Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Dive into Deep Learning

d2l.ai

Post reply on HN
Print view View original

21–30 of 94 posts

Re: Dive into Deep Learning

#21

Post by fareesh »

meritt wrote: post #13
fareesh wrote: post #8

As an engineer I find myself in this type of situation quite often - if anyone can point me to some good resources or has any advice, I'd be quite grateful: - Some non-technical stakeholder comes to me and says "can we solve this problem with Machine Learning?" usually it's something like "there need to be two supervisors on the factory floor at all times, and I want an email alert everytime there are less than 2 sup…

Why is this a machine learning problem? Does your factory not have keycard access? Or just require your supervisors to carry some sort of RFID/BLE tracking device. These are well-solved problems.

I had to lie about specifics in the example because I post under my real name and there are things I can't talk about :)

Apologies - I figured the primary intent of my comment - i.e. the questions at the end, would be the focus of most responses

Re: Dive into Deep Learning

#22

Post by voodootrucker »

fareesh wrote: post #8

As an engineer I find myself in this type of situation quite often - if anyone can point me to some good resources or has any advice, I'd be quite grateful: - Some non-technical stakeholder comes to me and says "can we solve this problem with Machine Learning?" usually it's something like "there need to be two supervisors on the factory floor at all times, and I want an email alert everytime there are less than 2 sup…

- We use GCP for labeling [1]

- Yolov3 is state of the art for speed. I think RetinaNet does better if you have the horse power.

- I can't recommend FastAI [2] enough for learning things to try.

- 60% on a frame by frame basis might be enough as long as you have a low false positive rate you can tell. Combine with OpenCV mean shift if you need real time.

- Start small. Show success with pre-trained models, then move on to transfer learning. Start with a small dataset. Agree on a metric beforehand.

- Use a notebook. [3] Play around, don't let it run for days then look at the result.

[1] https://cloud.google.com/ai-platform/data-labeling/docs/

[2] https://course.fast.ai/

[3] https://github.com/Mersive-Technologies/yolov3/blob/master/f...

Edit: formatting

Re: Dive into Deep Learning

#23

Post by throwlaplace »

fxtentacle wrote: post #15
fareesh wrote: post #8

As an engineer I find myself in this type of situation quite often - if anyone can point me to some good resources or has any advice, I'd be quite grateful: - Some non-technical stakeholder comes to me and says "can we solve this problem with Machine Learning?" usually it's something like "there need to be two supervisors on the factory floor at all times, and I want an email alert everytime there are less than 2 sup…

Most AI stuff is just horribly over-hyped, so the sad truth might be that what you are seeing is the state of the art and nobody else has found a better way yet. As a practical example, figuring out where a given pixel moves from one video frame to the next one, when working on real-world videos, the best known algorithms get about 50% of the pixels correct. With clever filtering, you can maybe bump that to 60 or 70%…

>As a practical example, figuring out where a given pixel moves from one video frame to the next one, when working on real-world videos, the best known algorithms get about 50% of the pixels correct. With clever filtering, you can maybe bump that to 60 or 70%, but in any case you will be left with a 30%+ error rate.

what do you mean by this? optical flow isn't really a learning problem? it's a classical problem with very good classical algorithms

https://www.mia.uni-saarland.de/Publications/brox-eccv04-of....

https://people.csail.mit.edu/celiu/OpticalFlow/

https://github.com/pathak22/pyflow

Re: Dive into Deep Learning

#24

Post by voodootrucker »

fareesh wrote: post #18
sickcodebruh wrote: post #16

Earlier quoted context omitted.

Have you tried fast.ai's Practical Deep Learning For Coders? https://course.fast.ai/ I think it's great for answering many of the exact questions you have. I was able to answer my own versions of many of those questions after the first few video lessons. It demonstrated to us that our data is a great fit for machine learning. I didn't feel comfortable turning my experiments into something production-worthy but I feel…

There seem to be a lot of courses in this space - I'll give this one a try since you're recommending it. Most of them seem to focus more on the theoretical / math aspects of stuff, which is quite interesting but I find it more interesting to implement these things and solve real-world problems.

FastAI has you detecting dog breads in lesson one :)

Re: Dive into Deep Learning

#25

Post by fareesh »

voodootrucker wrote: post #22
fareesh wrote: post #8

As an engineer I find myself in this type of situation quite often - if anyone can point me to some good resources or has any advice, I'd be quite grateful: - Some non-technical stakeholder comes to me and says "can we solve this problem with Machine Learning?" usually it's something like "there need to be two supervisors on the factory floor at all times, and I want an email alert everytime there are less than 2 sup…

- We use GCP for labeling [1] - Yolov3 is state of the art for speed. I think RetinaNet does better if you have the horse power. - I can't recommend FastAI [2] enough for learning things to try. - 60% on a frame by frame basis might be enough as long as you have a low false positive rate you can tell. Combine with OpenCV mean shift if you need real time. - Start small. Show success with pre-trained models, then move…

Thanks I will check out these resources

Re: Dive into Deep Learning

#26

Post by fareesh »

zmmmmm wrote: post #20
fareesh wrote: post #8

As an engineer I find myself in this type of situation quite often - if anyone can point me to some good resources or has any advice, I'd be quite grateful: - Some non-technical stakeholder comes to me and says "can we solve this problem with Machine Learning?" usually it's something like "there need to be two supervisors on the factory floor at all times, and I want an email alert everytime there are less than 2 sup…

The shocking thing that at least I ran into is the sheer quantity of training data you really need. The large companies doing this successfully are using utterly gigantic libraries of training data that are beyond anything others could ever come up with. It really brought home to me what a blunt intstrument deep learning really is.

Is there some kind of rule of thumb for a minimum of how much data is needed for various types of problems?

Re: Dive into Deep Learning

#27

Post by fxtentacle »

throwlaplace wrote: post #23
fxtentacle wrote: post #15

Earlier quoted context omitted.

Most AI stuff is just horribly over-hyped, so the sad truth might be that what you are seeing is the state of the art and nobody else has found a better way yet. As a practical example, figuring out where a given pixel moves from one video frame to the next one, when working on real-world videos, the best known algorithms get about 50% of the pixels correct. With clever filtering, you can maybe bump that to 60 or 70%…

>As a practical example, figuring out where a given pixel moves from one video frame to the next one, when working on real-world videos, the best known algorithms get about 50% of the pixels correct. With clever filtering, you can maybe bump that to 60 or 70%, but in any case you will be left with a 30%+ error rate. what do you mean by this? optical flow isn't really a learning problem? it's a classical problem with…

It used to be. Then the AI fanboys arrived and started treating it like a learning problem.

https://arxiv.org/abs/1612.01925

https://arxiv.org/abs/1709.02371

https://arxiv.org/abs/1904.09117

BTW, also the classical algorithms deal very badly with noise and repetitive textures, e.g. a video of a forest in the afternoon.

Re: Dive into Deep Learning

#28

Post by newfeatureok »

fareesh wrote: post #8

As an engineer I find myself in this type of situation quite often - if anyone can point me to some good resources or has any advice, I'd be quite grateful: - Some non-technical stakeholder comes to me and says "can we solve this problem with Machine Learning?" usually it's something like "there need to be two supervisors on the factory floor at all times, and I want an email alert everytime there are less than 2 sup…

I'm somewhat surprised at the responses for this.

I believe your issue can be easily solved - have supervisors wear a distinctive color from a non-supervisor. For example let's say it's yellow.

OK so now you have yellow wearing supervisors and everyone else. To resolve the issue you have described acquire a month or so of footage, with labels per minute describing how many yellow wearing supervisors and how many people (in total) there are.

So the data you have is:

1. Yellow wearing supervisors

2. Total amount of workers on the floor

Then with this data you can train a network to do what you're describing pretty easily. Assuming there are a lot of workers on the floor, trying to do person detection or face detection would require too much data. Just have a uniform enforced and train on the colors/presence.

Re: Dive into Deep Learning

#29

Post by whoisnnamdi »

Great guide - though unless I missed it I think this is missing the latest advancements around Transformers, BERT, ELMo, etc.

This stuff is pretty fresh, so it's understandable, but the NLP chapter would be greatly enhanced by covering these newer topics

Re: Dive into Deep Learning

#30

Post by fareesh »

newfeatureok wrote: post #28
fareesh wrote: post #8

As an engineer I find myself in this type of situation quite often - if anyone can point me to some good resources or has any advice, I'd be quite grateful: - Some non-technical stakeholder comes to me and says "can we solve this problem with Machine Learning?" usually it's something like "there need to be two supervisors on the factory floor at all times, and I want an email alert everytime there are less than 2 sup…

I'm somewhat surprised at the responses for this. I believe your issue can be easily solved - have supervisors wear a distinctive color from a non-supervisor. For example let's say it's yellow. OK so now you have yellow wearing supervisors and everyone else. To resolve the issue you have described acquire a month or so of footage, with labels per minute describing how many yellow wearing supervisors and how many peop…

Sorry but it was a scenario I imagined and not something that happened in reality. I can't talk about some of the real-world scenarios that I am asked to consult on, so I made up a rather poorly thought-out one.
Post reply on HN