DocProduct / Git / Diff of /README.md

Models:

philipB/

DocProduct

Downloads: 1

Diff of /README.md [d1fcb5] .. [0f6a60]

Switch to unified view


...

Collaboration between Santosh Gupta, Alex Sheng, and Junpeng Ye

Download trained models and embedding file [here](https://1drv.ms/u/s!An_n1-LB8-2dgoAYrXYnnBSA4d5dsg?e=i3mnFH).





Winner Top 6 Finalist of the ⚡#PoweredByTF 2.0 Challenge! https://devpost.com/software/nlp-doctor . Doc Product will be presented to the Tensorflow Engineering Team at Tensorflow Connect. Stay tuned for details. 

We wanted to use TensorFlow 2.0 to explore how well state-of-the-art natural language processing models like [BERT](https://arxiv.org/abs/1810.04805) and [GPT-2](https://openai.com/blog/better-language-models/) could respond to medical questions by retrieving and conditioning on relevant medical data, and this is the result.

## DISCLAIMER

	a/README.md		b/README.md
	...		...
4		4
5	Collaboration between Santosh Gupta, Alex Sheng, and Junpeng Ye	5	Collaboration between Santosh Gupta, Alex Sheng, and Junpeng Ye
6		6
7	Download trained models and embedding file [here](https://1drv.ms/u/s!An_n1-LB8-2dgoAYrXYnnBSA4d5dsg?e=i3mnFH).	7	Download trained models and embedding file [here](https://1drv.ms/u/s!An_n1-LB8-2dgoAYrXYnnBSA4d5dsg?e=i3mnFH).
8		8
9	<p align="center">
10	<img src="https://challengepost-s3-challengepost.netdna-ssl.com/photos/production/software_photos/000/806/964/datas/gallery.jpg?raw=true">
11	</p>
12
13	Winner Top 6 Finalist of the ⚡#PoweredByTF 2.0 Challenge! https://devpost.com/software/nlp-doctor . Doc Product will be presented to the Tensorflow Engineering Team at Tensorflow Connect. Stay tuned for details.	9	Winner Top 6 Finalist of the ⚡#PoweredByTF 2.0 Challenge! https://devpost.com/software/nlp-doctor . Doc Product will be presented to the Tensorflow Engineering Team at Tensorflow Connect. Stay tuned for details.
14		10
15	We wanted to use TensorFlow 2.0 to explore how well state-of-the-art natural language processing models like [BERT](https://arxiv.org/abs/1810.04805) and [GPT-2](https://openai.com/blog/better-language-models/) could respond to medical questions by retrieving and conditioning on relevant medical data, and this is the result.	11	We wanted to use TensorFlow 2.0 to explore how well state-of-the-art natural language processing models like [BERT](https://arxiv.org/abs/1810.04805) and [GPT-2](https://openai.com/blog/better-language-models/) could respond to medical questions by retrieving and conditioning on relevant medical data, and this is the result.
16		12
17	## DISCLAIMER	13	## DISCLAIMER