Upload app.py

app.py

@@ -20,7 +20,8 @@ with h5py.File('complete_artist_data.hdf5', 'r') as f:
     artist_names = [name.decode() for name in f['artist_names'][:]]
 
 def find_similar_artists(new_tags_string, top_n):
-    new_image_tags = [tag.strip() for tag in new_tags_string.split(",")]
+    #
+    new_image_tags = [tag.replace('_', ' ').strip() for tag in new_tags_string.split(",")]
     unseen_tags = set(new_image_tags) - set(vectorizer.vocabulary_.keys())
     unseen_tags_str = f'Unseen Tags: {", ".join(unseen_tags)}' if unseen_tags else 'No unseen tags.'
     
@@ -28,24 +29,24 @@ def find_similar_artists(new_tags_string, top_n):
     similarities = cosine_similarity(X_new_image, X_artist)[0]
     
     top_artist_indices = np.argsort(similarities)[-top_n:][::-1]
-    bottom_artist_indices = np.argsort(similarities)[:top_n]
-    
     top_artists = [(artist_names[i], similarities[i]) for i in top_artist_indices]
-    bottom_artists = [(artist_names[i], similarities[i]) for i in bottom_artist_indices]
     
-    top_artists_str = "\n".join([f"{rank+1}. {artist} - similarity score: {score:.4f}" for rank, (artist, score) in enumerate(top_artists)])
-    bottom_artists_str = "\n".join([f"{rank+1}. {artist} - similarity score: {score:.4f}" for rank, (artist, score) in enumerate(bottom_artists)])
+    top_artists_str = "\n".join([f"{rank+1}. {artist[3:]} ({score:.4f})" for rank, (artist, score) in enumerate(top_artists)])
+    dynamic_prompts_formatted_artists = "{" + "|".join([artist for artist, _ in top_artists]) + "}"
     
-    output_str = f"{unseen_tags_str}\n\nTop 10 artists:\n{top_artists_str}\n\nBottom 10 artists:\n{bottom_artists_str}"
-    return output_str
+    return unseen_tags_str, top_artists_str, dynamic_prompts_formatted_artists
 
 iface = gr.Interface(
     fn=find_similar_artists,
     inputs=[
-        gr.Textbox(label="Enter image tags", placeholder="fox, outside, detailed background"),
+        gr.Textbox(label="Enter image tags", placeholder="e.g. fox, outside, detailed background, ..."),
         gr.Slider(minimum=1, maximum=100, value=10, step=1, label="Number of artists")
     ],
-    outputs="text",
+    outputs=[
+        gr.Textbox(label="Unseen Tags", info="These tags are not used in the artist calculation. Even valid e6 tags may be \"unseen\" if they have insufficient data."), 
+        gr.Textbox(label="Top Artists", info="These are the artists most strongly associated with your tags.  The number in parenthes is a similarity score between 0 and 1, with higher numbers indicating greater similarity."),
+        gr.Textbox(label="Dynamic Prompts Format", info="For if you're using the Automatic1111 webui (https://github.com/AUTOMATIC1111/stable-diffusion-webui) with the Dynamic Prompts extension activated (https://github.com/adieyal/sd-dynamic-prompts) and want to try them all individually.") 
+    ],
     title="Tagset Completer",
     description="Enter a list of comma-separated e6 tags"
 )