latest

app.py

@@ -1,5 +1,4 @@
 import streamlit as st
-import wandb
 from transformers import pipeline
 from transformers import AutoTokenizer, AutoModelForTokenClassification
 import pandas as pd
@@ -33,12 +32,16 @@ def load_random_examples(dataset_name, num_examples=5):
     dataset = load_dataset("surrey-nlp/PLOD-CW")
     
     # Convert the dataset to a pandas DataFrame
-    df = pd.DataFrame(dataset['train'])
+    df = pd.DataFrame(dataset['test'])
     
     # Select random examples
-    random_examples = df.sample(n=num_examples)
-    
-    return random_examples
+    random_examples = df.sample(n=1)
+
+    tokens = random_examples.tokens
+    ner_tags = random_examples.ner_tags
+
+    return pd.DataFrame((tokens, ner_tags))
+
 
 def render_entities(tokens, entities):
     """
@@ -65,15 +68,19 @@ def render_entities(tokens, entities):
         th {
             background-color: #4CAF50;
             color: white;
+            width: 16.66%;
         }
         tr:hover {
             background-color: #f5f5f5;
         }
+        td {
+            width: 16.66%;
+        }
         </style>
         """, unsafe_allow_html=True)
 
     # Title and description
-    st.title("Token Entities Table")
+    st.title("Model predicted Token vs Entities Table")
     st.write("This table shows the entity corresponding to each token in a cool and chilled theme.")
 
     # Create the table
@@ -85,8 +92,6 @@ def render_random_examples():
     Render random examples from the PLOD-CW dataset in a Streamlit table.
     """
     # Load random examples
-    random_examples = load_random_examples("surrey-nlp/PLOD-CW")
-    
     
     # Custom CSS for chilled and cool theme
     st.markdown("""
@@ -108,16 +113,20 @@ def render_random_examples():
         th {
             background-color: #4CAF50;
             color: white;
+            width: 16.66%;
         }
         tr:hover {
             background-color: #f5f5f5;
         }
+        td {
+            width: 16.66%;
+        }
         </style>
         """, unsafe_allow_html=True)
 
     # Title and description
     st.title("Random Examples from PLOD-CW")
-    st.write("This table shows 5 random examples from the PLOD-CW dataset in a cool and chilled theme.")
+    st.write("This table shows 1 random examples from the PLOD-CW dataset in a cool and chilled theme.")
 
     # Add a button to select a different set of random samples
     if st.button('Show another set of random examples'):
@@ -130,13 +139,12 @@ def render_random_examples():
     # Display the table
     st.table(st.session_state['random_examples'])
 
-    
 def prep_page():
     model = load_trained_model()
 
     # Streamlit app
     # Page configuration
-    st.set_page_config(page_title="NER Token Entities", layout="centered")
+    #st.set_page_config(page_title="NER Token Entities", layout="centered")
 
     st.title("Named Entity Recognition with BERT on PLOD-CW")
     st.write("Enter a sentence to see the named entities recognized by the model.")
@@ -185,7 +193,7 @@ def prep_page():
 
         render_entities(text, entities)
 
-        render_random_examples()
+    render_random_examples()