Creating augmented suggester #54
Conversation
Signed-off-by: Grace Sng <[email protected]>
…sponding utils. Signed-off-by: Grace Sng <[email protected]>
Signed-off-by: Grace Sng <[email protected]>
Signed-off-by: Grace Sng <[email protected]>
Signed-off-by: Grace Sng <[email protected]>
Signed-off-by: Grace Sng <[email protected]>
Signed-off-by: Grace Sng <[email protected]>
Signed-off-by: Grace Sng <[email protected]>
Signed-off-by: Grace Sng <[email protected]>
Signed-off-by: Grace Sng <[email protected]>
Signed-off-by: Grace Sng <[email protected]>
There was a problem hiding this comment.
thanks for the PR @grace-sng7 I've shared a few inline comments.
I also notice that the build is failing. I'm working on a fix and will update this PR.
| { | ||
| "cell_type": "code", | ||
| "source": [ | ||
| "from pywhyllm.suggesters.augmented_model_suggester import AugmentedModelSuggester\n", |
There was a problem hiding this comment.
The notebook makes sense.
Can you add some description/text to the notebook so that readers can understand:
- Motivation: we are introducing this retrieval augmented suggester. It currently works on a specific cause pairs dataset.
- How it works: we fetch the pairs from that dataset.
- some documentation on the actual function call: can say users can give any two variable names, and also tell them what would happen if the pair was not found in the database.
|
|
||
|
|
||
| class AugmentedModelSuggester(SimpleModelSuggester): | ||
| def __init__(self, llm, file_path: str = 'data/causenet-precision.jsonl.bz2'): |
There was a problem hiding this comment.
needs docstrings for each method, so that future editors would be able to understand the code easily.
pywhyllm/utils/data_loader.py
Outdated
| International Conference on Information & Knowledge Management (CIKM '20). Association for | ||
| Computing Machinery, New York, NY, USA, 3023–3030. https://doi.org/10.1145/3340531.3412763 | ||
|
|
||
| TODO: Add license |
There was a problem hiding this comment.
yes, if license is known, please do add.
| import re | ||
|
|
||
| from .simple_model_suggester import SimpleModelSuggester | ||
| from pywhyllm.utils.data_loader import * |
There was a problem hiding this comment.
instead of *, it may be good to mention the three functions that are being imported from the data_loader module. This helps the reader understand which file a function comes from, since there are two files with import *
… dependencies. Signed-off-by: Grace Sng <[email protected]>
|
Hi Dr. Sharma, thanks for the comments. I've added a few updates to the PR. |
There was a problem hiding this comment.
thanks @grace-sng7 I just have one comment for the notebook.
| { | ||
| "cell_type": "markdown", | ||
| "source": [ | ||
| "Here we introduce the AugmentedModelSuggester class. Creating an instance of it enables the chosen LLM to utilize Retrieval Augmented Generation (RAG) to determine causality. It currently does this by searching the CauseNet dataset for a relevant causal pair and augmenting the LLM with the corresponding evidence/information stored in CauseNet." |
There was a problem hiding this comment.
it will be good to add a cite for causenet here
There was a problem hiding this comment.
it will be good to add a cite for causenet here
I have updated the notebook with a citation for CauseNet.
Signed-off-by: Grace Sng <[email protected]>
No description provided.