SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2101.09868 · ICLR · 2021

CPT: Efficient Deep Neural Network Training via Cyclic Precision

Yingyan Lin, Yonggan Fu, Han Guo, Xin Yang, Yining Ding, Meng Li, Vikas Chandra

arXiv · PDF · Open in the Atlas

Code that ran

2 of 3 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 0 have no contract verdict; 2 failures of our instrument; 1 no recorded run), measured 2026-11-22; 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/2101.09868.json · badge: /api/ran/2101.09868.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
RICE-EIC/CPT marked official in Papers with Code
Papers with Code archive
2 ran · 1 no recorded run
FunctionStatusWhere it lives
_deflatten_as Ran
ran; our contract draft was wrong, not the code
RICE-EIC/CPT/cpt_cifar/modules/quantize.py
code served (permissive licence) · get_code("3fed6b12751d0395")
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 4, dtype family float, shape (2, 4, 1, 1) 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)

# Analyze the function:
# - _deflatten_as(x, x_full) takes two tensors
# - It uses x.shape, x.dim(), x_full.dim(), and x.view()
# - It reshapes x to match the dimensionality of x_full by adding singleton dimensions

# Create two tensors with different dimensionalities
# x should have fewer dimensions than x_full
x = torch.randn(2, 4)  # 2D tensor
x_full = torch.randn(2, 4, 3, 5)  # 4D tensor

ARGS = {
    "x": x,
    "x_full": x_full,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, ndim 4, dtype family float, shape (2, 4, 1, 1) — shape (2, 4, 1, 1) 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
calculate_qparams Ran
ran; our contract draft was wrong, not the code
RICE-EIC/CPT/cpt_cifar/modules/quantize.py
code served (permissive licence) · get_code("750c588ade0c659c")
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 QParams. 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

torch.manual_seed(0)

# Create a small tensor for quantization parameter calculation
# Shape: (2, 4, 3) - batch_size=2, channels=4, spatial=3
x = torch.randn(2, 4, 3)

# Number of bits for quantization
num_bits = 8

# Flatten dimensions (default behavior)
flatten_dims = (1, -1)

# Reduction dimension
reduce_dim = 0

# Reduction type
reduce_type = 'mean'

# Keep dimension flag
keepdim = False

# True zero flag
true_zero = False

ARGS = {
    'x': x,
    'num_bits': num_bits,
    'flatten_dims': flatten_dims,
    'reduce_dim': reduce_dim,
    'reduce_type': reduce_type,
    'keepdim': keepdim,
    'true_zero': true_zero,
}
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 QParams — a QParams 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
UniformQuantize Not yet run
unverified (harvested, no recorded run)
RICE-EIC/CPT/cpt_cifar/modules/quantize.py
code served (permissive licence) · get_code("3b12a9ff5ac11191")
How it ran

no invocation recorded.

Also present in other papers' code

Of 3 samples lifted from this paper's linked repositories, 3 carry a function identity and 0 do not. Of the 3 distinct functions with an identity, 2 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 1 later) and 1 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
calculate_qparams
brick of this paper's assembly: helper
3 other papers
1 dated earlier, 2 later · 1 only as an assembly brick
2012.13113 (2020-12)
FracTrain: Fractionally Squeezing Bit Savings Both Temporally and Spatially for Efficient DNN Training
2112.11594 (2021-12)
GCoD: Graph Convolutional Network Acceleration via Dedicated Algorithm and Accelerator Co-Design
2211.05109 (2022-11)
ViTALiTy: Unifying Low-rank and Sparse Approximation for Vision Transformer Acceleration with a Linear Taylor Attention
_deflatten_as
brick of this paper's assembly: helper
1 other paper
1 dated earlier · 1 only as an assembly brick
2012.13113 (2020-12)
FracTrain: Fractionally Squeezing Bit Savings Both Temporally and Spatially for Efficient DNN Training

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

Low-precision deep neural network (DNN) training has gained tremendous attention as reducing precision is one of the most effective knobs for boosting DNNs' training time/energy efficiency. In this paper, we attempt to explore low-precision training from a new perspective as inspired by recent findings in understanding DNN training: we conjecture that DNNs' precision might have a similar effect as the learning rate during DNN training, and advocate dynamic precision along the training trajectory for further boosting the time/energy efficiency of DNN training. Specifically, we propose Cyclic Precision Training (CPT) to cyclically vary the precision between two boundary values which can be identified using a simple precision range test within the first few training epochs. Extensive simulations and ablation studies on five datasets and eleven models demonstrate that CPT's effectiveness is consistent across various models/tasks (including classification and language modeling). Furthermore, through experiments and visualization we show that CPT helps to (1) converge to a wider minima with a lower generalization error and (2) reduce training variance which we believe opens up a new design knob for simultaneously improving the optimization and efficiency of DNN training. Our codes are available at: https://github.com/RICE-EIC/CPT.

For agents

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

get_harvested_code_for_paper("2101.09868")
get_code_for_paper("2101.09868")
have("2101.09868")

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

curl https://syntology.ai/api/ran/2101.09868.json

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

[![Syntology run record](https://syntology.ai/api/ran/2101.09868.svg)](https://syntology.ai/paper/2101.09868)

Connect an agent — have() is free.