Base model: gpt-2 large
Epoch: 22
Loss: 0.1053
Dataset:covid-19 Article (https://github.com/ha-mulan/gpt2-FineTune-Tool/blob/master/dataset/cord19Article.txt)
it is the gpt2 large base model that generate covid article based on your input keyword
three ways to use Gpt2 covid article
1. CLI
2. Swagger
3. Demo
keyword: keyword of covid article you want to generate
generated text
curl -X GET "https://master-gpt2-covid-article-ha-mulan.endpoint.ainize.ai/api/?keyword=deadly" -H "accept: string"
deadly ill patients with confirmed COVID-19 by SARS-CoV-2 RT-PCR between March 1 and March 18, 2020.
Outcomes were followed up to March 25, 2020.
RESULTS: Among the 299 hospitalized patients with COVID-19, the median age was 59 years (interquartile range, 49-69; range, 20-91 years), and 138 (53.0%) were men.
Of the deceased patients, 64 (24.9%) had a history of chronic medical illness, including diabetes (54 [20.4%]), hypertension (40 [14.9%]), chronic cardiovascular disease (24 [9.8%]), chronic kidney disease (16 [ 5.0%]), and malignancy (16 [5.0%]).
The median time from hospital admission to death was 9 days (interquartile range, 7-12; range, 0-31).
At the time of hospital admission, dyspnea (55 [22.6%]), cough (40 [14.9%]), and fatigue (26 [10.7%]) were the most common symptoms.
The median time from illness onset to dyspnea was 7 days (interquartile range, 2-9; range, 0-13), and cough was present in 92.1% (152/299) of the deceased patients.
API page: Ainize
Demo page: End-point