Instructions to use Mandy234/myQAmodel with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Mandy234/myQAmodel with Transformers:
# Use a pipeline as a high-level helper # Warning: Pipeline type "question-answering" is no longer supported in transformers v5. # You must load the model directly (see below) or downgrade to v4.x with: # pip install "transformers<5.0.0" from transformers import pipeline pipe = pipeline("question-answering", model="Mandy234/myQAmodel")# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForQuestionAnswering tokenizer = AutoTokenizer.from_pretrained("Mandy234/myQAmodel") model = AutoModelForQuestionAnswering.from_pretrained("Mandy234/myQAmodel", device_map="auto") - Notebooks
- Google Colab
- Kaggle
|
Download README.md from Mandy234/myQAmodel: direct link, hf CLI and curl.
- Browser
- Download file 1.41 kB
-
https://huggingface.co/Mandy234/myQAmodel/resolve/main/README.md
- Command line
-
hf download hf://Mandy234/myQAmodel/README.md
-
curl -L -o README.md https://huggingface.co/Mandy234/myQAmodel/resolve/main/README.md
1.41 kB
| license: apache-2.0 | |
| Hey ! | |
| Well, you can use this model for the Q&A model- given a context and question corresponding from the context, this model can answer the question. | |
| Beware that this model can only answer closed-end questions. If you have given context based on the history of the USA from a Wikipedia page and asked the question | |
| 'Why the sky is Blue ?', chances will be that you wouldn't get the answer you expected. | |
| The exceptional thing is I fine-tuned this model with only 50 data points (out of 10000) due to RAM issues. Even though with such a small dataset, this | |
| is working exceptionally well with 98% accuracy! | |
| # Context | |
| You can input up to 30k words (That's equivalent to 100-page book content). But, it takes time to pre-process (expected!). | |
| # Question | |
| Ask the question which can be answered from the context only. This field can take upto 1000 words as input. | |
| # Answer | |
| This section provides you the answer from the above question. Note: Sometimes you may not get the right answer! | |
| # Score | |
| This number lies between 0 and 1, 1 being best possible answer, 0 being least possible answer. | |
| I used BERT model for this task. I've explored different methods and finally arrived at this model. This is pretty good one! | |
| PS: I'm working on solving Q&A for physics problems. If you are a big physics buff, please reach out to me to make this possible! (mail:manideepmaruthi.234@gmail.com) |