I am curios if there are examples on how to run TF-Slim models/slim using deployment/model_deploy.py across multiple GPU’s on multiple machines. The documentation is pretty good but I am missing a couple of pieces. Specifically what needs to be put in for worker_device and ps_device and what additionally needs to be run on each machine?
An example like the one at the bottom of the distributed page would be awesome.
https://www.tensorflow.org/how_tos/distributed/
Related
I would like to use PyTextRank for keyphrase extraction. How can I feed feed 5 million documents (each document consisting of a few paragraphs) to the package?
This is the example I see on the official tutorial.
text = "Compatibility of systems of linear constraints over the set of natural numbers. Criteria of compatibility of a system of linear Diophantine equations, strict inequations, and nonstrict inequations are considered. Upper bounds for components of a minimal set of solutions and algorithms of construction of minimal generating sets of solutions for all types of systems are given. These criteria and the corresponding algorithms for constructing a minimal supporting set of solutions can be used in solving all the considered types systems and systems of mixed types.\n"
doc = nlp(text)
for phrase in doc._.phrases:
ic(phrase.rank, phrase.count, phrase.text)
ic(phrase.chunks)
Is my option only to concatenate several million documents to a single string and pass it to nlp(text)? I do not think I could use nlp.pipe(texts) as I want to create one network by computing words/phrases from all documents.
No, instead it would almost certainly be better to run these tasks in parallel. Many use cases of pytextrank have used Spark, Dask, Ray, etc., to parallelize running documents through a spaCy pipeline with pytestrank to extract entities.
For an example of parallelization with Ray, see https://github.com/Coleridge-Initiative/rclc/blob/4d5347d8d1ac2693901966d6dd6905ba14133f89/bin/index_phrases.py#L45
One question would be how you are associating the extracted entities with documents? Are these being collected into a dataset, or perhaps a database or key/value store?
However these results get collected, you could then construct a graph of co-occurring phrases, and also include additional semantics to help structure the results. A sister project kglab https://github.com/DerwenAI/kglab was created for these kinds of use cases. There are some examples in the Jupyter notebooks included with the kglab project; see https://derwen.ai/docs/kgl/tutorial/
FWIW, we'll have tutorials coming up at ODSC West about using kglab and pytextrank and there are several videos online (under Graph Data Science) for previous tutorials at conferences. We also have monthly public office hours through https://www.knowledgegraph.tech/ – message me #pacoid on Tw for details.
When OptaPlanner is used in web service, which means an OptaPlanner app is required to solve multiple problems in parallel threads, are there any limitations to prevent OptaPlanner from doing this? Is synchronization required in any OptaPlanner functions? Thanks
OptaPlanner supports this: it's a common use case.
In a single JVM, look at SolverManager to solve mulitple datasets of the same use case in parallel. Even if the constraint weights differ per dataset (see ConstraintConfiguration). So even if some datasets disable/enable some of the constraints while others don't.
For different use cases in a single JVM, just create multiple SolverFactory or SolverManager instances. This is uncommon because usually each use case is a different app (= microservice?).
Across multiple JVMs (= pods), there are several good techniques. Our activemq quickstart scales beautifully horizontally. Read Radovan's blog about how the ActiveMQ is used to load balance the work across the solver pods.
I have hundreds of models, based on categories, projects,s, etc. Some of the models are heavily used while other models are not used very frequently.
How can I trigger a scale-up operation only in case needed (For the models that are not frequently used), instead of running hundreds of pods serving hundreds of models while most of them are not being used - which is a huge waste of computing resources.
What you are trying to do is to scale deployment to zero when these are not used.
K8s does not provide such functionality out of the box.
You can achieve it using Knative Pod Autoscaler.
Knative is probably the most mature solution available at the moment of writing this answer.
There are also some more experimental solutions like osiris or zero-pod-autoscaler you may find interesting and that may be a good fit for your usecase.
having read this article about a guy who uses tensorflow to sort cucumber into nine different classes I was wondering if this type of process could be applied to a large number of classes. My idea would be to use it to identify Lego parts.
At the moment, a site like Bricklink describes more than 40,000 different parts so it would be a bit different than the cucumber example but I am wondering if it sounds suitable. There is no easy way to get hundreds of pictures for each part but does the following process sound feasible :
take pictures of a part ;
try to identify the part using tensorflow ;
if it does not identify the correct part, take more pictures and feed the neural network with them ;
go on with the next part.
That way, each time we encounter a new piece we "teach" the network its reference so that it can better be recognized the next time. Like that and after hundreds of iterations monitored by a human, could we imagine tensorflow to be able to recognize the parts? At least the most common ones?
My question might sound stupid but I am not into neural networks so any advice is welcome. At the moment I have not found any way to identify a lego part based on pictures and this "cucumber example" sounds promising so I am looking for some feedback.
Thanks.
You can read about the work of Jacques Mattheij, he actually uses a customized version of Xception1 running on https://keras.io/.
The introduction is Sorting 2 Metric Tons of Lego.
In Sorting 2 Tons of Lego, The software Side you can read:
The hard challenge to deal with next was to get a training set large
enough to make working with 1000+ classes possible. At first this
seemed like an insurmountable problem. I could not figure out how to
make enough images and to label them by hand in acceptable time, even
the most optimistic calculations had me working for 6 months or longer
full-time in order to make a data set that would allow the machine to
work with many classes of parts rather than just a couple.
In the end the solution was staring me in the face for at least a week
before I finally clued in: it doesn’t matter. All that matters is that
the machine labels its own images most of the time and then all I need
to do is correct its mistakes. As it gets better there will be fewer
mistakes. This very rapidly expanded the number of training images.
The first day I managed to hand-label about 500 parts. The next day
the machine added 2000 more, with about half of those labeled wrong.
The resulting 2500 parts where the basis for the next round of
training 3 days later, which resulted in 4000 more parts, 90% of which
were labeled right! So I only had to correct some 400 parts, rinse,
repeat… So, by the end of two weeks there was a dataset of 20K images,
all labeled correctly.
This is far from enough, some classes are severely under-represented
so I need to increase the number of images for those, perhaps I’ll
just run a single batch consisting of nothing but those parts through
the machine. No need for corrections, they’ll all be labeled
identically.
A recent update is Sorting 2 Tons of Lego, Many Questions, Results.
1CHOLLET, François. Xception: Deep Learning with Depthwise Separable Convolutions. arXiv preprint arXiv:1610.02357, 2016.
I have started this using IBM Watson's Visual Recognition.
I had six different bricks to be recognized on the transport belt background.
I am actually thinking about tensorflow, since I can have it running locally.
The codelab : TensorFlow for Poets, describes almost exactly what you want to achieve,
For a demo of the Watson version:
https://www.ibm.com/developerworks/community/blogs/ibmandgoogle/entry/Lego_bricks_recognition_with_Watosn_lego_and_raspberry_pi?lang=en
I'm currently using a Processing Kinect library which supplies a depth map. I was wondering how I could take that and use it to create a 2D skeleton, if possible. Not looking for any code here, just a general process I could use to achieve those results.
Also, given that we've seen this in several of the Kinect games so far, would it be difficult to have multiple skeletons running at once?
Disclaimer: the reason why you still didn't get an answer for this question is probably because that's a current research problem. So I can't give you a direct answer but will try to help with some information and useful resources for this topic.
There are mainly 2 different approaches to create a skeleton from a depth map. The first one is to use machine learning, the second is purely algorithmic.
For the machine learning one, you'd need many samples of people doing a predetermined move, and use those samples to train your favorite learning algorithm. That's the approach that was taken and implemented by Microsoft in the XBox (source), it works really well BUT you need millions of samples to make it reliable... quite a drawback.
The "algorithmic" approach (understand without using a training set) can be done in many different ways and is a research problem. It's often based on modeling the possible body postures and trying to match that with the depth image received. That's the approach that was chosen by PrimeSense (the guys behind the kinect depth camera technology) for their skeleton tracking tool NITE.
The OpenKinect community maintains a wiki where they list some interesting research material about this topic. You might also be interested in this thread on the OpenNI mailing list.
If you're looking for an implementation of a skeleton tracking tool, PrimeSense released NITE (closed source), the one they made: it's part of the OpenNI framework. That's what's used in most of the videos you might have seen that involve skeleton tracking. I think it's able to handle up to 2 skeletons at the same time, but that requires confirmation.
The best solution is to use FAAST (http://projects.ict.usc.edu/mxr/faast/) which requires OpenNI. I have struggled to get OpenNI to work on my computer. I have not seen an approach yet using Code Laboratories' CL NUI.
An algorithmic approach is http://code.google.com/p/skeletonization/ but you may have a problem because your depthmap only represents surfaces and no closed objects.