Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
We don’t allow questions seeking recommendations for books, tools, software libraries, and more. You can edit the question so it can be answered with facts and citations.
Closed yesterday.
Improve this question
I have some job experience in machine learning, specifically in the branch of computer vision. However, when I develop at work I use tools that abstract away the compute and storage via some cloud API.
Now, I want to develop a deep NN in my spare time and on my PC - I cannot use the work's resources and code.
What is the best setup, from your experience, for compute and storage I can get on a budget? To be specific, I am aiming to:
have ~1M samples of data - images of decent size, let's say 500X500.
try a variety of models, CNN and transformer architectures included.
Of course, I want training to be done in reasonable time (up to a day).
I can save checkpoints and resume if that helps with budget but preferably I will want to train one time each model.
I know some people use Google Colab which offers GPU access. What are the pros and cons and what alternatives are out there?
Related
Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
We don’t allow questions seeking recommendations for books, tools, software libraries, and more. You can edit the question so it can be answered with facts and citations.
Closed 6 years ago.
Improve this question
I am trying to do a paper about the Machine learning been applied in NLP. Can you guys please suggest me applications that have already used the Machine learning with the NLP?
The list is broad since machine learning is becoming more and more mainstream.
Regarding text, images and video, a good list of APIs would be:
AT&T Speech, IBM Watson, Google Prediction, Wit.ai, AlchemyAPI, Diffbot and I guess Project Oxford as well.
Hope it helps.
If you want something generic you can use this tutorial: http://www.cs.columbia.edu/~mcollins/papers/tutorial_colt.pdf
It is probably not the more recent information but you could find it useful if you start to learn ML methods for NLP.
As it is mentionned in this tutorial, ML methods are generally linked to the NLP task (Information Extraction, Machine Translation, etc.).
IBM Watson project is an example of platform that uses NLP and ML.
Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
We don’t allow questions seeking recommendations for books, tools, software libraries, and more. You can edit the question so it can be answered with facts and citations.
Closed 7 years ago.
Improve this question
I am working on a project wherein I am supposed to detect human beings from a live video stream which I get from a UAV's camera module. I do not need to create any rectangles or boxes around detected subjects, but just need to reply with a yes or no. I am fairly new to Open-CV and have no prior experience.
What I have tried:
I started by training my SVM on HOG features. My team gathered a few images from a UAV we had, with people in it. I then trained the SVM from the crops of those people. We got unsatisfactory results when we used the trained detector on the a video from sky with people. Moreover processing each frame turned out to be very slow , therefore the system became unusable.(it did work on still images to some extent).
My question:
I wanted to know if there is some other technique, library etc I could try for achieving good results. Please point me to the next step.
Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
We don’t allow questions seeking recommendations for books, tools, software libraries, and more. You can edit the question so it can be answered with facts and citations.
Closed 7 years ago.
Improve this question
Does anyone know of any courses etc for teaching people how to learn how to apply technical analysis and trading mechanics to the development of an automated trading algorithm?
I'm a regular at Quantopian, and attended the most recent Quantcon. They had some seminars, but largely it's a huge topic (like learning "surgery") because of the multiple disciplines involved.
Different languages, different levels of profeciency with those languages, different time frames, different securities, and a general air of secrecy where no one wants to share strategies.
For a programmer, I'd focus on API integrations (if you need that, some strategies run once a month then you punch in your trades manually). For a noob-programmer, I'd focus on programming skills in C#.
Sorry for being so broad, but like I said it's a huge topic. There's miles between a long gamma hedge fund doing managed futures and an HFT with custom hardware chips.
Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
We don’t allow questions seeking recommendations for books, tools, software libraries, and more. You can edit the question so it can be answered with facts and citations.
Closed 5 years ago.
Improve this question
Disclaimer: although I know some things about big data and am currently learning some other things about machine learning, the specific area that I wish to study is vague, or at least appears vague to me now. I'll do my best to describe it, but this question could still be categorised as too vague or not really a question. Hopefully, I'll be able to reword it more precisely once I get a reaction.
So,
I have some experience with Hadoop and the Hadoop stack (gained via using CDH), and I'm reading a book about Mahout, which is a collection of machine learning libraries. I also think I know enough statistics to be able to comprehend the math behind the machine learning algorithms, and I have some experience with R.
My ultimate goal is making a setup that would make trading predictions and deal with financial data in real time.
I wonder if there're any materials that I can further read to help me understand ways of managing that problem; books, video tutorials and exercises with example datasets are all welcome.
Take ML course on coursera. It is a good introductery into ML algorithms which will tell you what ML could do\some general approaches:
https://www.coursera.org/course/ml
Also to get a broader picture I suggest coursera's DataSciense course:
https://www.coursera.org/course/datasci
Finally a good book is Mahout in action - it is more about solving practical matters with mahout and has lots of examples and case-studies.
I beleive after that you will have a better understanding of what you want to do next.
Closed. This question does not meet Stack Overflow guidelines. It is not currently accepting answers.
We don’t allow questions seeking recommendations for books, tools, software libraries, and more. You can edit the question so it can be answered with facts and citations.
Closed 9 days ago.
Improve this question
I am badly looking for a small size (around 50 nodes) link prediction dataset, with node attributes (preferably real-valued attributes).
The network can be directed or undirected (preferably directed), but the network preferably not be bipartite.
Of course I am looking for a free, ready to use dataset.
Perhaps related to a social network, author communications, etc.
Does any one knows a specific dataset? I really really appreciate showing me one.
This Google+ social networks with node attributes should be good for you. You can find it here: https://gonglab.pratt.duke.edu/google-dataset
The 'Social Network Analysis Interactive Dataset Library' at http://www.growmeme.com/sna/visual has over 150 datasets, and an ability to filter by directed/undirected and bipartite = true/false and whether community information is present (among other things).