Proposal for lightening talk: Transfer learning with pre-trained models

[suvarnak]

Training a deep neural network requires lot of labelled data, engineering (for choosing architecture and tuning hyper parameters) and of course, compute. If wish to reap the benefits for deep learning without the pain of training & tuning, pre-trained models can provide solution. This introductory talk can help in getting started with transfer learning.

Prerequisite

Familiarity with Python, Machine Learning and Neural Networks

Scope

Common steps to train a deep network Transfer Learning with Pretrained Models Repurposing a pretrained model for your own needs Example usecase for image classification

[mayankskb]

Hey Suvarna,

Thanks for making an issue. Just FYI, lightning talk will be of 15 mins max. You need to completed your talk with that period only.

Please give confirmation for the above info and your consent for the same.

Also, let me know about your availability for Oct 5.

Thanks Mayank Mishra

[suvarnak]

Hello Mayank,

I am ok with 15 minutes talk. I wont be able to have detailed code walkthrough for the usecase, but I can share the code for participants to explore later. I am available on Oct 5.

Regards

[mayankskb]

Thanks!! Scheduling your lightning talk for Oct 5, 2019.

Regards, Mayank Mishra

[mayankskb]

Hey,

We will be recording your session. So if you have slides/demo/paper. Kindly keep it handy.

Also, you need to install screen recording software which can capture the presentation on your laptop.

If you have any other alternate go with that if not then please go through the following readme to get an available screen recorder for your OS:

https://github.com/sigmakappa/ScreenRecorders

Let me know if a screen recorder for your OS is not available here.

Also, after installing do testing of it that it is working properly or not.

Also, please give confirmation, once done with screen recorder testing stuff.

Thanks Mayank Mishra

[suvarnak]

Thanks Mayank, I just tested screen recording with ShareX and its working fine. Regards Suvarna

[mayankskb]

Hi,

Please send a small write-up about you. We will use it for introducing you.

Thanks Mayank Mishra

[suvarnak]

Hello Mayank,

Just mailed you. Please check.

Regards

[mayankskb]

Hi,

Please fill this form: https://forms.gle/h784Vv3cd2D6HY4M9

Thanks Mayank Mishra

[suvarnak]

Filled. Thanks .

On Fri, Sep 27, 2019 at 10:42 AM Mayank Mishra notifications@github.com wrote:

Hi,

Please fill this form: https://forms.gle/h784Vv3cd2D6HY4M9

Thanks Mayank Mishra

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/PyDataPune/Talks/issues/30?email_source=notifications&email_token=ABPPIJ6FLPDB4KDOCGGI4QTQLWI2VA5CNFSM4IUERXI2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD7XXYBQ#issuecomment-535788550, or mute the thread https://github.com/notifications/unsubscribe-auth/ABPPIJYEC7ULMDGNU4V4YODQLWI2VANCNFSM4IUERXIQ .

[mayankskb]

Thanks Suvarna for your kind support. See you on Oct 05, 2019

[suvarnak]

I've shared an item with you:

TransferLearning_Suvarna.mp4 https://drive.google.com/file/d/1Tuj5z4rAnei3LlEGwAcmHPS7vq17-wIc/view?usp=sharing&ts=5d999a02

It's not an attachment -- it's stored online. To open this item, just click
the link above.

Hi,

This is screen recording of my talk on Transfer Learning

Best Regards

[suvarnak]

Hello Mayank,

PF the code and slides at https://github.com/suvarnak/TL_demo.git I have also uploaded the video of screen recording of the talk in my google drive.

Thanks & Best Regards Suvarna

[mayankskb]

Hi Suvarna,

Kindly upload the screen recording to this folder. Link: https://drive.google.com/drive/folders/1NvszNmVAP-T5QwLH4yuBpQJWIfHudBm7?usp=sharing

Let me know in case you do not have access to this folder.

Thanks Mayank Mishra

[mayankskb]

Thanks, Suvarna. Loved your presence at PyData Pune event. Your talk was quite informative. Would love to hear more in the future from you. Feel free to make talk proposals in the future.

Closing this issue. Thanks Mayank Mishra