search

johko / computer-vision-course

This repo is the homebase of a community driven course on Computer Vision with Neural Networks. Feel free to join us on the Hugging Face discord: hf.co/join/discord
MIT License
371 stars 123 forks source link

Unit 4. Multimodal models transfer learning. CLIP transfer learning notebook #176

Closed minemile closed 5 months ago

minemile commented 5 months ago

Hi all!

We have prepared a CLIP transfer learning notebook. In this notebook we review various fine-tuning techniques for the classification task on Oxford-pet dataset.

I will be glad to any feedback!

review-notebook-app[bot] commented 5 months ago

Check out this pull request on  ReviewNB

See visual diffs & provide feedback on Jupyter Notebooks.

Powered by ReviewNB

review-notebook-app[bot] commented 5 months ago

View / edit / reply to this conversation on ReviewNB

merveenoyan commented on 2024-01-11T11:41:10Z ----------------------------------------------------------------

you can make this without a header: "Let's start with importing necessary functions, libraries and environmental variables."


minemile commented on 2024-01-16T16:11:22Z ----------------------------------------------------------------

Fixed

review-notebook-app[bot] commented 5 months ago

View / edit / reply to this conversation on ReviewNB

merveenoyan commented on 2024-01-11T11:41:11Z ----------------------------------------------------------------

"Exploratory Data Analysis on the Dataset"


minemile commented on 2024-01-16T16:11:27Z ----------------------------------------------------------------

Fixed

review-notebook-app[bot] commented 5 months ago

View / edit / reply to this conversation on ReviewNB

merveenoyan commented on 2024-01-11T11:41:12Z ----------------------------------------------------------------

Please add a dot at the end of the sentences.


review-notebook-app[bot] commented 5 months ago

View / edit / reply to this conversation on ReviewNB

merveenoyan commented on 2024-01-11T11:41:13Z ----------------------------------------------------------------

This is technically not OpenAI implementation, it's a model released by openai converted to transformers format that is you're using. You can say "We will use the pre-trained model by OpenAI" instead


minemile commented on 2024-01-16T16:10:45Z ----------------------------------------------------------------

Fixed

review-notebook-app[bot] commented 5 months ago

View / edit / reply to this conversation on ReviewNB

merveenoyan commented on 2024-01-11T11:41:14Z ----------------------------------------------------------------

Do you mean the text encoder here?

You can say "Fine-tuning Text Encoder"


review-notebook-app[bot] commented 5 months ago

View / edit / reply to this conversation on ReviewNB

merveenoyan commented on 2024-01-11T11:41:15Z ----------------------------------------------------------------

"Fine-tuning Image Encoder"


review-notebook-app[bot] commented 5 months ago

View / edit / reply to this conversation on ReviewNB

merveenoyan commented on 2024-01-11T11:41:16Z ----------------------------------------------------------------

"Fine-tune 30 percent of the Last Layers of Vision-Language Model" 


review-notebook-app[bot] commented 5 months ago

View / edit / reply to this conversation on ReviewNB

MKhalusova commented on 2024-01-15T18:19:34Z ----------------------------------------------------------------

If you use 1. instead of 1), it'll be properly formatted as a numbered list.


minemile commented on 2024-01-16T16:11:01Z ----------------------------------------------------------------

Fixed

minemile commented 5 months ago

Fixed

View entire conversation on ReviewNB

minemile commented 5 months ago

Fixed

View entire conversation on ReviewNB

minemile commented 5 months ago

Fixed

View entire conversation on ReviewNB

minemile commented 5 months ago

Fixed

View entire conversation on ReviewNB

minemile commented 5 months ago

Thanks for your comments! I tried to fix everything.

  • © Githubissues.
  • Githubissues is a development platform for aggregating issues.