I am following this tutorial about Google Cloud Platform (GCP) for deep learning. According to the tutorial it says
The --train_steps option specifies the total number of training batches.
However, in the code of the same tutorial it says
'--train_steps', help='Steps to run the training job for.'
Now, I am confused because I found several questions regarding this topic here in StackOverflow and other sources saying that the training steps corresponds to the number of iterations that the optimizer does to find the minimum. Can someone confirm which one of these three definitions is correct?
xerx is correct in that steps and batches are pretty much the same thing, and can be treated as doing exactly the same thing for your model. I write more about the training process, batch sizes, and overfitting in my book about stock prediction, which you can find here - it's great for learning Deep Learning.
Good Luck!
Related
I'm trying to train a model for a sentence classification task. The input is a sentence (a vector of integers) and the output is a label (0 or 1). I've seen some articles here and there about using Bert and GPT2 for text classification tasks. However, I'm not sure which one should I pick to start with. Which of these recent models in NLP such as original Transformer model, Bert, GPT2, XLNet would you use to start with? And why? I'd rather to implement in Tensorflow, but I'm flexible to go for PyTorch too.
Thanks!
It highly depends on your dataset and is part of the data scientist's job to find which model is more suitable for a particular task in terms of selected performance metric, training cost, model complexity etc.
When you work on the problem you will probably test all of the above models and compare them. Which one of them to choose first? Andrew Ng in "Machine Learning Yearning" suggest starting with simple model so you can quickly iterate and test your idea, data preprocessing pipeline etc.
Don’t start off trying to design and build the perfect system.
Instead, build and train a basic system quickly—perhaps in just a few
days
According to this suggestion, you can start with a simpler model such as ULMFiT as a baseline, verify your ideas and then move on to more complex models and see how they can improve your results.
Note that modern NLP models contain a large number of parameters and it is difficult to train them from scratch without a large dataset. That's why you may want to use transfer learning: you can download pre-trained model and use it as a basis and fine-tune it to your task-specific dataset to achieve better performance and reduce training time.
I agree with Max's answer, but if the constraint is to use a state of the art large pretrained model, there is a really easy way to do this. The library by HuggingFace called pytorch-transformers. Whether you chose BERT, XLNet, or whatever, they're easy to swap out. Here is a detailed tutorial on using that library for text classification.
EDIT: I just came across this repo, pytorch-transformers-classification (Apache 2.0 license), which is a tool for doing exactly what you want.
Well like others mentioned, it depends on the dataset and multiple models should be tried and best one must be chosen.
However, sharing my experience, XLNet beats all other models so far by a good margin. Hence if learning is not the objective, i would simple start with XLNET and then try a few more down the line and conclude. It just saves time in exploring.
Below repo is excellent to do all this quickly. Kudos to them.
https://github.com/microsoft/nlp-recipes
It uses hugging face transformers and makes them dead simple. 😃
I have used XLNet, BERT, and GPT2 for summarization tasks (English only). Based on my experience, GPT2 works the best among all 3 on short paragraph-size notes, while BERT performs better for longer texts (up to 2-3 pages). You can use XLNet as a benchmark.
I have images of unique products that are used at my workplace. I can't imagine that the inception database already has similar items that it has been trained on.
I tried to train a model using YOLO. It was taking a very very long time. Maybe 7minutes between epochs; and I wanted to do 1000 epochs due to small data size.
I used tiny-yolov2-voc cfg/weight on 1.0 GPU. I had a video of the item but i broke it up into frames so i could annotate. I then attempted to train on the images (not video). The products are healthcare related. Basically anything that a hospital would use.
Ive also used the inception method on images I got from Google. I noticed that inception method was very fast and resulted in accurate predictions. However, i'm worried that my images are too unique for inception to work.
Which method is best to use?
If you recommend YOLO, can you please provide suggestions on how to speed up the training phase?
If you recommend inception, can you please provide an explanation why it would work on unique images? I guess i'm having trouble understanding how inception knows which item i'm trying to train on without me providing annotations.
Thanks in advance
Just my impression (no recommendation or even related experience)
Having a look at the Hardware recommendations related to darknet a assumption is that you might stock up your own hardware to get faster results.
I read about the currently three different versions of YOLO and expect there are lot's of GFLOPS training included if you download the recommended files, but if the models never fit to your products then for you they never might be very helpful.
I must admit I've neither been active with YOLO nor with Tensorflow, so my impression might not be helpful at all.
If you see some videos of YOLO you can remark that sometimes a camel is labeled with horse and the accuracy seems being bad but it depends on the threshold that is applied to the images, so the videos look amazing as it seems the recognition is done so fast but with higher accuracy the process would slow down - also depending on the trained motives.
They never hide it though, they explain on an image where a dog is labeled as cow and a horse as sheep (Version 2) that in combination with darknet it's getting much faster but less accurate too, so usage of darknet is an important aspect too.
The information about details seems being quite bad on the websites of YOLO, they present it more like you'd do with a popstar, in comparison the website of Tensorflow looks more academic and is informing about the mathematics behind the framework.
Concerning Tensorflow I don't know about the hardware-recommendations, but as you wrote your results are useful, probably they are a bit or even much less.
My impression is that YOLO is primary intended for real-time detection in (live-)videos and needs much training for high accuracy. So depending on your use-case it might be right but you'd to invest in hardware probably for professional usage.
This is not an opinion against Tensorflow but that I had to verify more and it seems taking more time to get an impression. Concerning Tensorflow in the moment I even can't say if it can be used for real-time-detection, how accurate it is then and if the results are then still better then those of YOLO.
My assumption is that concerning both solutions it's a matter of involved elements (like the decision if to include darknet for speed), configuration, training and adjustments. Probably there is always something to increase in speed and accuracy, so investing in a system for recognition won't be static process with fixed end in timeline, but a steady process.
This is just a short overview of my impressions, I've never any experience with any recognition-software and hardly recommend that you make any decision based on my words.
Just if you want to do use any recognition software professional, especially for real-time-recognition, then you've to invest in hardware probably.
To my understanding of your problem you need you need inception with the capability of identifying your unique images. In this circumstance you can use transfer-learning on the inception model. With transfer-learning you can still train inception your own pictures while retaining the previous knowledge of inception.
More on transfer-learning
Problem
Do you have a tutorial for LTSM or RNN time series anomaly detection using deep learning with CNTK? If not, can you make one or suggest a series of simple steps here for us to follow?
I am a software developer and a member of a team investigating using deep learning on time series data we have for anomaly detection. We have not found anything on your python docs that can help us. It seems most of the tutorials are for visual recognition problems and not specific to the problem domain of interest to us.
Using LTSM and RNN in Anomaly Detection
I have found the following
This link references why we are trying to use time series for anomaly detection
This paper convinced us that the first link is a respected approach to the problem in general
This link also outlined the same approach
I look around on CNTK here, but didn't find any similar question and so I hope this question helps other developers in the future.
Additional Notes and Questions
My problem is that I am finding CNTK not that simple to use or as well documented as I had hoped. Frankly, our framework and stack is heavy on .NET and Microsoft technologies. So I repeat the question again for emphasis with a few follow ups:
Do you have any resources you feel you can recommend to developers learning neural networks, deep learning, and so on to help us understand what is going on under the hood with CNTK?
Build 2017 mentions C# is supported by CNTK. Can you please point us in the direction of where the documentation and support is for this?
Most importantly can you please help get us unstuck on trying to do time series anomaly analysis for time series using CNTK?
Thank you very much for time and assistance in reading and asking this question
Thanks for your feedback. Your suggestions help improve the toolkit.
First Bullet
I would suggest that you can start with the CNTK tutorials.
https://github.com/Microsoft/CNTK/tree/master/Tutorials
They are designed from CNTK 101 to 301. Suggest that you work through them. Many of them even though uses image data, the concept and the models are amenable to build solutions with numerical data. 101-103 series are great to understand basics of the train-test-predict workflow.
Second Bullet:
Once you have trained the model (using Python recommended). The model evaluation can be performed using different language bindings, C# being one of them.
https://github.com/Microsoft/CNTK/wiki/CNTK-Evaluation-Overview
Third Bullet
There are different approaches suggested in the papers you have cited. All of them are possible to do in CNTK with some changes to the code in the tutorials.
The key tutorial for you would be CNTK 106, CNTK 105, and CNTK 202
Anomaly as classification: This would involve you label your target value as 1 of N classes, with one of the class being "anomaly". Then you can combine 106 with 202, to classify the prediction
Anomaly as an autoencoder: You can need to study 105 autoencoder. Now instead of a dense network, you could apply the concept for Recurrent networks. Train only on the normal data. Once trained, pass any data through the trained model. The difference between the input and autoencoded version will be small for normal data but the difference will be much larger for anomalies. The 105 tutorial uses images, but you can train these models with any numerical data.
Hope you find these suggestions helpful.
This is a newbie question for the tensorflow experts:
I reading lot of data from power transformer connected to an array of solar panels using arduinos, my question is can I use tensorflow to predict the power generation in future.
I am completely new to tensorflow, if can point me to something similar I can start with that or any github repo which is doing similar predictive modeling.
Edit: Kyle pointed me to the MNIST data, which I believe is a Image Dataset. Again, not sure if tensorflow is the right computation library for this problem or does it only work on Image datasets?
thanks, Rajesh
Surely you can use tensorflow to solve your problem.
TensorFlow™ is an open source software library for numerical
computation using data flow graphs.
So it works not only on Image dataset but also others. Don't worry about this.
And about prediction, first you need to train a model(such as linear regression) on you dataset, then predict. The tutorial code can be found in tensorflow homepage .
Get your hand dirty, you will find it works on your dataset.
Good luck.
You can absolutely use TensorFlow to predict time series. There are plenty of examples out there, like this one. And this is a really interesting one on using RNN to predict basketball trajectories.
In general, TF is a very flexible platform for solving problems with machine learning. You can create any kind of network you can think of in it, and train that network to act as a model for your process. Depending on what kind of costs you define and how you train it, you can build a network to classify data into categories, predict a time series forward a number of steps, and other cool stuff.
There is, sadly, no short answer for how to do this, but that's just because the possibilities are endless! Have fun!
The tensorflow text summarization model as described here https://github.com/tensorflow/models/tree/master/textsum requires a multi GPU architecture in order to train. My repeated attempts at training the model has resulted in memory exceptions, machine crashing for various reasons. Is the trained summarisation model available so can make use of the summarization model without the need for training? The summarization model is trained using the not free Gigaword dataset, if the trained model is not available from Google is this a factor in reason why ?
So as far as I can tell, no one has put the trained model out there that is referenced. I too was originally running into memory issues on my macbook pro and eventually ended up using my gaming laptop which had a much better GPU.
The other option of course is to take advantage of AWS and use something like their g2.2xlarge instance. They also have their P2 instances as well, but I have not checked that out yet.
With regards to the Gigaword dataset, it simply comes down to licensing. It is not a free license from LDC and often many of the academics working on this have the dataset provided to them via their Universities or companies. I have not had luck finding it, however LDC did get back to me and advised that they do have other article datasets that have a pricetag of around $300 which is much more reasonable for those of use just trying to learn TF. That said, if you didn't want to buy anything, you can always write your own page scraper and format the data for the textsum model. https://github.com/tensorflow/models/pull/379/files
Hope this helps some. Good luck!