cifar10-simclr-a40

DIA-MVP/cifar10-simclr-a40

cifar10-simclr-a40 is a machine learning model from DIA-MVP. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.

A demo model from the Data & Impact Accounting (DIA) lab. It performs self-supervised contrastive pretraining (CIFAR-10) via from-scratch SimCLR encoder, with the base model trained from scratch, trained on NVIDIA A40.

Downloads · 30 days

Access

Public

Updated Jun 29, 2026

Repo size

44.8 MB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.pt44.8 MB · 100%

At a glance

License: apache-2.0
Access: Public
Created: Jun 29, 2026
Updated: Jun 29, 2026
SHA: 401798e2

Model card

From the Hugging Face model README

SimCLR ResNet-18 — CIFAR-10 (NVIDIA A40)

A demo model from the Data & Impact Accounting (DIA) lab. It performs self-supervised contrastive pretraining (CIFAR-10) via from-scratch SimCLR encoder, with the base model trained from scratch, trained on NVIDIA A40.

The point of this repo is not the model itself but its dia_report — a standardized record of the energy, carbon, and water used to train it, embedded in this card's metadata.

This footprint feeds the DIA dashboard, which rolls up a base model and all its derivatives to show the cumulative carbon, water, and energy cost of a model family.

Training footprint

Metric	Value
Hardware	1× NVIDIA A40
Compute	0.0663 GPU-hours
Energy	0.0137 (measured) kWh
Carbon	0.0009 (measured) kgCO₂eq
Water	0.025–0.055 (estimated-from-default-wue) L
Grid region	ca-on

Energy and carbon are measured with CodeCarbon; water is estimated from a default water-usage-effectiveness range. Carbon uses the local grid's intensity (Ontario, ~0.03 kgCO₂eq/kWh).

Reproduce

REPO=DIA-MVP/cifar10-simclr-a40 python scripts/train_simclr_cifar.py