Update app.py

app.py

@@ -16,7 +16,9 @@ This space provides a gradio demo and an easy-to-run wrapper of the pre-trained
 
 Information about project you an fine on the website of [University of Oslo](https://www.mn.uio.no/ifi/english/research/projects/sant/)
 
-The model can be easily used for predicting sentiment as follows: 
+## How to do inference?
+
+Specify in config.json which model from saved_models you want to use. The model can be easily used for predicting sentiment as follows: 
 ```python
 >>> from sentiment_wrapper import PredictionModel
 >>> model = PredictionModel()
@@ -24,7 +26,9 @@ The model can be easily used for predicting sentiment as follows:
 [5,5]
 ```
 
-Any model can be fine-tuned. For this run fine-tune.py and specify required arguments:
+## How to fine-tune?
+
+For this run fine-tune.py and specify required arguments:
 -dataframe: pandas dataframe with columns ['text', 'label', 'split'] with 3 possible values in 'split' ['train','dev','test']
 -model: pre-traied model from huggingface or path to local folder with config.json in case you want to use custom wrapper
 <ul>