Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
    Garvit Agarwal
    @garvit984
    @mloning Can you please tell me what's the selection process for Outreachy? Is it the same just like the GSoC's selection procedure comprising of interview?
    mahakkothari
    @mahakkothari
    @mloning I have submitted a PR (#84) that adds multivariate LSTM Classifier in the sktime-dl repository, but due to some issue with the import of tabularize (Issue #76), some checks were not successful. Can you please review that and let me know if I should change anything.
    Taiwo Owoseni
    @thayeylolu
    Hello Everyone and @mloning . My name is Taiwo Owoseni from Nigeria. I am an Outreachy Applicant. I am Data Analyst and would be pursuing a Masters in Data Science by September. I built a staff promotion model using Python and scikit-learn. I am comfortable building applications with Python and analysing Data with SQL. I am not familiar with time series analysis but I am opened to learning. I am new to open source contribution and would love to make resourceful use of this opportunity to learn and contribute. Thank you for having me.
    Markus Löning
    @mloning
    Hi @thayeylolu and @garvit984, thanks for introducing yourself and welcome! I've created a placeholder here: https://github.com/alan-turing-institute/sktime/wiki/Outreachy-2021 which we will populate with more details over the next few days
    2 replies
    Preeti
    @Leovarmak1_twitter
    Hello all, Preeti Sharma an Outreachy applicant this is my first time contributing to a large organization I'm really looking forward to contributing and collaborate in this project.
    Markus Löning
    @mloning
    Welcome @Leovarmak1_twitter :wave:
    OLUFUNMILAYO RUTH AFORIJIKU
    @olufunmiruth
    Hello everyone.. I am Olufunmi. I am an outreachy applicant. I am looking forward to contributing and learning. I am glad to be here.
    Markus Löning
    @mloning
    Welcome @olufunmiruth :wave:
    Anastasia Rizzo
    @arizzogithub
    Hello all, I am Ana, can't wait to join this project, good luck to everyone!
    Astha Jakher
    @Aexousia
    Hello, I am Astha, I'm really looking forward to contributing here! :)
    Markus Löning
    @mloning
    Hi there @arizzogithub and @Aexousia
    Ankita Priya
    @AnkitaxPriya
    Hello everyone! I am Ankita Priya, an Outreachy applicant pursuing Bachelor's degree in Computer Science Engineering at Birla Institute of Technology, Mesra, Ranchi, India. I am currently working on a research paper in the field of data science. I'm very new to the concept of "time series analysis" but I wish to learn about it. Having a basic knowledge of Python, I wish to contribute to the project "Extend classification API to handle time-constant variables and predict Covid-19 mortality risk". I'd be more than happy if you could guide me @mloning . Thanks.
    Riddhima Singh
    @riddhima:matrix.org
    [m]
    Hello everyone! I am Riddhima Singh from India.I am an Outreachy applicant. I am a sophomore at Indian Institute of Engineering Science and Technology,Shibpur.I have been working with python and its libraries for a couple of months. I have used scikit-learn previously for linear regression and logistics regression but time series analysis is a bit new to me but I can definetly learn it. I am new to open source as well hence it would be very helpful if the mentor could guide me to make a first contribution tso that I can get started.
    Riddhima Singh
    @riddhima:matrix.org
    [m]
    :point_up: Edit: Hello everyone! I am Riddhima Singh from India.I am an Outreachy applicant. I am a sophomore at Indian Institute of Engineering Science and Technology,Shibpur.I have been working with python and its libraries for a couple of months. I have used scikit-learn previously for linear regression and logistics regression but time series analysis is a bit new to me but I can definetly learn it. I am new to open source as well hence it would be very helpful if the mentor could guide me to make a first contribution so that I can get started.
    Guzal Bulatova
    @GuzalBulatova
    Hi there everybody, nice to meet you all!
    I'm Guzal. I'm studying object-oriented programming in Gothenburg. Very interested in Machine Learning and explore Python by myself. Looking forward to contribute to the classification API project :)
    Ifeanyi30
    @Ifeanyi30
    Hey guys! I am happy to be an outreachy applicant as it has given me the opportunity to learn as part of a community, thanks to @mloning for accepting this mentorship position and I look forward to your intellectual guidance.
    Sree Vidhya Isukapalli
    @vidhya830_twitter
    Hi, I am Vidhya , a CS undergraduate from India. I am outreachy aspirant experienced in python programming and open source. I have used scikit-learn in machine learning models. Looking forward to contribute here :)
    sonali945
    @sonali945
    Hello everyone, I am Sonali Gupta from India. I am an Outreachy applicant. I am a sophomore at Indian Institute of Technology Kanpur. I am glad to be here and looking forward to make my contribution.
    aishagurung
    @aishagurung
    Hey everyone, I am Aisha(Aishwarya) Gurung. I am from the city of Kathmandu in Nepal where the smogs are currently taking over. During this time I hope to stay inside and make contributions to the sktime community projects. I am very interested in ML and learning more about time series analysis. Stay safe and please send any good vibes and suggestions my way!
    Ifeanyi30
    @Ifeanyi30
    @mloning I have been working on this issue "alan-turing-institute/sktime#583", tried running the command "pip install -r build_tools/requirements.txt" but some packages could not be installed. What should I do?
    Ifeanyi30
    @Ifeanyi30
    @mloning I made a PR for this issue "alan-turing-institute/sktime#583" but the pre-commit package gives a the flask8 error without an actual error message. Can you help me out?
    jowallner
    @jowallner
    Hi everyone! Is there a workaround for the sktime-dl installation bug? I get the error message "cannot import name 'check_X_y' from 'sktime.utils.validation' " when trying to import "from sktime_dl.classifiers.deeplearning import CNNClassifier"
    Markus Löning
    @mloning
    @jowallner we're looking for some help with sktime-dl to bring it up to speed with sktime
    Abhipsha Das
    @chiral-carbon
    Hi! I am Abhipsha Das, an Outreachy applicant from India. I graduated last year and am interested in ML and applications of ML. I am proficient in Python and have experience in writing ML code. The sktime projects piqued my interest, so I look forward to contributing and working with everyone! :smile:
    Yash Lamba
    @yashlamba
    Hi @mloning, for alan-turing-institute/sktime#421 what way are we looking forward to? I think subclassing from statsmodel's theta model might be a better way because it won't break our current api implementation. I'll also look for issues with fit_trend if I find any.
    Markus Löning
    @mloning
    @yashlamba best to comment on the issue itself
    1 reply
    Mark Zurbruegg
    @Betaglutamate
    Hi I am a biology student at the University of Edinburgh and am looking into time series segmentation of bacterial growth curves. I see you guys have time series segmentation on your road map but as I understand it, it is not implemented yet?
    Markus Löning
    @mloning
    @Betaglutamate that's right, it's not implemented yet, do you have labelled data (supervised learning)?
    Nicolaj Noergaard Muehlbach
    @muhlbach
    Hi everyone! Quick question: For simulation purposes, I want to iteratively predict one-step ahead (fh=1) but with updated input data at each step. No refitting is needed. I see a few different ways to implement this, but I wanted to ask if there's a preferred way that could be implemented across all forecasting methods available via sktime? Thanks in advance!
    5 replies
    Guzal Bulatova
    @GuzalBulatova

    Side note for fellows who watch the tutorial (https://www.youtube.com/watch?v=wqQKFu41FIw&ab_channel=CodingTech):
    the sktime version available now is a bit different from the one in the video. Some functions shown in the tutorial notebook don't work anymore.

    It's better to follow along with the code published on github (https://github.com/alan-turing-institute/sktime/blob/main/examples/01_forecasting.ipynb)

    11 replies
    Mark Zurbruegg
    @Betaglutamate
    Hi @mloning thanks for replying. Right now I have about 600 time series recordings of bacterial growth. I am interested in identifying the period of exponential growth in these traces. I can annotate this manually quite easily. Do you have any suggestions of how I could use this data to then predict the exponential growth phase in my future traces?
    7 replies
    Ifeanyi30
    @Ifeanyi30
    Happy Easter guys
    1 reply
    Markus Löning
    @mloning
    Hi all, for those interested in the Outreachy program, we've posted some application info here: https://github.com/alan-turing-institute/sktime/wiki/Outreachy-2021
    Ifeanyi30
    @Ifeanyi30
    Hi @mloning I made another PR: alan-turing-institute/sktime#776 . I'd like you to review it
    I would also like to try out the issue: alan-turing-institute/sktime#729
    Markus Löning
    @mloning
    If you're interested in GSoC or Outreachy and want to ask some questions, you're very welcome to join our coffee chat on Thursday, 2:30pm BST https://turing-uk.zoom.us/j/99045960066?pwd=V1dOOUhGSVlFaFRpY0hqOVRzMXREUT09 - let me know if you have any questions in advance that I should discuss :)
    4 replies
    Riya Elizabeth John
    @Riyabelle25
    Hi guys, are there some mentors reviewing pre-proposals for GSoC? I mailed mine (google docs) last week, but didn't get a response, so got concerned (as the deadline for proposal submission is approaching)
    2 replies
    Markus Löning
    @mloning
    For those interested in GSoC, please make sure to upload your proposal drafts on the official GSoC platform too.
    Khoa Lai
    @imkhoa99
    image.png
    Khoa Lai
    @imkhoa99
    image.png
    Hi, thanks for the wonderful work!
    I'm trying to fit both time-series dataframe into the TimeSeriesForestClassifier model to classify detect synthetic data from real data. I ran into a problem: "Tabularization failed, it's possible that not all series were of equal length". I worked around by taking the same length subset of data and fit the model but still failed. I saw that someone managed to solve it by "tabularizing each column individually". I am wondering what does it mean? Or are there other reasons that might be causing the issue? Both of my data set are as above. Thanks!
    2 replies
    Yash Lamba
    @yashlamba
    Hi! So while framing my proposal around time series clustering, I came across https://github.com/alan-turing-institute/sktime/tree/uea_dev/sktime this branch. It already has most of the clustering utils and algorithms already implemented. @mloning, are we looking forward to merging this in the main branch? Most of the basics are already implemented so I'll probably have to reframe my proposal.
    Markus Löning
    @mloning
    For those applying to GSoC, we've sent out some feedback today to the submitted draft proposals - please let us know if we have missed you!
    7 replies
    SHREYA SINGH
    @SinghShreya05
    Hi everyone! I am Shreya. I am an Outreachy applicant. I have been previously contributing in opensource in Python, Deep learning, Natural language processing, and other tech-stack in my domain. Can anyone guide me to get started and share the link where I have to make contribution?
    MJimitater
    @MJimitater
    Hello Team, thank you for pushing ever forward with improvements! I saw the latest release is v0.5.3, I'd like to propose a fresh release with all the improvements, for convenient pip install usage. This would allow the broad community to take advantage of your great work! Thanks
    Markus Löning
    @mloning
    Hi @MJimitater thanks for feedback - we should be able to make a new release soon, there are still some PRs that we want to get merged first. Let me know if you're interested in contributing!
    Hi @SinghShreya05 yes you can find out more here: https://github.com/alan-turing-institute/sktime/wiki/Outreachy-2021
    Taiwo Owoseni
    @thayeylolu
    Hi Everyone. I want to extract features from a dataset and transform it. I am using RandomIntervalFeatureExtractor(n_intervals="sqrt", features=[np.mean, np.std, _slope]) to extract features but I don't know where to import _slope from. Kindly assist. Thank you :)
    1 reply
    Markus Löning
    @mloning
    For everyone applying to GSoC, please make sure you submit your final (not the draft) proposal before the deadline!
    Taiwo Owoseni
    @thayeylolu
    Hello everyone. I am currently having a challenge understanding how to transform a data frame with shape (114,1) to a nested data frame.
    2 replies