Show newer

@freeschool I don't think so. Yolo is the core method to detect birds in a video, and my laptop's 3070 is too weak. A powerful GPU instance will cost money, but will be much faster...

OK... So now basically every VPS provider requires additional verification in order to use their GPU instance.

Just start doing it with AWS. It said my account needed to be verified before I can create GPU instances. Then I got my account verified. After that they said I don't have the quota to create any GPU instance. Now I'm working on increasing that f*ck quota.

I don't know why. After all this processing, I still need to pay for my bill.

我现在感觉模型训练这东西秘密在于一堆技巧和经验。它可能很难算作一个科学,有时候paper和实际差距深远,怎么补?靠经验。所以再过一些年模型自己训练自己可能比人训的好,因为经验也就那些(((

现在唯一的障碍是这种经验要烧h100才能获得。所以我也开始理解为什么这一行人工甚高,因为你要烧掉h100才能获得经验,而h100 hours烧起来比一般工程师工资高多了,那么已经烧过的人确实值高一些的工资。

:ablobnervous:

天空вℓσи∂  
I switched to yolov8 L instead of X and I got 4.5 FPS instead of 3 FPS. My current algorithm is to move the clip to the boring folder if there is n...

I switched to yolov8 L instead of X and I got 4.5 FPS instead of 3 FPS.

My current algorithm is to move the clip to the boring folder if there is no bird detected. If the model says there is a bird, then move to the bird folder.

I checked the results, the boring folder is pretty good. very video file is boring.

The bird folder, however, gives 147 results, but 33 of them are nothing. Maybe it detects some shadow of something. Maybe I should count how many frames have birds and set a minimum requirement before moving to the bird folder.

昨天和朋友去了山里看了小溪小瀑布,之后去附近的可爱小镇野餐闲逛。回家路上顺路停下逛了两家二手店,快到城里车太堵就顺势右拐去朋友家吃了Costco的冷冻chicken tikka masala,玩了overcooked。
做失业游民就是这么痛并快乐着。

Well... Another problem: my RTX3070 mobile is too weak to handle the workload. It's overheating and can only process 3 frames per second, which I have already discarded 75% of the frames.

I probably need to rent a gpu server from vultr or something.

I mean, the heavy lift is done by onnx and yolo, I just invoke them and copy files based on the result.

But still, it's exciting to see my code actually working. BTW I'm using kotlin/jvm. So no python nonsense.

YOLOv8 detect birds in video clips.

I think the main issue of this toot is I use the default 640x480 resolution, which makes the picture too small to distinguish the details. Running on 1080P natively does cost a lot of RAM and time, but it can tell it's a bird very quickly and confidently.

The best result I can get with my laptop RTX 3070 8GB is 1.85 FPS, processing 1920x1088 videos, yolov8x

Maybe I should try the smaller model?

Tonight I'm going to stay up late writing code and squeezing the last bit of performance out of my poor graphics card as much as I can. I DONT need to go to bed early, I DONT need to live a healthy life, I NEED to satisfy my curiosity, and I want to see my code run correctly. Healthy living is like God and angels, they only come out when you're dead; and my curiosity, it's the real Mephistopheles. He brings me the true satisfaction and joy.

*that me desperately trying to make my code work*

I recently changed to openSUSE tumbleweed on my gamin PC after using EndeavourOS for a year.
I can say that Tumbleweed works much better than Endeavour, I've had no problems running any games under proton compared to when I used Endeavor.

Good to see my first linux distro still being the best

Please reboost!

Trying something new, everyone is guaranteed an interview! Open interviews! For a limited time no one will be skipped (except for clear cases of abuse).

So we still have about 10 more 100% remote positions to hire for full-time market-fair positions here at QOTO/CleverThis.

100% remote, work from anywhere, even the beach, market-fair offers. Ethics first, we treat our people like family.

We have an urgent need for Machine learning experts with a background in NLP and Deep Learning (Natural Language Processing and Neural Networks). There is a focus on Knowledge Graphs, Mathematics, Java, C, looking for Polyglots.

We are an open-source first company, we give back heavily to the OSS community.

We need everything from jr to sr, data scientist to programmer. If your IT and your good, you might be a fit.

I will personally be both your direct boss, and hiring manager. I am also the founder and inventor.

The NLP position can be found at this link, other positions can be found on the menu bar on the left:

docs.cleverthis.com/en/human_r

If you would like to submit yourself for an interview, which for a limited time I am guaranteeing you will get a first stage interview, then you can submit your application here, and even schedule your interview as you apply, instantly!

cal.com/team/cleverthis/interv

For those of you who cant schedule during core hours you can schedule in my free time if you’d like a chance (the company doesnt have fixed hours):

cal.com/team/cleverthis/interv

I2P 2.5.2 has been released. It includes fixing a bug introduced in 2.5.0 causing truncation of some HTTP content, also an update rrd4j to 3.9.1-preview in the console, and a bug fix for tunnel selection of peers with expired RIs:
geti2p.net/en/blog/post/2024/0

Show older
Qoto Mastodon

QOTO: Question Others to Teach Ourselves
An inclusive, Academic Freedom, instance
All cultures welcome.
Hate speech and harassment strictly forbidden.