New Notebook showcasing how to perform prompt tuning using the PEFT library. #33

peremartra · 2024-02-18T12:34:47Z

What does this PR do?

Is a new notebook showcasing the use of the PEFT library to perform Prompt Tuning. I also provide a brief explanation of what prompt-tuning is and how it works.

Since this is my first PR request, I hope I have created the notebook in the correct style. If not, I will try to adapt and make any necessary modifications.

Thank you for reviewing it. @MKhalusova

…ook-images dataset

review-notebook-app · 2024-02-18T12:34:52Z

Check out this pull request on

See visual diffs & provide feedback on Jupyter Notebooks.

Powered by ReviewNB

MKhalusova · 2024-02-19T19:39:27Z

@pacman100 Can you please review a notebook on prompt-tuning with peft?

MKhalusova · 2024-02-19T19:40:41Z

@peremartra Thank you for contributing a notebook! This looks very useful! There are a few more steps required:

Add the notebook to the _toctree.yml
Add the notebook to the list of the latest notebooks in the index.md. As this is the most recent addition, put it at the top of the list.
Right after the first header (notebook title), add yourself as an author, like this: _Authored by: [Your Name](https://huggingface.co/your_profile)_ Feel free to use either your Hugging Face profile, or GitHub profile, it's up to you which one to link.

I also fixed the link to the other notebooks on index.md, all of them are missing the .ipynb and the links were broken.

peremartra · 2024-02-19T22:32:03Z

@peremartra Thank you for contributing a notebook! This looks very useful! There are a few more steps required:

Add the notebook to the _toctree.yml

Add the notebook to the list of the latest notebooks in the index.md. As this is the most recent addition, put it at the top of the list.

Right after the first header (notebook title), add yourself as an author, like this: _Authored by: [Your Name](https://huggingface.co/your_profile)_ Feel free to use either your Hugging Face profile, or GitHub profile, it's up to you which one to link.

@MKhalusova Modifications Done! Thanks for the feedback :-)

peremartra · 2024-02-21T22:37:12Z

Hi @MKhalusova, I just finished another notebook about fine-tuning with QLoRA. I don't know how to proceed, should I do a second PR, or just wait to close this PR before opening a new one?

MKhalusova · 2024-02-22T13:20:41Z

Hi @MKhalusova, I just finished another notebook about fine-tuning with QLoRA. I don't know how to proceed, should I do a second PR, or just wait to close this PR before opening a new one?

Feel free to open a separate PR for a different notebook. We can review them in parallel and it's better to keep one notebook per PR.

pacman100 · 2024-02-23T09:22:06Z

notebooks/en/Prompt_Tuning_PEFT.ipynb

@@ -0,0 +1,7324 @@
+{


Nice overview!

Reply via ReviewNB

pacman100 · 2024-02-23T09:22:06Z

notebooks/en/Prompt_Tuning_PEFT.ipynb

@@ -0,0 +1,7324 @@
+{


Using sampling with temperature=0.2 and top_p=0.95 would result in better generations.
def get_outputs(model, inputs, max_new_tokens=100): outputs = model.generate( input_ids=inputs["input_ids"], attention_mask=inputs["attention_mask"], max_new_tokens=max_new_tokens, temperature=0.2, top_p=0.95, do_sample=True, repetition_penalty=1.2, #Avoid repetition. early_stopping=True, #The model can stop before reach the max_length eos_token_id=tokenizer.eos_token_id ) return outputs

Reply via ReviewNB

Thank you very much! I was aiming for a response as consistent as possible to ensure the explanation in the notebook consistently aligns with the generated answer. With these values, a different response is obtained in each execution.

Would it be acceptable if I leave it as is, with these parameters commented, indicating that for better but more varied responses, users can uncomment and adjust the parameters?

I think it's ok to leave as is, and add this context in markdown before the code. It's always great to see the reasoning behind parameter choices, and alternative options.

pacman100 · 2024-02-23T09:22:06Z

notebooks/en/Prompt_Tuning_PEFT.ipynb

@@ -0,0 +1,7324 @@
+{


"I want you to act as an English translator," -> "I want you to act as a motivational coach. "

Reply via ReviewNB

ups... modified!

pacman100 · 2024-02-23T09:22:06Z

notebooks/en/Prompt_Tuning_PEFT.ipynb

@@ -0,0 +1,7324 @@
+{


Instead of this, you can do the below which would avoid loading a second copy of the base model in memory.
loaded_model_prompt.load_adapter(output_directory_sentences, adapter_name="quotes") loaded_model_prompt.set_adapter("quotes")

Reply via ReviewNB

Amazing! Thaks I changed the code!

pacman100 · 2024-02-23T09:22:06Z

notebooks/en/Prompt_Tuning_PEFT.ipynb

@@ -0,0 +1,7324 @@
+{


this generation is nsfw.

Reply via ReviewNB

I changed the input sentence to avoid this generation.

Can you also update the output?

Yes! For sure, I'm just executing the notebook again, and checking the responses obtained with different configurations. I will upload the notification in few minutes.

This is the generation: 'There are two nice things that should matter to you: the weather and your health.' now.

I checked with different configurations and all responses are sfw :-)

pacman100

Thank you @peremartra for the insightful notebook on prompt tuning! 🔥

Left few comments.

MKhalusova · 2024-02-23T17:17:33Z

notebooks/en/Prompt_Tuning_PEFT.ipynb

@@ -0,0 +1,7324 @@
+{


Peft->PEFT

Reply via ReviewNB

MKhalusova · 2024-02-23T17:17:33Z

notebooks/en/Prompt_Tuning_PEFT.ipynb

@@ -0,0 +1,7324 @@
+{


Let's remove the output of this cell

Reply via ReviewNB

MKhalusova · 2024-02-23T17:17:33Z

notebooks/en/Prompt_Tuning_PEFT.ipynb

@@ -0,0 +1,7324 @@
+{


Same here, let's remove the output

Reply via ReviewNB

MKhalusova · 2024-02-23T17:17:33Z

notebooks/en/Prompt_Tuning_PEFT.ipynb

@@ -0,0 +1,7324 @@
+{


Again, we can remove the output of this cell

Reply via ReviewNB

MKhalusova · 2024-02-23T17:17:33Z

notebooks/en/Prompt_Tuning_PEFT.ipynb

@@ -0,0 +1,7324 @@
+{


I would suggest placing the last line (display(train_sample_prompt)) into a separate cell, so that we could show it's output but remove the output of all the downloads

Reply via ReviewNB

MKhalusova · 2024-02-23T17:17:33Z

notebooks/en/Prompt_Tuning_PEFT.ipynb

@@ -0,0 +1,7324 @@
+{


Same here, if you place the last line in a separate cell, we can show the output of the display, and remove the output with downloading

Reply via ReviewNB

MKhalusova · 2024-02-23T17:17:33Z

notebooks/en/Prompt_Tuning_PEFT.ipynb

@@ -0,0 +1,7324 @@
+{


Fine-tuning (capitalization)

Reply via ReviewNB

MKhalusova · 2024-02-23T17:18:54Z

I left a few comments to polish things, and we can aim to publish the notebook early next week :)

HuggingFaceDocBuilderDev · 2024-02-23T17:21:40Z

The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update.

peremartra · 2024-02-23T18:35:09Z

Thanks for your corrections! I think all is fixed now.

MKhalusova

Great work! Let's merge :)

peremartra added 2 commits February 18, 2024 12:57

Demonstrates how to perform prompt tuning using the PEFT library.

562e6d7

Prompt Tuning With PEF - Modified Image Directory to use it from coob…

025bd73

…ook-images dataset

MKhalusova requested a review from pacman100 February 19, 2024 19:39

Modified _toctree.yml, index.md and added author to notebook

ad69cde

I also fixed the link to the other notebooks on index.md, all of them are missing the .ipynb and the links were broken.

First Version QLoRA notebook

d1b3ac1

Creado con Colaboratory

08285d2

MKhalusova and others added 4 commits February 22, 2024 08:21

Merge branch 'main' into main

b746d20

Delete notebooks/en/QLoRA_Tuning_PEFT.ipynb

5ad53b9

Update index.md

579954c

Merge branch 'main' into main

2596743

pacman100 reviewed Feb 23, 2024

View reviewed changes

MKhalusova reviewed Feb 23, 2024

View reviewed changes

Implementation of proposed corrections.

944d67e

Rename Prompt_Tuning_PEFT.ipynb to prompt_tuning_peft.ipynb

67ab8ae

MKhalusova approved these changes Feb 26, 2024

View reviewed changes

MKhalusova merged commit 8e662a7 into huggingface:main Feb 26, 2024
1 check passed

New Notebook showcasing how to perform prompt tuning using the PEFT library. #33

New Notebook showcasing how to perform prompt tuning using the PEFT library. #33

Conversation

peremartra commented Feb 18, 2024

What does this PR do?

review-notebook-app bot commented Feb 18, 2024

MKhalusova commented Feb 19, 2024

MKhalusova commented Feb 19, 2024 • edited Loading

peremartra commented Feb 19, 2024

peremartra commented Feb 21, 2024

MKhalusova commented Feb 22, 2024

pacman100 Feb 23, 2024 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

pacman100 Feb 23, 2024 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

pacman100 Feb 23, 2024 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

pacman100 Feb 23, 2024 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

pacman100 Feb 23, 2024 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

pacman100 left a comment

Choose a reason for hiding this comment

MKhalusova Feb 23, 2024 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

MKhalusova Feb 23, 2024 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

MKhalusova Feb 23, 2024 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

MKhalusova Feb 23, 2024 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

MKhalusova Feb 23, 2024 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

MKhalusova Feb 23, 2024 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

MKhalusova Feb 23, 2024 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

MKhalusova commented Feb 23, 2024

HuggingFaceDocBuilderDev commented Feb 23, 2024

peremartra commented Feb 23, 2024

MKhalusova left a comment

Choose a reason for hiding this comment

MKhalusova commented Feb 19, 2024 •

edited

Loading

pacman100 Feb 23, 2024 •

edited

Loading

pacman100 Feb 23, 2024 •

edited

Loading

pacman100 Feb 23, 2024 •

edited

Loading

pacman100 Feb 23, 2024 •

edited

Loading

pacman100 Feb 23, 2024 •

edited

Loading

MKhalusova Feb 23, 2024 •

edited

Loading

MKhalusova Feb 23, 2024 •

edited

Loading

MKhalusova Feb 23, 2024 •

edited

Loading

MKhalusova Feb 23, 2024 •

edited

Loading

MKhalusova Feb 23, 2024 •

edited

Loading

MKhalusova Feb 23, 2024 •

edited

Loading

MKhalusova Feb 23, 2024 •

edited

Loading