Methods › Computer Vision › Image Model Blocks › Fire Module

Fire Module

95 papers tagged archive 2025-07-28

archive 2025-07-28 Description, source and code snippet are the archive's method entry.

A Fire Module is a building block for convolutional neural networks, notably used as part of SqueezeNet. A Fire module is comprised of: a squeeze convolution layer (which has only 1x1 filters), feeding into an expand layer that has a mix of 1x1 and 3x3 convolution filters. We expose three tunable dimensions (hyperparameters) in a Fire module: s₁ₓ₁, e₁ₓ₁, and e₃ₓ₃. In a Fire module, s₁ₓ₁ is the number of filters in the squeeze layer (all 1x1), e₁ₓ₁ is the number of 1x1 filters in the expand layer, and e₃ₓ₃ is the number of 3x3 filters in the expand layer. When we use Fire modules we set s₁ₓ₁ to be less than (e₁ₓ₁ + e₃ₓ₃), so the squeeze layer helps to limit the number of input channels to the 3x3 filters.

Source: SqueezeNet: AlexNet-level accuracy with 50x fewer...See Code · pytorch/vision

Papers archive 2025-07-28

30 shown of 95, newest first. Repository counts are the archive's code-links table. A Syntology line states what Syntology ran from that paper's harvested code; it is per sample and not a correctness claim.

Tasks archive 2025-07-28

20 shown of 90 tasks the archive attaches to papers tagged with this method, by distinct papers. A task without a page in the catalog is plain text.

TaskPapers
Transfer Learning16
General Classification11
Object Detection11
object-detection11
Image Classification10
Classification8
GPU7
image-classification7
Data Augmentation6
Deep Learning5
Face Recognition4
Object4
Quantization4
Specificity4
CPU3
Diagnostic3
Face Verification3
Network Pruning3
Object Recognition3
Semantic Segmentation3

Usage over time archive 2025-07-28

Papers per year tagged with Fire Module: 2016 to 2025, peak 18 18 0 2016: 4 papers 2016 2017: 8 papers 2017 2018: 18 papers 2018 2019: 10 papers 2019 2020: 17 papers 2020 2021: 15 papers 2021 2022: 6 papers 2022 2023: 6 papers 2023 2024: 8 papers 2024 2025: 3 papers 2025
Papers per year the archive tags with this method, by the paper's archive date (95 dated). Bars are counts, not a trend claim.

Components: the archive holds no method-to-method composition, so PwC's Components table cannot be rebuilt; the Papers list carries no Results column for the same reason (the archive does not join its leaderboard rows to method tags).

Categories archive 2025-07-28

Image Model Blocks

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections