SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1904.12043 · 2019

Dynamic Mini-batch SGD for Elastic Distributed Training: Learning in the Limbo of Resources

arXiv · PDF · Open in the Atlas

Code that ran

13 of 20 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 12 have no contract verdict; 1 failure of our instrument; 7 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/1904.12043.json · badge: /api/ran/1904.12043.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
awslabs/dynamic-training-with-apache-mxnet-on-aws community (classifier label)
classifier basis not recorded on the link
9 ran · 5 no recorded run
AlanFokCo/compensation-tools community (classifier label)
classifier basis not recorded on the link
4 ran · 2 no recorded run
FunctionStatusWhere it lives
bulk Ran
ran on a synthesized input
awslabs/dynamic-training-with-apache-mxnet-on-aws/python/mxnet/engine.py
pointer only (licence: Apache-2.0) · get_code("f235e1530ce3d53d")
How it ran

The body of this sample is withheld: Apache-2.0, as GitHub detects it on this paper's repository, awslabs/dynamic-training-with-apache-mxnet-on-aws, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with it. 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.

c_array Ran
ran on a synthesized input
awslabs/dynamic-training-with-apache-mxnet-on-aws/amalgamation/python/mxnet_predict.py
pointer only (licence: Apache-2.0) · get_code("3bc3ed8fa916a182")
How it ran

The body of this sample is withheld: Apache-2.0, as GitHub detects it on this paper's repository, awslabs/dynamic-training-with-apache-mxnet-on-aws, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with it. 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.

c_array Ran
ran on a synthesized input
awslabs/dynamic-training-with-apache-mxnet-on-aws/python/mxnet/base.py
pointer only (licence: Apache-2.0) · get_code("c037ab026fd36b3a")
How it ran

The body of this sample is withheld: Apache-2.0, as GitHub detects it on this paper's repository, awslabs/dynamic-training-with-apache-mxnet-on-aws, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with it. 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.

c_str Ran
ran on a synthesized input
awslabs/dynamic-training-with-apache-mxnet-on-aws/amalgamation/python/mxnet_predict.py
pointer only (licence: Apache-2.0) · get_code("313d72408f5f6086")
How it ran

The body of this sample is withheld: Apache-2.0, as GitHub detects it on this paper's repository, awslabs/dynamic-training-with-apache-mxnet-on-aws, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with it. 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.

classproperty Ran
ran on a synthesized input
awslabs/dynamic-training-with-apache-mxnet-on-aws/python/mxnet/base.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from hpi-xnor/BMXNet-v2, is pointer only (licence: Apache-2.0) · get_code("47ebd1622534322e")
How it ran

The body of this sample is withheld: Apache-2.0, as GitHub detects it on this paper's repository, awslabs/dynamic-training-with-apache-mxnet-on-aws, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with it. 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.

do_checkpoint Ran
ran on a synthesized input
awslabs/dynamic-training-with-apache-mxnet-on-aws/python/mxnet/callback.py
pointer only (licence: Apache-2.0) · get_code("df77b3eca6bd676b")
How it ran

The body of this sample is withheld: Apache-2.0, as GitHub detects it on this paper's repository, awslabs/dynamic-training-with-apache-mxnet-on-aws, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with it. 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.

load_data Ran
ran on a synthesized input
AlanFokCo/compensation-tools/elastic_demos/data_process.py
code served (permissive licence) · get_code("3c9ed3332c1f6ca8")
How it ran

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

import numpy as np
import tempfile
import os

# Create a temporary .npz file with the expected structure
temp_dir = tempfile.mkdtemp()
temp_path = os.path.join(temp_dir, "test_data.npz")

# Create sample data: 2 images of shape (64, 64, 3) and labels
sample_data = np.random.rand(2, 64, 64, 3).astype(np.float32)
sample_labels = np.array([0, 1])

# Save as .npz file with named arrays 'data' and 'labels'
np.savez(temp_path, data=sample_data, labels=sample_labels)

ARGS = {
    "path": temp_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
load_data_single Ran
ran on a synthesized input
AlanFokCo/compensation-tools/elastic_demos/add_data.py
code served (permissive licence) · get_code("2d6b2877daa0effe")
How it ran

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

import pickle
import numpy as np
import tempfile
import os

# Create a temporary pickle file with the expected structure
temp_dir = tempfile.mkdtemp()
temp_file = os.path.join(temp_dir, 'test_data.pkl')

# Create sample data matching CIFAR-10 format
sample_data = {
    b'data': np.random.randint(0, 256, (2, 3072), dtype=np.uint8),
    b'labels': [0, 1]
}

with open(temp_file, 'wb') as f:
    pickle.dump(sample_data, f)

ARGS = {
    'file': temp_file
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
log_train_metric Ran
ran on a synthesized input
awslabs/dynamic-training-with-apache-mxnet-on-aws/python/mxnet/callback.py
pointer only (licence: Apache-2.0) · get_code("5bc9dd9627f0eb1c")
How it ran

The body of this sample is withheld: Apache-2.0, as GitHub detects it on this paper's repository, awslabs/dynamic-training-with-apache-mxnet-on-aws, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with it. 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.

map_func Ran
ran on a synthesized input
AlanFokCo/compensation-tools/elastic_demos/tensorflow_v2_cifar_resnet.py
code served (permissive licence) · get_code("d46ae594e2adad67")
How it ran

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

import tensorflow as tf
import numpy as np

# Create a valid TFRecord example
def create_tfrecord_example():
    # Create sample data: 32x32x3 image flattened to bytes
    image_data = np.random.randint(0, 256, (32, 32, 3), dtype=np.uint8)
    image_bytes = image_data.tobytes()
    
    # Create a feature dict
    feature = {
        'data': tf.train.Feature(bytes_list=tf.train.BytesList(value=[image_bytes])),
        'label': tf.train.Feature(int64_list=tf.train.Int64List(value=[5]))
    }
    
    # Create an Example proto
    example_proto = tf.train.Example(features=tf.train.Features(feature=feature))
    
    # Serialize to bytes
    serialized_example = example_proto.SerializeToString()
    return serialized_example

# Generate the example
example = create_tfrecord_example()

ARGS = {
    'example': example
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
module_checkpoint Ran
ran on a synthesized input
awslabs/dynamic-training-with-apache-mxnet-on-aws/python/mxnet/callback.py
pointer only (licence: Apache-2.0) · get_code("4950353b99f3413d")
How it ran

The body of this sample is withheld: Apache-2.0, as GitHub detects it on this paper's repository, awslabs/dynamic-training-with-apache-mxnet-on-aws, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with it. 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.

record Ran
ran on a synthesized input
awslabs/dynamic-training-with-apache-mxnet-on-aws/python/mxnet/autograd.py
pointer only (licence: Apache-2.0) · get_code("b86308bc3bc53b49")
How it ran

The body of this sample is withheld: Apache-2.0, as GitHub detects it on this paper's repository, awslabs/dynamic-training-with-apache-mxnet-on-aws, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with it. 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.

unpickle Ran
ran; our contract draft was wrong, not the code
AlanFokCo/compensation-tools/elastic_demos/add_data.py
code served (permissive licence) · get_code("91b57896044b9479")
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 pickle
import tempfile
import os

# Create a temporary file with pickled data
temp_dir = tempfile.mkdtemp()
temp_file = os.path.join(temp_dir, "test.pkl")

# Create some sample data and pickle it
sample_data = {
    b'key1': [1, 2, 3],
    b'key2': {'nested': 'value'},
    b'key3': 42
}

with open(temp_file, 'wb') as f:
    pickle.dump(sample_data, f)

# Define the arguments for the unpickle function
ARGS = {
    'file': temp_file
}
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
create Not yet run
unverified (harvested, no recorded run)
awslabs/dynamic-training-with-apache-mxnet-on-aws/python/mxnet/kvstore.py
pointer only (licence: Apache-2.0) · get_code("9ff2673d6a779a1a")
How it ran

The body of this sample is withheld: Apache-2.0, as GitHub detects it on this paper's repository, awslabs/dynamic-training-with-apache-mxnet-on-aws, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with it. 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.

load_val_data Not yet run
unverified (harvested, no recorded run)
AlanFokCo/compensation-tools/elastic_demos/data_process.py
code served (permissive licence) · get_code("e987ba7540887ac6")
How it ran

no invocation recorded.

normalization Not yet run
unverified (harvested, no recorded run)
AlanFokCo/compensation-tools/elastic_demos/data_process.py
code served (permissive licence) · get_code("b98592423270ab35")
How it ran

no invocation recorded.

set_bulk_size Not yet run
unverified (harvested, no recorded run)
awslabs/dynamic-training-with-apache-mxnet-on-aws/python/mxnet/engine.py
pointer only (licence: Apache-2.0) · get_code("8bcbec3a30a50379")
How it ran

The body of this sample is withheld: Apache-2.0, as GitHub detects it on this paper's repository, awslabs/dynamic-training-with-apache-mxnet-on-aws, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with it. 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.

set_recording Not yet run
unverified (harvested, no recorded run)
awslabs/dynamic-training-with-apache-mxnet-on-aws/python/mxnet/autograd.py
pointer only (licence: Apache-2.0) · get_code("84a2b5addc51f455")
How it ran

The body of this sample is withheld: Apache-2.0, as GitHub detects it on this paper's repository, awslabs/dynamic-training-with-apache-mxnet-on-aws, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with it. 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.

set_training Not yet run
unverified (harvested, no recorded run)
awslabs/dynamic-training-with-apache-mxnet-on-aws/python/mxnet/autograd.py
pointer only (licence: Apache-2.0) · get_code("565479a708f0d1c6")
How it ran

The body of this sample is withheld: Apache-2.0, as GitHub detects it on this paper's repository, awslabs/dynamic-training-with-apache-mxnet-on-aws, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with it. 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.

with_metaclass Not yet run
unverified (harvested, no recorded run)
awslabs/dynamic-training-with-apache-mxnet-on-aws/python/mxnet/base.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from hpi-xnor/BMXNet-v2, is pointer only (licence: Apache-2.0) · get_code("05550ef10b65470b")
How it ran

The body of this sample is withheld: Apache-2.0, as GitHub detects it on this paper's repository, awslabs/dynamic-training-with-apache-mxnet-on-aws, is in the permissive set, but a licence decision keeps the body a pointer: the reason for that decision is not recorded with it. 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 20 samples lifted from this paper's linked repositories, 20 carry a function identity and 0 do not. Of the 20 distinct functions with an identity, 12 are also present in other papers whose code we hold (6 with at least one paper dated earlier, 0 the same month, 11 later) and 8 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
do_checkpoint 3 other papers
1 dated earlier, 2 later
1512.01274 (2015-12)
MXNet: A Flexible and Efficient Machine Learning Library for Heterogeneous Distributed Systems
1905.03960 (2019-05)
Priority-based Parameter Propagation for Distributed DNN Training
2110.06804 (2021-10)
A comprehensive review of Binary Neural Network
record 3 other papers
1 dated earlier, 2 later
1512.01274 (2015-12)
MXNet: A Flexible and Efficient Machine Learning Library for Heterogeneous Distributed Systems
1905.03960 (2019-05)
Priority-based Parameter Propagation for Distributed DNN Training
2110.06804 (2021-10)
A comprehensive review of Binary Neural Network
set_recording 3 other papers
1 dated earlier, 2 later
1512.01274 (2015-12)
MXNet: A Flexible and Efficient Machine Learning Library for Heterogeneous Distributed Systems
1905.03960 (2019-05)
Priority-based Parameter Propagation for Distributed DNN Training
2110.06804 (2021-10)
A comprehensive review of Binary Neural Network
set_training 3 other papers
1 dated earlier, 2 later
1512.01274 (2015-12)
MXNet: A Flexible and Efficient Machine Learning Library for Heterogeneous Distributed Systems
1905.03960 (2019-05)
Priority-based Parameter Propagation for Distributed DNN Training
2110.06804 (2021-10)
A comprehensive review of Binary Neural Network
c_array 2 other papers
2 later
1905.03960 (2019-05)
Priority-based Parameter Propagation for Distributed DNN Training
2110.06804 (2021-10)
A comprehensive review of Binary Neural Network
c_str 2 other papers
2 later
1905.03960 (2019-05)
Priority-based Parameter Propagation for Distributed DNN Training
2110.06804 (2021-10)
A comprehensive review of Binary Neural Network
log_train_metric 2 other papers
1 dated earlier, 1 later
1512.01274 (2015-12)
MXNet: A Flexible and Efficient Machine Learning Library for Heterogeneous Distributed Systems
1905.03960 (2019-05)
Priority-based Parameter Propagation for Distributed DNN Training
module_checkpoint 2 other papers
2 later
1905.03960 (2019-05)
Priority-based Parameter Propagation for Distributed DNN Training
2110.06804 (2021-10)
A comprehensive review of Binary Neural Network
unpickle 2 other papers
2 dated earlier
1502.03044 (2015-02)
Show, Attend and Tell: Neural Image Caption Generation with Visual Attention
1806.00451 (2018-06)
Do CIFAR-10 Classifiers Generalize to CIFAR-10?
c_array 1 other paper
1 later
1905.03960 (2019-05)
Priority-based Parameter Propagation for Distributed DNN Training
classproperty 1 other paper
1 later
2110.06804 (2021-10)
A comprehensive review of Binary Neural Network
with_metaclass 1 other paper
1 later
2110.06804 (2021-10)
A comprehensive review of Binary Neural Network

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

With an increasing demand for training powers for deep learning algorithms and the rapid growth of computation resources in data centers, it is desirable to dynamically schedule different distributed deep learning tasks to maximize resource utilization and reduce cost. In this process, different tasks may receive varying numbers of machines at different time, a setting we call elastic distributed training. Despite the recent successes in large mini-batch distributed training, these methods are rarely tested in elastic distributed training environments and suffer degraded performance in our experiments, when we adjust the learning rate linearly immediately with respect to the batch size. One difficulty we observe is that the noise in the stochastic momentum estimation is accumulated over time and will have delayed effects when the batch size changes. We therefore propose to smoothly adjust the learning rate over time to alleviate the influence of the noisy momentum estimation. Our experiments on image classification, object detection and semantic segmentation have demonstrated that our proposed Dynamic SGD method achieves stabilized performance when varying the number of GPUs from 8 to 128. We also provide theoretical understanding on the optimality of linear learning rate scheduling and the effects of stochastic momentum.

For agents

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

get_harvested_code_for_paper("1904.12043")
get_code_for_paper("1904.12043")
have("1904.12043")

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

curl https://syntology.ai/api/ran/1904.12043.json

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

[![Syntology run record](https://syntology.ai/api/ran/1904.12043.svg)](https://syntology.ai/paper/1904.12043)

Connect an agent — have() is free.