Update README.md

README.md

@@ -1,19 +1,21 @@
-
-**This model was obtained by fine-tuning deepset/bert-base-cased-squad2 on Cord19 Dataset.**
-
-## Overview
-```
-Language model: deepset/bert-base-cased-squad2
-Language: English
-Downstream-task: Q&A
-Datasets: CORD-19 from 31rd January 2022
-Code: Haystack and FARM
-Infrastructure: Tesla T4
-```
-
-
+---
+language: en
+tags:
+- pytorch
+- question-answering
+datasets:
+- squad2
+- cord19
+metrics:
+- exact
+- f1
 widget:
 - text: "How can I protect myself against covid-19?"
+  context: "This paper explores how trust in formal information sources (government and media) and informal information sources (interpersonal) about COVID-19 influences compliance with preventive measures. This cross-sectional study uses convenience sampling of 478 adult participants. Data analyses using structural equation modeling with multigroup comparisons examine hypothesized relationships between trust in information sources and preventative behaviors and social distancing. Results suggest that understanding of COVID-19 causes is related to trust in formal information sources, but not to trust in informal information. Self-efficacy for prevention is related to trust in informal information sources, but not to trust in formal information sources. Worry about contracting COVID-19 is related to trust in formal information sources, but not to informal ones. Engaging in preventive measures is linked to both self-efficacy for prevention and worry, while social distancing was related only to worry. These findings have important implications for public health policy guidelines centered on clear and truthful media messages. The findings also facilitate comparative analyses of reactions to information sources across a decade of evolving attitudes toward media and government, between two cultures (Hong Kong vs. the USA), and between two different global pandemics."
+---
+
+## Model description 
+This model was obtained by fine-tuning deepset/bert-base-cased-squad2 on Cord19 Dataset.
 
 ## How to use
 
@@ -28,6 +30,15 @@ inputs = {
 nlp(inputs)
 ```
 
+## Overview
+```
+Language model: deepset/bert-base-cased-squad2
+Language: English
+Downstream-task: Q&A
+Datasets: CORD-19 from 31rd January 2022
+Code: Haystack and FARM
+Infrastructure: Tesla T4
+```
 
 ## Hyperparameters
 ```
@@ -37,4 +48,4 @@ max_seq_len = max_length
 learning_rate = AdamW: 1e-5
 ```
 
-  
+