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
Related
I'm playing the game Factorio, where you build a factory.
For the time being, I made a kind-of flowchart using libreoffice calc to calculate how many machines I need to produce a certain material.
Example image from the spreadsheet
Each block has a recipe saved (blue). This recipe includes what and how much it produces and needs and how much time it takes.
It takes the demand from the previous Block (yellow) and, using the recipe, calculates how many machines (green) it needs to fulfill this demand.
Based on the amount of machines it calculates its own demands (orange).
Then the following blocks do the same, until it has reached the last block.
Doing this in a spreadsheet does work, but it is quite a tedious task.
I showed this to my dad, as I'm quite proud of what I made, and he said that maybe a database would be more suitable.
I definitely see its advantages. For example I could easily summarize the final demands of raw resources, or the total power consumption, etc.
So I got myself Microsoft Access, and I'm pretty lost now. I know the basics of Databases and some SQL-Coding, but I'm not quite sure how I would make this.
My first attempt was:
one table for machines. It includes the machines production speed and other relevant stats.
one table for recipes. Each recipe clearly states what it produces, what it needs, the amount of each, and whether or not it is a basic. Basic means that it is a raw resources, i.e. the production chain would end with this.
one table for units. Each unit has a machine, a recipe and an amount. For example I would have one unit using basic assemblers to produce iron gears. This unit also says how many machines there are, so it needs more and produces more.
I did manage to make a query that calculates the total in and outputs of all units based on their machine and recipe, as well as a total energy consumption.
However, that is nowhere near the spreadsheet I made.
For now we can probably set the Graphical overlay aside, that would probably be quite a bit overkill. However what I do want to be able to make:
enter how much I want of a certain resources
based on that entry the database would create a new table. The first entry would be the unit that produces the requested resources. The second would fulfill the firsts demand, the third fulfills the seconds demand, and so on.
So in the end I would end up with a list of units that will produce my requested resource.
I hope someone can help me. There are programs out there that already do this kind of stuff, but I want to do this myself. If this is a problem that a database isn't suited for, then please tell me so.
Thanks for any help!
I'm relatively new and am still learning the basics. I've used NVIDIA DIGITS in the past, and am now looking at Tensorflow. While I've been able to fumble my way around creating some models for a few projects I'm working on, I really want to start diving deeper into what I'm doing, how I'm doing it, and ultimately a better understanding of why.
One area that I would like to start with is the Images that I'm using for training and testing. Can anyone point me to a blog, an article, a paper, or give me some insight in what I need to consider when selecting images to train a new model on. Up until recently, I've been using datasets that have already been selected and that are available for download. Lets say I'm going to start working on a project that involves object detection of ships from a variety of distances and angles.
So my thoughts would be
1) I need a large quantity of images.
2) The images need to contain ships of the different types I would like to detect. (lets just say one class, ships, don't care what type of ships)
3) I also need to have images that have a great variety of distance perspective for the different types of ships.
Ultimately, my thoughts are that the images need to reflect the distance, perspective, and types of ships I would ideally want to identify from the video. Seems simple enough.
However, there are a number of questions
Does the images need to be the same/similar resolution as the camera I'll be using, for best results?
Does the images all need to be the same resolution?
Can I use a single image and just digitally zoom out on the image to give the illusion of different distances?
I'm sure there are a number of other questions that I'm not asking, or should be asking. Are there any guide lines available for creating a solid collection of images to use when creating the collection of images for training and validation?
I recommend thinking through end to end, like would you need to classify ship models as a next step? I recommend going through well known public datasets and actually work with the structure, how to store data, labels, how to handle preprocessing etc.
More importantly, what are you trying to achieve? Talking to experts in the topic does help greatly while preparing your own dataset.
Use open source images if you can, e.g. flickr, google, imagenet.
No, you don't need them to be the same resolution.
It is not ideal to zoom in/out images to use in different categories. Preprocessing images and data augmentation already does this to create more distant representations of the same class. This is why I would recommend hands on approach with an existing dataset first.
Yes, what you need is many, different representations of classes, and a roughly balanced dataset of classes. If you define your data structure well in the beginning, it will save you a ton of time as you won't have to make changes often.
I'm looking into designing a software platform that will aid linguists and anthropologists in their study of previously unstudied languages. Statistics show that around 1,000 languages exist that have never been studied by a person outside of their respective speaker groups.
My goal is to utilize TensorFlow to make a platform that will allow linguists to study and document these languages more efficiently, and to help them create written systems for the ones that don't have a written system already. One of their current methods of accomplishing such a task is three-fold: 1) Record a native speaker conversing in the language, 2) Listening to that recording and trying to transcribe it into the IPA, 3) From the phonetics, analyzing the phonemics and phonotactics of the language to eventually create a written system for the speaker.
My proposed platform would cut that research time down from a minimum of a year to a maximum of six months. Before I start, I have some questions...
What would be required to train TensorFlow to transcribe live audio into the IPA? Has this already been done? and if so, how would I utilize a previous solution for this project? Is a project like this even possible with TensorFlow? if not, what would you recommend using instead?
My apologies for the magnitude of this question. I don't have much experience in the realm of machine learning, as I am just beginning the research process for this project. Any help is appreciated!
I guess I will take a first shot at answering this. Since the question is pretty general, my answer will have to be pretty general as well.
What would be required. At the very least you would have to have a large dataset of pre-transcribed data. Ideally a large amount of spoken language audio mapped to characters in the phonetic alphabet, so the system could learn the sound of individual characters rather than whole transcribed words. If such a dataset doesn't exist, a less granular dataset could be used, mapping single words to their transcriptions. Then you would need a model, that is the actual neural network architecture implemented in code. And lastly you would need some computing resources. This is not something you can train casually, you would either have to buy some time in a cloud based machine learning framework (like Google Cloud ML) or build a fairly expensive machine to train at home.
Has this been done? I don't know. I don't think so. There have been published papers reporting various degrees of success at training systems to transcribe speech. Here is one, for example, http://deeplearning.stanford.edu/lexfree/lexfree.pdf It seems that since the alphabet you want to transcribe to is specifically designed to capture the way words sound rather than just write down the words you might have more success at training such a model.
Is it possible with TensorFlow. Yes, most likely. TensorFlow is well suited for implementing most modern deep learning architectures. Unless you end up designing some really weird and very original model for this purpose, TensorFlow should work just fine.
Edit: after some thought in part 1, you would have to use a dataset mapping spoken words to their transcriptions, since I expect that the same sound pronounced separately would be different from when the same sound is used in a word.
This has actually been done, albeit in PyTorch, by a group at CMU: https://github.com/xinjli/allosaurus
This issue is seen when performing training against my own dataset which was converted to binary via data_convert_example.py. After a week of training I get decode results that don't make sense when comparing the decode and ref files.
If anyone has been successful and gotten results similar to what is posted in the Textsum readme using their own data, I would love to know what has worked for you...environment, tf build, number of articles.
I currently have not had luck with 0.11, but have gotten some results with 0.9 however the decode results are similar to those shown below which I have no idea where they are even coming from.
I currently am running Ubuntu 16.04, TF 0.9, CUDA 7.5 and CuDnn 4. I tried TF 0.11 but was dealing with other issues so I went back to 0.9. It does seem that the decode results are being generated from valid articles, but the reference file and decode file indicies have NO correlation.
If anyone can provide any help or direction, it would be greatly appreciated. Otherwise, should I figure anything out, I will post here.
A few final questions. Regarding the vocab file referenced. Does it at all need to be sorted by word frequency at all? I never performed anything along these lines when generating it and just wasn't sure if this would throw something off as well.
Finally, I made the assumption in generating the data that the training data articles should be broken down into smaller batches. I separated out the articles into multiple files of 100 articles each. These were then named data-0, data-1, etc. I assume this was a correct assumption on my part? I also kept all the vocab in one file which has not seemed to throw any errors.
Are the above assumptions correct as well?
Below are some ref and decode results which you can see are quite odd and seem to have no correlation.
DECODE:
output=Wild Boy Goes About How I Can't Be Really Go For Love
output=State Department defends the campaign of Iran
output=John Deere sails profit - Business Insider
output=to roll for the Perseid meteor shower
output=Man in New York City in Germany
REFERENCE:
output=Battle Chasers: Nightwar Combines Joe Mad's Stellar Art With Solid RPG Gameplay
output=Obama Meets a Goal That Could Literally Destroy America
output=WOW! 10 stunning photos of presidents daughter Zahra Buhari
output=Koko the gorilla jams out on bass with Flea from Red Hot Chili Peppers
output=Brenham police officer refused service at McDonald's
Going to answer this one myself. Seems the issue here was the lack of training data. In the end I did end up sorting my vocab file, however it seems this is not necessary. The reason this was done, was to allow the end user to limit the vocab words to something like 200k words should they wish.
The biggest reason for the problems above were simply the lack of data. When I ran the training in the original post, I was working with 40k+ articles. I thought this was enough but clearly it wasn't and this was even more evident when I got deeper into the code and gained a better understanding as to what was going on. In the end I increased the number of articles to over 1.3 million, I trained for about a week and a half on my 980GTX and got the average loss to about 1.6 to 2.2 I was seeing MUCH better results.
I am learning this as I go, but I stopped at the above average loss because some reading I performed stated that when you perform "eval" against your "test" data, your average loss should be close to what you are seeing in training. This helps to determine whether you are getting close to over-fitting when these are far apart. Again take this with a grain of salt, as I am learning but it seems to make sense logically to me.
One last note that I learned the hard way is this. Make sure you upgrade to the latest 0.11 Tensorflow version. I originally trained using 0.9 but when I went to figure out how to export the model for tensorflow, I found that there was no export.py file in that repo. When I upgrades to 0.11, I then found that the checkpoint file structure seems to have changed in 0.11 and I needed to take another 2 weeks to train. So I would recommend just upgrading as they have resolved a number of the problems I was seeing during the RC. I still did have to set the is_tuple=false but that aside, all has worked out well. Hope this helps someone.
I have developed two differing methods in MATLAB which aim to analyse a pop song and then automatically create a 30 second audio thumbnail (a preview clip) containing part of the chorus section.
Both methods have varying results:
The first method can create a thumbnail for each track, managing to find a chorus section in 40 out of 50 tested songs
The second method only managed to work on 30 out of the 50 songs, and it found the chorus section 21 times out the 30.
Obviously I know which method is superior, but I need to describe and explain the results in a report which requires the demonstration of proper statistical testing.
Other academic papers have previously used an f-test to do this, but because their methods are vastly superior, their aims are usually involve the detection of chorus onset times with 100% accuracy.
My aim is more relaxed as I am just looking for the generated thumbnails to contain any part of the chorus, regardless of onset.
Can anyone suggest some objective tests that I could possibly explore with regards to my project? This is my first time conducting an investigation like this so my experience/knowledge is incredibly low.
Thank you!
Possibly, the way for you is formating your song track with time cuts for relevant information about type of sound(chorus, etc). In sound editor like CoolEdit, you can set time cuts and assign names for theirs like 'chorus', 'pause','music'... Then, you must extract cut information to import in Matlab. For Windows 32 can be used utility Wav2labs from http://www.pallier.org/ressources/wspot/sig2wav/toolswav.html; http://www.pallier.org/ressources/wspot/sig2wav/Wav2labs.exe This program extract cuts to text file and you can read with Matlab textscan function.
After all, only segmentation accuracy must be proceed, like percent time when signal type(chorus/not chorus) was recognized correctly
Or specify your question more exactly