SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1702.05373 · 2017

EMNIST: an extension of MNIST to handwritten letters

arXiv · PDF · Open in the Atlas

Code that ran

13 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 11 have no contract verdict; 0 failures 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/1702.05373.json · badge: /api/ran/1702.05373.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
Lornatang/TensorFlow-MNIST community (classifier label)
classifier basis not recorded on the link
9 ran · 1 no recorded run
Chizuchizu/amplify-hackathon community (classifier label)
classifier basis not recorded on the link
3 ran · 0 no recorded run
zmr1128/EMNIST reimplementation (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
FunctionStatusWhere it lives
AlexNet Ran
ran on a synthesized input
Lornatang/TensorFlow-MNIST/v0_2/models.py
code served (permissive licence) · get_code("083164bfbeb0d9ca")
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 Functional. 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
from tensorflow.python.keras import layers

ARGS = {
    'include_top': True,
    'input_tensor': None,
    'input_shape': (224, 224, 3),
    'pooling': None,
    'classes': 10,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Functional — a Functional output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Functional 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
AlexNet Ran
ran on a synthesized input
Lornatang/TensorFlow-MNIST/v0_1/models/alexnet.py
code served (permissive licence) · get_code("5aca9cbf66d5ea31")
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 Functional. 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
from tensorflow.python.keras import layers

ARGS = {
    'include_top': True,
    'input_tensor': None,
    'input_shape': (224, 224, 3),
    'pooling': None,
    'classes': 10
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Functional — a Functional output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Functional 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
CNN Ran
ran on a synthesized input
Lornatang/TensorFlow-MNIST/v0_2/models.py
code served (permissive licence) · get_code("640fbef861b1331b")
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 Functional. 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
from tensorflow.python.keras import backend
from tensorflow.python.keras import layers

# Construct valid arguments for the CNN function
ARGS = {
    'include_top': True,
    'input_tensor': None,
    'input_shape': (32, 32, 3),
    'pooling': None,
    'classes': 10
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Functional — a Functional output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Functional 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
LeNet Ran
ran on a synthesized input
Lornatang/TensorFlow-MNIST/v0_2/models.py
code served (permissive licence) · get_code("95aaaa71dd002c63")
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 Functional. 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
from tensorflow.python.keras import layers

# Test with include_top=True and classes != 1000
ARGS = {
    "include_top": True,
    "input_tensor": None,
    "input_shape": (32, 32, 1),
    "pooling": None,
    "classes": 10,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Functional — a Functional output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Functional 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
VGG16 Ran
ran on a synthesized input
Lornatang/TensorFlow-MNIST/v0_1/models/vgg16.py
code served (permissive licence) · get_code("ce11fac8cedd9a15")
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 Functional. It ran in syntology-harvest-sandbox:v5-deps (sha256:6a292aa1aed1) at 2026-09-26T21:51:12Z. 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
from tensorflow.python.keras import layers

ARGS = {
    "include_top": True,
    "input_tensor": None,
    "input_shape": (224, 224, 3),
    "pooling": None,
    "classes": 10,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Functional — a Functional output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Functional output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image syntology-harvest-sandbox:v5-deps (sha256:6a292aa1aed1); 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
VGG19 Ran
ran on a synthesized input
Lornatang/TensorFlow-MNIST/v0_1/models/vgg19.py
code served (permissive licence) · get_code("d3303b2b9b7cd348")
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 Functional. 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
from tensorflow.python.keras import layers

# Create a small valid input tensor for testing
img_input_tensor = layers.Input(shape=(224, 224, 3))

ARGS = {
    "include_top": True,
    "input_tensor": None,
    "input_shape": (224, 224, 3),
    "pooling": None,
    "classes": 10,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
tensorflow — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Functional — a Functional output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Functional 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
get_weight Ran
ran on a synthesized input
Chizuchizu/amplify-hackathon/src/utils.py
code served (permissive licence) · get_code("6eee3173de8e2ce0")
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

torch.manual_seed(0)

w = [1, 2, 3, 4]
k = 5

ARGS = {
    "w": w,
    "k": k
}
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
minmax Ran
ran on a synthesized input
Chizuchizu/amplify-hackathon/src/utils.py
code served (permissive licence) · get_code("49fda2edf096ba10")
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 (4,) 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 minmax takes a single argument x and computes min/max normalization
# It calls min(x), max(x), and performs arithmetic operations on x
# x should be a sequence or tensor that supports these operations

x = torch.tensor([1.0, 2.0, 5.0, 8.0])

ARGS = {
    "x": x
}
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 (4,) — shape (4,) 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
process_image Ran
ran on a synthesized input
Lornatang/TensorFlow-MNIST/v0_2/dataset.py
code served (permissive licence) · get_code("8f4ab24d39a5686c")
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 small image tensor (batch_size=2, height=4, width=3, channels=3)
# Values in range [0, 255] to simulate raw image data
image_data = np.random.randint(0, 256, size=(2, 4, 3, 3), dtype=np.uint8)
image = tf.convert_to_tensor(image_data, dtype=tf.float32)

# Create a label tensor (batch_size=2)
label = tf.convert_to_tensor([0, 1], dtype=tf.int64)

# Define height and width for resizing
height = 32
width = 32

ARGS = {
    "image": image,
    "label": label,
    "height": height,
    "width": width
}
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
process_image_with_kmnist Ran
ran on a synthesized input
Lornatang/TensorFlow-MNIST/v0_1/datasets/kmnist.py
code served (permissive licence) · get_code("32796d02041ad8bf")
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

# Create a small test image tensor (2, 4, 3) - batch of 2, height 4, width 3, 3 channels
image = tf.constant([
    [[[100, 150, 200], [120, 160, 210], [110, 155, 205]],
     [[105, 152, 202], [115, 162, 212], [108, 157, 207]],
     [[102, 151, 201], [112, 161, 211], [107, 156, 206]],
     [[103, 153, 203], [113, 163, 213], [109, 158, 208]]],
    [[[50, 100, 150], [60, 110, 160], [55, 105, 155]],
     [[52, 102, 152], [62, 112, 162], [57, 107, 157]],
     [[51, 101, 151], [61, 111, 161], [56, 106, 156]],
     [[53, 103, 153], [63, 113, 163], [58, 108, 158]]]
], dtype=tf.uint8)

# Create label tensor (2,) with int64 dtype
label = tf.constant([0, 1], dtype=tf.int64)

# Define height and width parameters
height = 32
width = 32

ARGS = {
    "image": image,
    "label": label,
    "height": height,
    "width": width
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
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
process_image_with_mnist Ran
ran on a synthesized input
Lornatang/TensorFlow-MNIST/v0_1/datasets/mnist.py
code served (permissive licence) · get_code("9cbd84255dff2fe5")
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

# Create a small test image tensor (batch_size=2, height=224, width=224, channels=3)
# Using uint8 values in range [0, 255] as typical for image data
image = tf.constant([
    [[[100, 150, 200], [120, 160, 210]], [[110, 155, 205], [125, 165, 215]]],
    [[[50, 100, 150], [60, 110, 160]], [[55, 105, 155], [65, 115, 165]]]
], dtype=tf.uint8)

# Create a small label tensor (batch_size=2)
label = tf.constant([3, 7], dtype=tf.int64)

ARGS = {
    "image": image,
    "label": label,
    "height": 32,
    "width": 32
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
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
rotate Ran
ran, honoured the contract we drafted
zmr1128/EMNIST/emnist.py
pointer only (licence: NONE) · get_code("c905d83a8c7c8cfe")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, zmr1128/EMNIST 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.

standardization Ran
ran, honoured the contract we drafted
Chizuchizu/amplify-hackathon/src/utils.py
code served (permissive licence) · get_code("2dc8b2021e28ec53")
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 2, dtype family float, shape (2, 4) 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 standardization takes a single parameter 'x'
# It calls x.mean(), x.std(), and performs arithmetic operations
# So x needs to support these methods - a torch tensor works perfectly

x = torch.randn(2, 4)

ARGS = {
    "x": x
}
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 2, dtype family float, shape (2, 4) — shape (2, 4) 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
process_image_with_emnist Not yet run
unverified (harvested, no recorded run)
Lornatang/TensorFlow-MNIST/v0_1/datasets/emnist.py
code served (permissive licence) · get_code("20087ccac441201c")
How it ran

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, 1 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 1 later) and 13 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
standardization 2 other papers
2 later
2205.13760 (2022-05)
Tranception: protein fitness prediction with autoregressive transformers and inference-time retrieval
2407.00002 (2024-07)
Kermut: Composite kernel regression for protein variant effects

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

More are linked than shown here.

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

The MNIST dataset has become a standard benchmark for learning, classification and computer vision systems. Contributing to its widespread adoption are the understandable and intuitive nature of the task, its relatively small size and storage requirements and the accessibility and ease-of-use of the database itself. The MNIST database was derived from a larger dataset known as the NIST Special Database 19 which contains digits, uppercase and lowercase handwritten letters. This paper introduces a variant of the full NIST dataset, which we have called Extended MNIST (EMNIST), which follows the same conversion paradigm used to create the MNIST dataset. The result is a set of datasets that constitute a more challenging classification tasks involving letters and digits, and that shares the same image structure and parameters as the original MNIST task, allowing for direct compatibility with all existing classifiers and systems. Benchmark results are presented along with a validation of the conversion process through the comparison of the classification results on converted NIST digits and the MNIST digits.

For agents

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

get_harvested_code_for_paper("1702.05373")
get_code_for_paper("1702.05373")
have("1702.05373")

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

curl https://syntology.ai/api/ran/1702.05373.json

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

[![Syntology run record](https://syntology.ai/api/ran/1702.05373.svg)](https://syntology.ai/paper/1702.05373)

Connect an agent — have() is free.