True)
        with open(path, 'w') as f:
            f.write("mock_weights_data")
        self.saved_weights_path = path

# Create temporary directory for model storage
temp_dir = tempfile.mkdtemp()

# Construct ARGS dictionary
ARGS = {
    'models_dir': temp_dir,
    'model': MinimalModel(),
    'model_size': 'small',
    'curr_epochs': 5,
    'total_epochs': 10,
    'train_temp': 0.8,
    'val_acc': 0.92,
    'train_acc': 0.95
}
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 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
save_model Ran
ran on a synthesized input
softsys4ai/neural-distiller/src/experiments/eskd_teacher_train_and_collect_logits.py
code served (permissive licence) · get_code("1ff169b5b9d1cd91")
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 str. 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 os
import tempfile
import torch
import torch.nn as nn

torch.manual_seed(0)

# Create a temporary directory for saving models
models_dir = tempfile.mkdtemp()

# Create a minimal model class that has a save method
class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
    
    def forward(self, x):
        return self.linear(x)
    
    def save(self, path):
        """Mock save method that writes to file"""
        with open(path, 'w') as f:
            f.write("model_data")

# Instantiate the model
model = SimpleModel()

# Define all arguments
ARGS = {
    "models_dir": models_dir,
    "model": model,
    "model_size": "small",
    "curr_epochs": 5,
    "total_epochs": 10,
    "val_acc": 0.85,
    "train_acc": 0.90
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
save_weights Ran
ran on a synthesized input
softsys4ai/neural-distiller/src/experiments/eskd_baseline_train.py
code served (permissive licence) · get_code("72fd99b24d34b031")
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 str. 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 os
import torch
import torch.nn as nn
import tempfile

torch.manual_seed(0)

# Create a temporary directory for models
models_dir = tempfile.mkdtemp()

# Create a minimal model with save_weights method
class MinimalModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
    
    def forward(self, x):
        return self.linear(x)
    
    def save_weights(self, path):
        """Save model weights to an h5 file"""
        torch.save(self.state_dict(), path)

model = MinimalModel()

ARGS = {
    'models_dir': models_dir,
    'model': model,
    'model_size': 'small',
    'model_num': 1,
    'total_epochs': 10,
    'val_acc': 0.95,
    'train_acc': 0.98
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
save_weights Ran
ran on a synthesized input
softsys4ai/neural-distiller/src/experiments/eskd_knowledge_distillation_train.py
code served (permissive licence) · get_code("5751220b37e6b368")
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 str. 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 tempfile
import os


class SimpleModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
    
    def forward(self, x):
        return self.linear(x)
    
    def save_weights(self, path):
        """Mock method to simulate model weight saving"""
        torch.save(self.state_dict(), path)


torch.manual_seed(0)

model = SimpleModel()
models_dir = tempfile.gettempdir()
model_size = "small"
curr_epochs = 5
total_epochs = 10
train_temp = 0.95
val_acc = 0.87
train_acc = 0.92

ARGS = {
    "models_dir": models_dir,
    "model": model,
    "model_size": model_size,
    "curr_epochs": curr_epochs,
    "total_epochs": total_epochs,
    "train_temp": train_temp,
    "val_acc": val_acc,
    "train_acc": train_acc,
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
save_weights Ran
ran on a synthesized input
softsys4ai/neural-distiller/src/experiments/eskd_teacher_train_and_collect_logits.py
code served (permissive licence) · get_code("3a54923ed094ac96")
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 str. 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 tempfile
import os


class MockKerasModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 3)
    
    def save_weights(self, path):
        """Mock save_weights method that mimics Keras API"""
        torch.save(self.state_dict(), path)


# Create a temporary directory for models
models_dir = tempfile.mkdtemp()

# Create a mock model with save_weights method
model = MockKerasModel()

# Define other parameters
curr_epochs = 5
val_acc = 0.92
train_acc = 0.95

ARGS = {
    "models_dir": models_dir,
    "model": model,
    "curr_epochs": curr_epochs,
    "val_acc": val_acc,
    "train_acc": train_acc,
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
step_decay Ran
ran on a synthesized input
softsys4ai/neural-distiller/src/experiments/eskd_baseline_train.py
code served (permissive licence) · get_code("3c6c1687bc93411d")
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

torch.manual_seed(0)

ARGS = {
    'epoch': 75
}
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
step_decay Ran
ran on a synthesized input
softsys4ai/neural-distiller/src/experiments/eskd_teacher_train_and_collect_logits.py
code served (permissive licence) · get_code("39a9b931bf412ce2")
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)

ARGS = {
    "epoch": 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 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
ternarize Ran
ran on a synthesized input
softsys4ai/neural-distiller/src/experiments/ternary_train.py
code served (permissive licence) · get_code("b822412a30b86c96")
How it ran

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

import torch
import numpy as np

torch.manual_seed(0)

# The function ternarize takes a single numeric value w
# and returns -1, 0, or 1 based on threshold comparisons
# We need to provide a value for w that can be compared with floats

w = 0.5

ARGS = {
    "w": w
}
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 int, ndim 0, dtype family int, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
get_gate_sparse_loss Not yet run
unverified (harvested, no recorded run)
youzhonghui/gate-decorator-pruning/prune/universal.py
code served (permissive licence) · get_code("4f18686562b60d83")
How it ran

no invocation recorded.

Also present in other papers' code

Of 18 samples lifted from this paper's linked repositories, 18 carry a function identity and 0 do not. Of the 18 distinct functions with an identity, 2 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 2 later) and 16 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
lr_schedule 7 other papers
6 dated earlier, 1 later
1511.06807 (2015-11)
Adding Gradient Noise Improves Learning for Very Deep Networks
1603.05027 (2016-03)
Identity Mappings in Deep Residual Networks
1807.06521 (2018-07)
CBAM: Convolutional Block Attention Module
and 4 more not named here
resnet20 2 other papers
2 later
2305.18497 (2023-05)
Collaborative Learning via Prediction Consensus
2403.19863 (2024-03)
DeNetDM: Debiasing by Network Depth Modulation

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

Filter pruning is one of the most effective ways to accelerate and compress convolutional neural networks (CNNs). In this work, we propose a global filter pruning algorithm called Gate Decorator, which transforms a vanilla CNN module by multiplying its output by the channel-wise scaling factors, i.e. gate. When the scaling factor is set to zero, it is equivalent to removing the corresponding filter. We use Taylor expansion to estimate the change in the loss function caused by setting the scaling factor to zero and use the estimation for the global filter importance ranking. Then we prune the network by removing those unimportant filters. After pruning, we merge all the scaling factors into its original module, so no special operations or structures are introduced. Moreover, we propose an iterative pruning framework called Tick-Tock to improve pruning accuracy. The extensive experiments demonstrate the effectiveness of our approaches. For example, we achieve the state-of-the-art pruning ratio on ResNet-56 by reducing 70% FLOPs without noticeable loss in accuracy. For ResNet-50 on ImageNet, our pruned model with 40% FLOPs reduction outperforms the baseline model by 0.31% in top-1 accuracy. Various datasets are used, including CIFAR-10, CIFAR-100, CUB-200, ImageNet ILSVRC-12 and PASCAL VOC 2011. Code is available at github.com/youzhonghui/gate-decorator-pruning

For agents

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

get_harvested_code_for_paper("1909.08174")
get_code_for_paper("1909.08174")
have("1909.08174")

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

curl https://syntology.ai/api/ran/1909.08174.json

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

[![Syntology run record](https://syntology.ai/api/ran/1909.08174.svg)](https://syntology.ai/paper/1909.08174)

Connect an agent — have() is free.