update interface

app.py

@@ -1,41 +1,38 @@
-import gradio as gr
-import numpy as np
-import pandas as pd
-from datetime import datetime
 import os
 import re
+from datetime import datetime
 
-from huggingface_hub import hf_hub_url, cached_download
+import gradio as gr
+import numpy as np
+import pandas as pd
 from gensim.models.fasttext import load_facebook_model
-
-ACCESS_KEY = os.environ.get('ACCESS_KEY')
+from huggingface_hub import hf_hub_download
 
 
-# Setup model
-url = hf_hub_url(repo_id="simonschoe/call2vec", filename="model.bin")
-cached_download(url)
-model = load_facebook_model(cached_download(url))
+ACCESS_KEY = os.environ.get('ACCESS_KEY')
+url = hf_hub_download(repo_id="simonschoe/call2vec", filename="model.bin") #hf_hub_url(repo_id="simonschoe/call2vec", filename="model.bin")
+model = load_facebook_model(url)
 
-def semantic_search(_input, n):
+def semantic_search(_input, neighbours):
     """ Perform semantic search """
 
     _input = re.split('[,;\n]', _input)
     _input = [s.strip().lower().replace(' ', '_') for s in _input if s]
 
     if _input[0] != ACCESS_KEY:
-        with open('log.txt', 'a') as f:
+        with open('log.txt', 'a', encoding='utf-8') as f:
             f.write(str(datetime.now()) + '+++' + '___'.join(_input) + '\n')
 
     if len(_input) > 1:
         avg_input = np.stack([model.wv[w] for w in _input], axis=0).mean(axis=0)
-        nearest_neighbours = model.wv.most_similar(positive=avg_input, topn=n)
+        nearest_neighbours = model.wv.most_similar(positive=avg_input, topn=neighbours)
         frequencies = [model.wv.get_vecattr(nn[0], 'count') for nn in nearest_neighbours]
     else:
-        nearest_neighbours = model.wv.most_similar(positive=_input[0], topn=n)
+        nearest_neighbours = model.wv.most_similar(positive=_input[0], topn=neighbours)
         frequencies = [model.wv.get_vecattr(nn[0], 'count') for nn in nearest_neighbours]
-    
+
     if _input[0] == ACCESS_KEY:
-        with open('log.txt', 'r') as f:
+        with open('log.txt', 'r', encoding='utf-8') as f:
             prompts = f.readlines()
             prompts = [p.strip().split('+++') for p in prompts]
         result = pd.DataFrame(prompts,
@@ -43,40 +40,28 @@ def semantic_search(_input, n):
     else:
         result = pd.DataFrame([(a[0],a[1],b) for a,b in zip(nearest_neighbours, frequencies)],
                               columns=['Token', 'Cosine Similarity', 'Corpus Frequency'])
-    
+
     result.to_csv('result.csv')
     return result, 'result.csv', '\n'.join(_input)
 
-app = gr.Blocks()
+app = gr.Blocks(theme=gr.themes.Default(), css='#component-0 {max-width: 730px; margin: auto; padding-top: 1.5rem}')
 
 with app:
-    gr.Markdown("# Call2Vec")
-    gr.Markdown("## Semantic Search in Quarterly Earnings Conference Calls")
-    with gr.Row():
-        with gr.Column():
+    gr.Markdown(
+        """
+        # Call2Vec
+        ## Semantic Search in Quarterly Earnings Conference Calls
+        """
+    )
+
+    with gr.Tabs() as tabs:
+        with gr.TabItem("🔍 Model", id=0):
             text_in = gr.Textbox(lines=1, placeholder="Insert text", label="Search Query")
             with gr.Row():
                 n = gr.Slider(value=50, minimum=5, maximum=250, step=5, label="Number of Neighbours")
-                compute_bt = gr.Button("Search")
+                btn = gr.Button("Search")
             df_out = gr.Dataframe(interactive=False)
             f_out = gr.File(interactive=False, label="Download")
-        with gr.Column():
-            gr.Markdown(
-                """
-                #### Project Description
-                Call2Vec is a [fastText](https://fasttext.cc/) word embedding model trained via [Gensim](https://radimrehurek.com/gensim/). It maps each token in the vocabulary into a dense, 300-dimensional vector space, designed for performing semantic search.
-                The model is trained on a large sample of quarterly earnings conference calls, held by U.S. firms during the 2006-2022 period. In particular, the training data is restriced to the (rather sponentous) executives' remarks of the Q&A section of the call. The data has been preprocessed prior to model training via stop word removal, lemmatization, named entity masking, and coocurrence modeling.
-                """
-            )
-            gr.Markdown(
-                """
-                #### App usage
-                The model is intented to be used for **semantic search**: It encodes the search query (entered in the textbox on the right) in a dense vector space and finds semantic neighbours, i.e., token which frequently occur within similar contexts in the underlying training data.
-                The model allows for two use cases:
-                1. *Single Search:* The input query consists of a single word. When provided a bi-, tri-, or even fourgram, the quality of the model output depends on the presence of the query token in the model's vocabulary. N-grams should be concated by an underscore (e.g., "machine_learning" or "artifical_intelligence").
-                2. *Multi Search:* The input query may consist of several words or n-grams, seperated by comma, semi-colon or newline. It then computes the average vector over all inputs and performs semantic search based on the average input token.
-                """
-            )
             gr.Examples(
                 examples = [
                     ["transformation", 20],
@@ -88,15 +73,32 @@ with app:
                 fn = semantic_search,
                 cache_examples=True
             )
-    gr.Markdown(
-        """
-        <p style="text-align: center;">
-            Call2Vec by X and Y
-            <br>
-            <img 'id="visitor-badge" alt="visitor badge" src="https://visitor-badge.glitch.me/badge?page_id=simonschoe.call2vec&left_color=green&right_color=blue" />
-        </p>
-        """
-    )
-    compute_bt.click(semantic_search, inputs=[text_in, n], outputs=[df_out, f_out, text_in])
+        with gr.TabItem("📝 Usage", id=1):
+            gr.Markdown(
+                """
+                #### App usage
+                The model is intended to be used for **semantic search**: It encodes the search query (entered in the textbox on the right) in a dense vector space and finds semantic neighbours, i.e., token which frequently occur within similar contexts in the underlying training data.
+                The model allows for two use cases:
+                1. *Single Search:* The input query consists of a single word. When provided a bi-, tri-, or even fourgram, the quality of the model output depends on the presence of the query token in the model's vocabulary. N-grams should be concated by an underscore (e.g., "machine_learning" or "artifical_intelligence").
+                2. *Multi Search:* The input query may consist of several words or n-grams, seperated by comma, semi-colon or newline. It then computes the average vector over all inputs and performs semantic search based on the average input token.
+                """
+            )
+        with gr.TabItem("📖 About", id=2):
+            gr.Markdown(
+                """
+                #### Project Description
+                Call2Vec is a [fastText](https://fasttext.cc/) word embedding model trained via [Gensim](https://radimrehurek.com/gensim/). It maps each token in the vocabulary into a dense, 300-dimensional vector space, designed for performing semantic search.
+                The model is trained on a large sample of quarterly earnings conference calls, held by U.S. firms during the 2006-2022 period. In particular, the training data is restriced to the (rather sponentous) executives' remarks of the Q&A section of the call. The data has been preprocessed prior to model training via stop word removal, lemmatization, named entity masking, and coocurrence modeling.
+                """
+            )
+
+    with gr.Accordion("📙 Citation", open=False):
+        citation_button = gr.Textbox(
+            value='Placeholder',
+            label='Copy to cite these results.',
+            show_copy_button=True
+        )
+
+    btn.click(semantic_search, inputs=[text_in, n], outputs=[df_out, f_out, text_in])
 
 app.launch()