DeepLearning4J - Acquiring Data and Train Model - neural-network

I try to create the easiest of a NeuralNetwork and training it with some data:
Therefore I created a test.csv with a the following pattern:
number,number+1;
number2,number2+1
...
I try to make a linear regression with the network...
But I do not find a way to acquire the data, DataSetIterator does not work.
How to fit the Data, how to test the Data?

In our examples, we encourage people to use datavec + recordreaderdatasetiterator.
Datavec has all of the various data loading components.
I'm not sure what you mean about "datasetiterator not working" wihtout seeing any code, but it seems like you didn't really look at our examples.
In there are multiple examples of a csv record reader you can use for both regression and classification use cases.
Consider reorienting your data pipeline to use those.
Those examples are always found here:
https://github.com/deeplearning4j/dl4j-examples
If you follow any of those, the same pattern emerges:
Record reader for whatever data format -> RecordReaderDataSetIterator
The iterator allows you to specify common constructors such as whether it is a regression or not, which column your label is etc.

Related

How to implement Featuretools into my ML Process?

I am exploring the possibility of implementing Featuretools into my pipeline, to be able to create new features from my Df.
Currently I am using a GridSearchCV, with a Pipeline embedded inside it. Since Featuretools is creating new features with aggregation on columns, like STD(column) etc, I feel like it is suspectible to data leakage. In their FAQ, they are giving an example approach to tackle it, which is not suitable for a Pipeline structure I am using.
Idea 0: I would love to integrate it directly into my Pipeline but it seems like not compatible with Pipelines. It would use fold train data to construct features, transform fold test data. K times. At the end, it would use whole data to construct, during Refit= True stage of GridSearchCV. If you have any example opposed to this fact, you are very welcome.
Idea 1: I can switch to a manual CV structure, not embedded into pipeline. And inside it, I can use Train data to construct new features, and test data to transform with these. It will work K times. At the end, all data can be used to construct Ultimate model.
It is the safest option, with time and complexity disadvantages.
Idea 2: Using it with whole data, ignore the leakage possibility. I am not in favor of this of course. But when I look at Project Github page, all the examples are combining Train and Test data, creating these features with whole data. Then go on with Train-Test division for modeling.
https://github.com/Featuretools/predict-taxi-trip-duration/blob/master/NYC%20Taxi%203%20-%20Simple%20Featuretools.ipynb
Actually if the developers of the project think like that, I could give it a chance with whole data.
What do you think, I would love to hear about your experiences on FeatureTools.

What is currently the best way to add a custom dictionary to a neural machine translator that uses the transformer architecture?

It's common to add a custom dictionary to a machine translator to ensure that terminology from a specific domain is correctly translated. For example, the term server should be translated differently when the document is about data centers, vs when the document is about restaurants.
With a transformer model, this is not very obvious to do, since words are not aligned 1:1. I've seen a couple of papers on this topic, but I'm not sure which would be the best one to use. What are the best practices for this problem?
I am afraid you cannot easily do that. You cannot easily add new words to the vocabulary because you don't know what embedding it would get during training. You can try to remove some words, or alternatively you can manually change the bias in the final softmax layer to prevent some words from appearing in the translation. Anything else would be pretty difficult to do.
What you want to do is called domain adaptation. To get an idea of how domain adaptation is usually done, you can have a look at a survey paper.
The most commonly used approaches are probably model finetuning or ensembling with a language model. If you want to have parallel data in your domain, you can try to fine-tune your model on that parallel data (with simple SGD, small learning rate).
If you only have monolingual data in the target language, you train a language model on that data. During the decoding, you can mix the probabilities from the domain-specific language and the translation model. Unfortunately, I don't know of any tool that could do this out of the box.

Convert PySpark ML Word2Vec model to Gensim Word2Vec model

I've generated a PySpark Word2Vec model like so:
from pyspark.ml.feature import Word2Vec
w2v = Word2Vec(vectorSize=100, minCount=1, inputCol='words', outputCol = 'vector')
model = w2v.fit(df)
(The data that I used to train the model on isn't relevant, what's important is that its all in the right format and successfully yields a pyspark.ml.feature.Word2VecModel object.)
Now I need to convert this model to a Gensim Word2Vec model. How would I go about this?
If you still have the training data, re-training the gensim Word2Vec model may be the most straightforward approach.
If you only need the word-vectors, perhaps PySpark's model can export them in the word2vec.c format that gensim can load with .load_word2vec_format().
The only reason to port the model would be to continue training. Such incremental training, while possible, involves considering a lot of tradeoffs in balancing the influence of the older and later training to get good results.
If you are in fact wanting to do this conversion in order to do more training in such a manner, it again suggests that using the original training to reproduce a similar model could be plausible.
But, if you have to convert the model, the general approach would be to study the source code and internal data structures of the two models, to discover how they alternatively represent each of the key aspects of the model:
the known word-vectors (model.wv.vectors in gensim)
the known-vocabulary of words, including stats about word-frequencies and the position of individual words (model.wv.vocab in gensim)
the hidden-to-output weights of the model (`model.trainables' and its properties in gensim)
other model properties describing the model's modes & metaparameters
A reasonable interactive approach could be:
Write some acceptance tests that take models of both types, and test whether they are truly 'equivalent' for your purposes. (This is relatively easy for just checking if the vectors for individual words are present and identical, but nearly as hard as the conversion itself for verifying other ready-to-be-trained-more behaviors.)
Then, in an interactive notebook, load the source model, and also create a dummy gensim model with the same vocabulary size. Consulting the source code, write Python statements to iteratively copy/transform key properties over from the source into the target, repeatedly testing if they verify as equivalent.
When they do, take those steps you did manually and combine them into a utility method to do the conversion. Again verify its operation then try using the converted model however you'd hoped – perhaps discovering overlooked info or discovering other bugs in the process, and then improving the verification method and conversion method.
It's possible that the PySpark model will be missing things the gensim model expects, which might require synthesizing workable replacement values.
Good luck! (But re-train the gensim model from the original data if you want things to just be straightforward and work.)

Spark/Mllib Train many GaussianMixture models in a distributed way

I've been playing around with the Gaussian Mixture Models provided for spark/mllib.
I found it really nice to generate a GaussianMixture from an enormous number of vectors/points. However, this is not always the case in ML. Very often you do not need to generate a model from numberless vectors, but to generate a numberless models -each one- from a few vectors (i.e., building a GMM for each user of a database with hundred of users).
At this point, I do not know how to proceed with the mllib, as I cannot see an easy way to distribute in both by users and by data.
Example:
Let featuresByUser = RDD[user, List[Vectors]],
the natural way to train a GMM for each user might be something like
featuresByUser.mapValues(
feats => new GaussianMixture.set(nGaussians).run(sc.parallelize(feats))
)
However, it is well-known that this is forbidden in spark. The inside sc.parallelize is not in the driver, so this leads to an error.
So the question are,
should the Mllib methods accept Seq[Vector] as input apart from
RDD[Vector] Thus, the programmer could choose one of the other depending on the problem.
Is there any other workaround that I'm missing to deal with this case (using mllib)?
Mllib unfortunately is currently not meant to create many models, but only one at the time, which was confirmed at a recent Spark meetup in London.
What you can do is launch a separate job for each model in a separate thread in the driver. This is described in the job scheduling documentation. So you would create one RDD per user and run a Gaussian mixture on each, running the 'action' that makes the thing run for each on a separate thread.
Another option, if the amount of data per user fits on one instance, you can do a Gaussian mixture on each user with something else than Mllib. This approach was described in the meetup in a case where sklearn was used within PySpark to create multiple models. You'd do something like:
val users: List[Long] = getUsers
val models = sc.parallelize(users).map(user => {
val userData = getDataForUser(user)
buildGM(userData)
})

Continue training a Doc2Vec model

Gensim's official tutorial explicitly states that it is possible to continue training a (loaded) model. I'm aware that according to the documentation it is not possible to continue training a model that was loaded from the word2vec format. But even when one generates a model from scratch and then tries to call the train method, it is not possible to access the newly created labels for the LabeledSentence instances supplied to train.
>>> sentences = [LabeledSentence(['first', 'sentence'], ['SENT_0']), LabeledSentence(['second', 'sentence'], ['SENT_1'])]
>>> model = Doc2Vec(sentences, min_count=1)
>>> print(model.vocab.keys())
dict_keys(['SENT_0', 'SENT_1', 'sentence', 'first', 'second'])
>>> sentence = LabeledSentence(['third', 'sentence'], ['SENT_2'])
>>> model.train([sentence])
>>> print(model.vocab.keys())
# At this point I would expect the key 'SENT_2' to be present in the vocabulary, but it isn't
dict_keys(['SENT_0', 'SENT_1', 'sentence', 'first', 'second'])
Is it at all possible to continue the training of a Doc2Vec model in Gensim with new sentences? If so, how can this be achieved?
My understand is that this is not possible for any new labels. We can only continue training when the new data has the same labels as the old data. As a result, we are training or retuning the weights of the already learned vocabulary, but are not able to learn a new vocabulary.
There is a similar question for adding new labels/words/sentences during training: https://groups.google.com/forum/#!searchin/word2vec-toolkit/online$20word2vec/word2vec-toolkit/L9zoczopPUQ/_Zmy57TzxUQJ
Also, you might want to keep an eye on this discussion:
https://groups.google.com/forum/#!topic/gensim/UZDkfKwe9VI
Update: If you want to add new words to an already trained model, take a look at online word2vec here:
http://rutumulkar.com/blog/2015/word2vec/
According to gensim documentation online/incremental training is not supported for doc2vec.
refer to https://github.com/RaRe-Technologies/gensim/issues/1019
I could still add new documents to an existing doc2vec model( but some it crashes due to segmentation fault) but most similar query does not work on newly added document(so this approach seems useless).