CrossStagePartialNetworks

Cross Stage Partial Networks

Systems & Infrastructure Networking

Explore on GitHub Read Hacker News discussions

2Distinct people

3Mentions

0Explicit recommenders

914GitHub stars

REPOSITORY AT A GLANCE

What to know before exploring

What it is

Cross Stage Partial Networks

GitHub

Where it is used

No documented product use has been verified yet.

What people are saying

No recommendation phrase was found in the indexed comments.

No classified warning is attached to this page. Read the discussions before treating silence as approval.

Also in these discussions

No other repository is repeatedly linked in the same discussions.

This week on Hacker News

Nov 7–Nov 14 UTC · Direct repository links and separately labeled related project stories. Indexed through 11/14/2026, 3:56:07 PM UTC.

No direct repository links were found in the indexed HN discussions over the past seven days. Older discussions are available below.

Mentions over time

HN mentions by hour3 dated mentions

Observed mentions: 2020-06-10 20:32:21 UTC to 2020-06-11 09:47:10 UTC.

Counts come from the repository’s dated source evidence. Index cutoff: 2026-11-02 08:16:59 UTC.

View mention counts and date ranges

UTC ranges include their start and exclude their end. Select a period to filter the source evidence.
Period	Start (UTC)	End (exclusive, UTC)	Mentions	Comments
Jun 10, 20:00	2020-06-10 20:00:00 UTC	2020-06-10 21:00:00 UTC	1	1
Jun 10, 21:00	2020-06-10 21:00:00 UTC	2020-06-10 22:00:00 UTC	1	1
Jun 10, 22:00	2020-06-10 22:00:00 UTC	2020-06-10 23:00:00 UTC	0	0
Jun 10, 23:00	2020-06-10 23:00:00 UTC	2020-06-11 00:00:00 UTC	0	0
Jun 11, 00:00	2020-06-11 00:00:00 UTC	2020-06-11 01:00:00 UTC	0	0
Jun 11, 01:00	2020-06-11 01:00:00 UTC	2020-06-11 02:00:00 UTC	0	0
Jun 11, 02:00	2020-06-11 02:00:00 UTC	2020-06-11 03:00:00 UTC	0	0
Jun 11, 03:00	2020-06-11 03:00:00 UTC	2020-06-11 04:00:00 UTC	0	0
Jun 11, 04:00	2020-06-11 04:00:00 UTC	2020-06-11 05:00:00 UTC	0	0
Jun 11, 05:00	2020-06-11 05:00:00 UTC	2020-06-11 06:00:00 UTC	0	0
Jun 11, 06:00	2020-06-11 06:00:00 UTC	2020-06-11 07:00:00 UTC	0	0
Jun 11, 07:00	2020-06-11 07:00:00 UTC	2020-06-11 08:00:00 UTC	0	0
Jun 11, 08:00	2020-06-11 08:00:00 UTC	2020-06-11 09:00:00 UTC	0	0
Jun 11, 09:00	2020-06-11 09:00:00 UTC	2020-06-11 10:00:00 UTC	1	1

What Hacker News says

3 sources

anthiras·Jun 11, 2020

It's about time for Roboflow to pull this article. It seems highly unlikely that a 90 % smaller model would provide a similar accuracy, and the result seems to come from a small custom dataset only. Please make a real COCO comparison instead. The YoloV5 repo itself shows performance comparable to YoloV3: https://github.com/ultralytics/yolov5#pretrained-checkpoints Another comparison suggests YoloV5 is slightly WORSE than YoloV4: https://github.com/WongKinYiu/CrossStagePartialNetworks/issu...

In YOLOv5: State-of-the-art object detection at 140 FPS

sillysaurusx·Jun 10, 2020

YOLOv5 seems to have one important advantage over v4, which your post helped highlight: Fourth, YOLOv5 is small. Specifically, a weights file for YOLOv5 is 27 megabytes. Our weights file for YOLOv4 (with Darknet architecture) is 244 megabytes. YOLOv5 is nearly 90 percent smaller than YOLOv4. This means YOLOv5 can be deployed to embedded devices much more easily. Naming controversy aside, it's nice to have some model that can get close to the same accuracy at 10% of the size. Naming it v5 was certainly ... bold ... though. If it can't outperform v4 in any scenario, is it really worthy of the name? (On the other hand, if v5 can beat v4 in inference time or accuracy, that should be highlighted somewhere.) FWIW I doubt anyone who looks into this will think roboflow had anything to do with the current controversies. You just showed off what so

Read the rest

meone else made, which is both legit and helpful. It's not like you were the ones that named it v5. On the other hand... visiting https://models.roboflow.ai/ does show YOLOv5 as "current SOTA", with some impressive-sounding results: SIZE: YOLOv5 is about 88% smaller than YOLOv4 (27 MB vs 244 MB) SPEED: YOLOv5 is about 180% faster than YOLOv4 (140 FPS vs 50 FPS) ACCURACY: YOLOv5 is roughly as accurate as YOLOv4 on the same task (0.895 mAP vs 0.892 mAP) Then it links to https://blog.roboflow.ai/yolov5-is-here/ but there doesn't seem to be any clear chart showing "here's v5 performance vs v4 performance under these conditions: x, y, z" Out of curiosity, where did the "180% faster" and 0.895 mAP vs 0.892 mAP numbers come from? Is there some way to reproduce those measurements? The benchmarks at https://github.com/WongKinYiu/CrossStagePartialNetworks/issu... seem to show different results, with v4 coming out ahead in both accuracy and speed at 736x736 res. I'm not sure if they're using a standard benchmarking script though. Thanks for gathering together what's currently known. The field does move fast.

In YOLOv5: State-of-the-art object detection at 140 FPS

sillysaurusx·Jun 10, 2020

There are some benchmarks here: https://github.com/WongKinYiu/CrossStagePartialNetworks/issu... It's hard to interpret benchmarks in a fair way, but it's sort of sounding like YOLOv4 might be superior to YOLOv5, at least for certain resolutions. Does YOLOv5 outperform YOLOv4 at all? Faster inference time or higher accuracy?

In YOLOv5: State-of-the-art object detection at 140 FPS