SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1901.08164 · 2019

Decoupled Greedy Learning of CNNs

arXiv · PDF · Open in the Atlas

Code that ran

8 of 14 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; 2 honoured the contract we drafted, 0 violated it and 4 have no contract verdict; 2 failures of our instrument; 6 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/1901.08164.json · badge: /api/ran/1901.08164.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
eugenium/DGL canonical (classifier label)
classifier basis: classifier_2025
1 ran · 1 no recorded run
batuozt/gleam community (classifier label)
classifier basis not recorded on the link
7 ran · 5 no recorded run
FunctionStatusWhere it lives
approximately_split Ran
ran on a synthesized input
batuozt/gleam/datasets/utils/data_partition.py
code served (permissive licence) · get_code("9d2d1a47f9b68657")
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. 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 random
from copy import deepcopy
from typing import Any, List, Union
import numpy as np

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

# Create test data
values = ['item_0', 'item_1', 'item_2', 'item_3', 'item_4', 'item_5', 'item_6', 'item_7', 'item_8', 'item_9']
split_percentages = [0.5, 0.3, 0.2]

ARGS = {
    'values': values,
    'split_percentages': split_percentages,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list — a list output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
convert_basic_c2_names Ran
ran; our contract draft was wrong, not the code
batuozt/gleam/ss_recon/checkpoint/c2_model_loading.py
code served (permissive licence) · get_code("e8526a516b4f9206")
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 copy
import re

# Analyze the function: it takes a list of strings (original_keys) and applies
# various string transformations to them. It doesn't use any complex objects,
# just list operations and string manipulations.

ARGS = {
    'original_keys': [
        'pred_b',
        'pred_w',
        'conv_b',
        'conv_w',
        'bn_s',
        'bn_bias',
        'bn_rm',
        'bn_riv',
        'bn_gamma',
        'bn_beta',
        'gn_s',
        'gn_bias',
        'res_conv1_norm_weight',
        'conv1_weight',
        'layer_branch1_weight',
        'layer_branch2a_bias',
        'layer_branch2b_weight',
        'layer_branch2c_bias',
        'body_conv_fcn_weight',
        'AnnIndex_lowres',
        'Index_UV_lowres',
        'U_lowres',
        'V_lowres',
    ]
}
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 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
convert_c2_detectron_names Ran
ran; our contract draft was wrong, not the code
batuozt/gleam/ss_recon/checkpoint/c2_model_loading.py
code served (permissive licence) · get_code("52810b504f0d1066")
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 torch
import logging
import copy

torch.manual_seed(0)

# Set up logging to avoid issues
logging.basicConfig(level=logging.INFO)

# Create a minimal weights dict with C2-style naming conventions
# The function processes weight names, so we just need dummy tensors
weights = {
    'res_conv1_bn_s': torch.randn(4),
    'res_conv1_bn_bias': torch.randn(4),
    'pred_w': torch.randn(3, 4),
    'pred_b': torch.randn(3),
    'conv_rpn_fpn2_weight': torch.randn(4, 4),
    'rpn_bbox_pred_weight': torch.randn(4, 4),
    'rpn_cls_logits_weight': torch.randn(4, 4),
    'fc6_weight': torch.randn(4, 4),
    'fc7_bias': torch.randn(4),
    'fpn_inner_res2_2_sum_lateral_weight': torch.randn(4, 4),
    'fpn_res2_2_sum_weight': torch.randn(4, 4),
    'fpn_inner_res3_3_sum_lateral_bias': torch.randn(4),
    'fpn_res3_3_sum_bias': torch.randn(4),
    'bbox_pred_weight': torch.randn(4, 4),
    'cls_score_bias': torch.randn(4),
    'head_conv_weight': torch.randn(4, 4),
}

ARGS = {
    'weights': weights,
}
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
find_wandb_exp_id Ran
ran on a synthesized input
batuozt/gleam/ss_recon/engine/defaults.py
code served (permissive licence) · get_code("4caeef87f87a4bb7")
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
from pathlib import Path

class CfgNode:
    """Minimal config class with required attributes."""
    def __init__(self, output_dir):
        self.OUTPUT_DIR = output_dir

# Create a temporary directory structure for testing
temp_dir = tempfile.mkdtemp()

# Test case 1: Create wandb_id file
os.makedirs(os.path.dirname(os.path.join(temp_dir, 'wandb_id')), exist_ok=True)
wandb_id_path = os.path.join(temp_dir, 'wandb_id')
with open(wandb_id_path, 'w') as f:
    f.write('test-exp-id-123')

cfg = CfgNode(output_dir=temp_dir)

ARGS = {
    'cfg': cfg
}
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 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
format_as_iter Ran
ran, honoured the contract we drafted
batuozt/gleam/ss_recon/engine/dgl_trainer.py
code served (permissive licence) · get_code("dbc9cac27d1c2130")
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
from typing import Sequence, Union

torch.manual_seed(0)

# Test case 1: single epoch value
vals_1 = 5
iters_per_epoch_1 = 100
time_scale_1 = "epoch"

# Test case 2: sequence of iteration values
vals_2 = [-10, -20, -30]
iters_per_epoch_2 = 50
time_scale_2 = "iter"

# Test case 3: single negative iteration value
vals_3 = -5
iters_per_epoch_3 = 100
time_scale_3 = "iter"

# Test case 4: sequence of epoch values
vals_4 = [2, 3, 4]
iters_per_epoch_4 = 75
time_scale_4 = "epoch"

# Using the first test case as the main ARGS
ARGS = {
    "vals": 5,
    "iters_per_epoch": 100,
    "time_scale": "epoch"
}
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 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_files Ran
ran on a synthesized input
batuozt/gleam/datasets/format_fastmri.py
code served (permissive licence) · get_code("f8eba979963c16e9")
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 os
import tempfile
import shutil

# Create a temporary directory with test files
temp_dir = tempfile.mkdtemp()

# Create some valid fastMRI files
open(os.path.join(temp_dir, 'file_001.h5'), 'w').close()
open(os.path.join(temp_dir, 'file_002.h5'), 'w').close()
open(os.path.join(temp_dir, 'file_test.h5'), 'w').close()

# Create some invalid files that should be ignored
open(os.path.join(temp_dir, 'invalid.h5'), 'w').close()
open(os.path.join(temp_dir, 'file_001.txt'), 'w').close()

ARGS = {
    'dir_path': temp_dir,
    'ignore_missing_dir': False,
    'filenames': None
}
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 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
is_valid_file Ran
ran on a synthesized input
batuozt/gleam/datasets/format_fastmri.py
code served (permissive licence) · get_code("7173d4cd30e0bb88")
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 bool, 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 os
import tempfile

def is_valid_file(fpath):
    """Returns `True` if file corresponds to a valid fastMRI file."""
    fname = os.path.basename(fpath)
    return fname.startswith('file') and fname.endswith('.h5')

# Create a temporary file with a valid fastMRI filename
with tempfile.TemporaryDirectory() as tmpdir:
    valid_fpath = os.path.join(tmpdir, 'file_sample.h5')
    open(valid_fpath, 'w').close()
    
    ARGS = {
        'fpath': valid_fpath
    }
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 bool, 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
lr_scheduler Ran
ran, honoured the contract we drafted
eugenium/DGL/dni_comparisons/cifar_cnn_dni.py
pointer only (licence: NONE) · get_code("f81566b35c748e30")
How it ran

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

approximately_split_weighted Not yet run
unverified (harvested, no recorded run)
batuozt/gleam/datasets/utils/data_partition.py
code served (permissive licence) · get_code("6ca55eea797cb3a8")
How it ran

no invocation recorded.

fftc Not yet run
unverified (harvested, no recorded run)
batuozt/gleam/datasets/utils/fftc.py
code served (permissive licence) · get_code("cc12efe97cafdfd8")
How it ran

no invocation recorded.

fftnc Not yet run
unverified (harvested, no recorded run)
batuozt/gleam/datasets/utils/fftc.py
code served (permissive licence) · get_code("fdc31294150d9ef4")
How it ran

no invocation recorded.

ifftnc Not yet run
unverified (harvested, no recorded run)
batuozt/gleam/datasets/utils/fftc.py
code served (permissive licence) · get_code("617c7d033c431ebe")
How it ran

no invocation recorded.

init_wandb_run Not yet run
unverified (harvested, no recorded run)
batuozt/gleam/ss_recon/engine/defaults.py
code served (permissive licence) · get_code("d916b52fc1564a94")
How it ran

no invocation recorded.

one_hot Not yet run
unverified (harvested, no recorded run)
eugenium/DGL/dni_comparisons/cifar_cnn_dni.py
pointer only (licence: NONE) · get_code("148b238d5ad66615")
How it ran

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

Also present in other papers' code

Of 14 samples lifted from this paper's linked repositories, 14 carry a function identity and 0 do not. Of the 14 distinct functions with an identity, 7 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 7 later) and 7 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
convert_basic_c2_names 45 other papers
2 dated earlier, 43 later
1703.06870 (2017-03)
Mask R-CNN
1804.08864 (2018-04)
Learning to See the Invisible: End-to-End Trainable Amodal Instance Segmentation
1904.07850 (2019-04)
Objects as Points
and 42 more not named here
convert_c2_detectron_names 29 other papers
2 dated earlier, 27 later
1703.06870 (2017-03)
Mask R-CNN
1804.08864 (2018-04)
Learning to See the Invisible: End-to-End Trainable Amodal Instance Segmentation
1904.07850 (2019-04)
Objects as Points
and 26 more not named here
fftc 1 other paper
1 later
1903.07824 (2019-03)
Compressed Sensing: From Research to Clinical Practice with Data-Driven Learning
fftnc 1 other paper
1 later
1903.07824 (2019-03)
Compressed Sensing: From Research to Clinical Practice with Data-Driven Learning
find_wandb_exp_id 1 other paper
1 later
2111.02549 (2021-11)
VORTEX: Physics-Driven Data Augmentations Using Consistency Training for Robust Accelerated MRI Reconstruction
format_as_iter 1 other paper
1 later
2111.02549 (2021-11)
VORTEX: Physics-Driven Data Augmentations Using Consistency Training for Robust Accelerated MRI Reconstruction
ifftnc 1 other paper
1 later
1903.07824 (2019-03)
Compressed Sensing: From Research to Clinical Practice with Data-Driven Learning

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

A commonly cited inefficiency of neural network training by back-propagation is the update locking problem: each layer must wait for the signal to propagate through the full network before updating. Several alternatives that can alleviate this issue have been proposed. In this context, we consider a simpler, but more effective, substitute that uses minimal feedback, which we call Decoupled Greedy Learning (DGL). It is based on a greedy relaxation of the joint training objective, recently shown to be effective in the context of Convolutional Neural Networks (CNNs) on large-scale image classification. We consider an optimization of this objective that permits us to decouple the layer training, allowing for layers or modules in networks to be trained with a potentially linear parallelization in layers. With the use of a replay buffer we show this approach can be extended to asynchronous settings, where modules can operate with possibly large communication delays. We show theoretically and empirically that this approach converges. Then, we empirically find that it can lead to better generalization than sequential greedy optimization. We demonstrate the effectiveness of DGL against alternative approaches on the CIFAR-10 dataset and on the large-scale ImageNet dataset.

For agents

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

get_harvested_code_for_paper("1901.08164")
get_code_for_paper("1901.08164")
have("1901.08164")

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

curl https://syntology.ai/api/ran/1901.08164.json

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

[![Syntology run record](https://syntology.ai/api/ran/1901.08164.svg)](https://syntology.ai/paper/1901.08164)

Connect an agent — have() is free.