Updated app

app.py

@@ -5,8 +5,6 @@
 # Install necessary libraries
 from transformers import AutoTokenizer, AutoModelWithLMHead, pipeline
 import streamlit as st
-from pprint import pprint
-import json
 
 # Read the config
 with open("config.json") as f:
@@ -29,22 +27,22 @@ def load_model(model_name):
     return model, tokenizer
 
 # Side bar
-img = st.sidebar.image("images/tamil_logo.jpg", width=380)
+img = st.sidebar.image("images/tamil_logo.jpg", width=300)
 
 # Choose the model based on selection
 page = st.sidebar.selectbox("Model", config["models"])
 data = st.sidebar.selectbox("Data", config[page])
 
 # Main page
-st.header("Tamil Language Demos")
+st.title("Tamil Language Demos")
 st.markdown(
     "This demo uses [GPT2 trained on Oscar dataset](https://huggingface.co/flax-community/gpt-2-tamil) "
     "and [GPT2 trained on Oscar & Indic Corpus dataset] (https://huggingface.co/abinayam/gpt-2-tamil) "
-    "to show language generation"
+    "to show language generation!"
 )
 
 if page == 'Text Generation' and data == 'Oscar':
-    st.title('Tamil text generation with GPT2')
+    st.header('Tamil text generation with GPT2')
     st.markdown('A simple demo using gpt-2-tamil model trained on Oscar data')
     model, tokenizer = load_model(config[data])
     # Set default options
@@ -56,12 +54,12 @@ if page == 'Text Generation' and data == 'Oscar':
         try:
             with st.spinner('Generating...'):
                 generator = pipeline('text-generation', model=model, tokenizer=tokenizer)
-                seqs = generator(seed, max_length=max_len) # num_return_sequences=seq_num)
+                seqs = generator(seed, max_length=max_len)[0]['generated_text']# num_return_sequences=seq_num)
             st.write(seqs)
         except Exception as e:
             st.exception(f'Exception: {e}')
 elif page == 'Text Generation' and data == "Oscar + Indic Corpus":
-    st.title('Tamil text generation with GPT2')
+    st.header('Tamil text generation with GPT2')
     st.markdown('A simple demo using gpt-2-tamil model trained on Oscar data')
     model, tokenizer = load_model(config[data])
     # Set default options
@@ -73,7 +71,7 @@ elif page == 'Text Generation' and data == "Oscar + Indic Corpus":
         try:
             with st.spinner('Generating...'):
                 generator = pipeline('text-generation', model=model, tokenizer=tokenizer)
-                seqs = generator(seed, max_length=max_len) #num_return_sequences=seq_num)
+                seqs = generator(seed, max_length=max_len)[0]['generated_text'] #num_return_sequences=seq_num)
             st.write(seqs)
         except Exception as e:
             st.exception(f'Exception: {e}')

config.json

@@ -3,5 +3,8 @@
     "Text Generation": ["Oscar", "Oscar + Indic Corpus"],
     "Text Classification": ["News Data"],
     "Oscar": "flax-community/gpt-2-tamil",
-    "Oscar + Indic Corpus": "abinayam/gpt-2-tamil"
+    "Oscar + Indic Corpus": "abinayam/gpt-2-tamil",
+    "examples": ["ஒரு ஊரிலே ஒரு காக்கைக்கு",
+                "அன்பிர்க்கும் உன்டோ அடைக்கும்",
+                "தெனாலி ராமன், ஒரு பெரிய விகடகவி"]
 }