SP

keyu-tian/SparK

[ICLR'23 SpotlightšŸ”„] The first successful BERT/MAE-style pretraining on any convolutional network; Pytorch impl. of "Designing BERT for Convolutional Networks: Sparse and Hierarchical Masked Modeling"

1.4k 84 +0/wk
GitHub
bert cnn convnet convolutional-neural-networks deep-learning iclr iclr2023 instance-segmentation mae mask-rcnn masked-autoencoder masked-image-modeling
Trend 0

Star & Fork Trend (18 data points)

Stars
Forks

Multi-Source Signals

Growth Velocity

keyu-tian/SparK has +0 stars this period . Velocity data will be available after more historical data is collected.

Deep analysis is being generated for this repository.

Signal-backed technical analysis will be available soon.

Metric SparK cnn_graph ppl.nn TorchSSL
Stars 1.4k 1.4k1.4k1.4k
Forks 84 390221187
Weekly Growth +0 +0+0+0
Language Python Jupyter NotebookC++Python
Sources 1 111
License MIT MITApache-2.0MIT

Capability Radar vs cnn_graph

SparK
cnn_graph
Maintenance Activity 0

Last code push 807 days ago.

Community Engagement 31

Fork-to-star ratio: 6.1%. Lower fork ratio may indicate passive usage.

Issue Burden 70

Issue data not yet available.

Growth Momentum 30

No measurable growth in the current period (first-day cold start expected).

License Clarity 95

Licensed under MIT. Permissive — safe for commercial use.

Risk scores are computed from real-time repository data. Higher scores indicate healthier metrics.