Can I use BERT or BART for machine translation? [closed] - machine-learning

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 1 year ago.
Improve this question
I am working on a project to use a pre-trained model and finetune it for customized language translations, for example from English to French. Is it possible to load these models in Tensorflow and run them to see how translations turn out and fine-tune afterward?

Probably the fastest way to do so is relying on the HuggingFace transformers library. If you're not familiar with it, you may take a look at their official documentation. To fine-tune a BART for NMT you can use directly this provided script (it works with some other pre-trained models too).

Related

Kaggle Beginner problems [closed]

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 2 years ago.
Improve this question
I am looking for beginner Machine Learning Linear Regression problems. I searched in Kaggle, but couldn't find a proper one. Can you please suggest me a beginner problem from Kaggle, or from any other platform?
Thanks in advance.
Kaggle has tons of linear regression notebooks and datasets to learn from, most popular ones are probably about house pricing (given certain house features predict it's price).
Here's a new one I'm looking forward to solve:
Ben & Jerry's flavours and ratings ---> products.csv
The main goal would be predict wich ice cream flavours are better accepted based on it's ingredients.

Making an Application Using YOLO [closed]

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 4 years ago.
Improve this question
I'm new to YOLO and trying to make car counting application using YOLO. The cars is from video file. Is there any reference? Thank you
There are multiple places from which I suggest you try to learn:
1) The CNN course from coursera https://www.coursera.org/learn/convolutional-neural-networks
This course has a good explanation on yolo(There assignment is on car detection as well which can easily be extended to car counting) and the rest of the course is quite nice as well
2)https://towardsdatascience.com/yolo-you-only-look-once-real-time-object-detection-explained-492dc9230006
The article focus on a few implementation details and talks about the papers yolo and yolov2 , and helped me clear a out a few issues i had when i was trying to implement yolo
3)The original paper (although this may be too advanced ): https://arxiv.org/pdf/1506.02640v5.pdf
4)A keras implmentation : https://github.com/experiencor/keras-yolo2
A simple git clone if you wish to simply have the code , although i do not recommend this as it has very little actual learning and is simply a download and use option

What should I do first for NLP related task? [closed]

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 4 years ago.
Improve this question
I am a sort of newbie to NLP world.
But anyway, I have just started my NLP project.
My task is about inferring hidden sentence in a paragraph.
Let me show you an example question.
a multiple choice question about inferring a clause in the blank
I want my machine learning model to extract some meaningful phrase from the given text(in above image, a paragraph)
I know that my question sounds quite ambiguous for you all. I just want to know even a small clue.
Thank you for your response in advance.
Skip-thought vectors are a system for predicting sentences from a context, by essentially constructing sentence-wide vectors. Might be useful, especially so in combination with context2vec if you want to build a custom model.

hierarchical classification in sklearn [closed]

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 1 year ago.
Improve this question
I would like to know if there is an implementation of hierarchical classification in the scikit-learn package or in any other python package.
Thank you so much in advance.
I couldn't find an implementation of Hierarchical Classification on scikit-learn official documentation. But I found this repository recently. This module is based on scikit-learn's interfaces and conventions. I hope this will be useful.
https://github.com/globality-corp/sklearn-hierarchical-classification
The globality-corp repo was deleted, but I found this fork, so I am posting it:
https://github.com/fbenites/sklearn-hierarchical-classification
As above, it is a module following the scikit-learn's conventions and is used along with sklearn classifiers.

Cheat Sheet or Flowchart for Computer Vision [closed]

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 8 years ago.
Improve this question
I was curious if anyone knew of something like this flowchart but for Computer Vision tasks? Specifically for OpenCV would be most ideal.
Or any references with best practices, and common patterns for Computer Vision problems?
That's a monumental task. The best I could find is from this article and it's a little bit old:
Maybe it's a good time to commit to FlexCV on Kickstarter.com, a GUI for OpenCV that allows you to create complex algorithms in a matter of minutes by connecting graphical elements together. It's an alternative for Adaptive Vision, but purely based on OpenCV features.

Resources