Dropout: SVHN

Closed weights University of Toronto 47.8M parameters June 2014

No estimate

No hardware requirements for this model

The weights for this model have not been published, so it cannot be downloaded or run on your own hardware at any size. It is reachable only through its provider, and no graphics card changes that.

On record

Full specification

Everything on record for this model. Most of it describes how it was trained rather than how it runs — useful context for judging how much work went into it, and how it compares with models built at a different scale.

Origin

Who built this model, where, and when it was published.

Organisation
University of Toronto
Organisation type
Academia
Country
Canada
Published
1 June 2014
Authors
Nitish Shrivasta, Geoffrey Hinton, Alex Krizhevsky, Ilya Sutskever, Ruslan Salakhutdinov

What it does

The problem areas the model was built for. A model can carry several of each.

Domain
Vision
Task
Image classification, Digit recognition

Size

How large the model is and how much data it was trained on. Parameters are the figure that decides whether it fits on a given graphics card.

Parameters
47.8M

"The best architecture that we found uses three convolutional layers each followed by a max-pooling layer. The convolutional layers have 96, 128 and 256 filters respectively. Each convolutional layer has a 5 × 5 receptive field applied with a stride of 1 pixel. Each max pooling layer pools 3 × 3 regions at strides of 2 pixels. The convolutional layers are followed by two fully connected hidden layers having 2048 units each." Inputs: 32 x 32 x 3 conv_1: 96 x 5 x 5 x 3 = 7,200 conv_2: 128 x 5 x 5 …

Training data
600,000 tokens

Appendix B.2: "The SVHN data set consists of approximately 600,000 training images and 26,000 test images" dimensionality 3072 (32 × 32 color)

Availability

Whether you can obtain the model and run it on your own hardware, which is what decides if any of the graphics-card figures on this page apply.

Weights
Closed — provider access only
Model access
Unreleased
Training code
Open (non-commercial)

http://www.cs.toronto.edu/~nitish/dropout see model files

How it is classified

Labels the source dataset applies when tracking notable models, and how confident it is in the entry.

Why it is tracked
Highly cited,SOTA improvement

SOTA on the Street View House Numbers dataset

Record confidence
Confident
Citations
41,962

Sources

Where this record came from and when it was last checked.

Reference
Dropout: A Simple Way to Prevent Neural Networks from Overfitting
Last updated
1 January 2026

What the numbers mean

About this model

Dropout: SVHN was published by University of Toronto, in the country recorded as Canada, during June 2014. It comes out of an organisation categorised as academia.

It works in the domain of Vision, and is recorded as performing the task of image classification, Digit recognition.

This is a closed model: the trained values stayed with whoever produced them, and there is no local version to run.

What went into building it

The training set ran to roughly 600,000 tokens of text.

It is tracked in the underlying dataset for one reason in particular: highly cited,SOTA improvement.

Answers

Dropout: SVHN — common questions

01

Dropout: SVHN— is it open source?

No. Its weights have not been published, so it exists only as a service controlled by its owner.

02

Dropout: SVHN— how many parameters does it have?

It has a parameter count of 47.8M. "The best architecture that we found uses three convolutional layers each followed by a max-pooling layer. The convolutional layers have 96, 128 and 256 filters respectively. Each convolutional layer has a 5 × 5 receptive field applied with a stride of 1 pixel. Each max pooling layer pools 3 × 3 regions at strides of 2 pixels. The convolutional layers are followed by two fully connected hidden layers having 2048 units each." Inputs: 32 x 32 x 3 conv_1: 96 x 5 x 5 x 3 = 7,200 conv_2: 128 x 5 x 5 x 96 = 307,200 conv_3: 256 x 5 x 5 x 128 = 819,200 (output shape after CNN 3 will be: 20 x 20 x 256) max_pool has no learnable parameters but further reduces output shape to 9 x 9 x 256 FFN_1: (9 x 9 x 256) x 2048 = 42,467,328 FFN_2: 2048 x 2048 = 4,194,304 So total number of parameters appear to be 47,795,232. That figure is the total, and it is what decides how much memory the model needs — roughly half a gigabyte per billion at the compression most people use.

03

Dropout: SVHN— who created it?

It was published by University of Toronto, based in Canada, an organisation categorised as academia.

04

Dropout: SVHN— when was it released?

It was published in June 2014. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.

05

Dropout: SVHN— what is it used for?

It works in the domain of Vision, and is recorded as handling the task of image classification, Digit recognition. A model can carry several of each, so these are the areas it was built for rather than a limit on what it will attempt.

06

Dropout: SVHN— what GPU do I need to run it?

None. This is a closed model — its weights were never published, so it cannot be downloaded or run on your own hardware at any price. It is reachable only through its provider.

Source

Original publication

Record last updated 1 January 2026

The other direction

Looking at it from the other side?

This page starts from the model. If you already own a card and want to know everything it will run, start from the hardware instead.