testing LFTK with a sentence in code

app.py

@@ -14,24 +14,23 @@ vicuna_model = AutoModelForCausalLM.from_pretrained("lmsys/vicuna-7b-v1.3")
 llama_tokenizer = AutoTokenizer.from_pretrained("daryl149/llama-2-7b-chat-hf")
 llama_model = AutoModelForCausalLM.from_pretrained("daryl149/llama-2-7b-chat-hf")
 
-def linguistic_features(message):
+def linguistic_features_fn(message):
     # Load a trained spaCy pipeline
     nlp = spacy.load("en_core_web_sm")
 
     # Create a spaCy doc object
-    doc = nlp(message)
+    doc = nlp("I love research but my professor is strange.")
 
     # Initiate LFTK extractor by passing in the doc
-    LFTK_extractor = lftk.Extractor(docs=doc)
+    LFTK = lftk.Extractor(docs=doc)
 
     # Customize LFTK extractor (optional)
-    LFTK_extractor.customize(stop_words=True, punctuations=False, round_decimal=3)
+    LFTK.customize(stop_words=True, punctuations=False, round_decimal=3)
 
     # Use LFTK to dynamically extract handcrafted linguistic features
-    features_to_extract = lftk.search_features(family="wordsent", language="general", return_format="list_key")
-    extracted_features = LFTK_extractor.extract(features=features_to_extract)
+    extracted_features = LFTK.extract(features = ["a_word_ps", "a_kup_pw", "n_noun"])
 
-    print('Linguistic Features:', extracted_features)
+    print(extracted_features)
 
     return extracted_features
 
@@ -263,10 +262,7 @@ def interface():
             
             ling_ents_btn.click(fn=update_textbox, inputs=ling_ents_prompt, outputs=user_prompt_1, api_name="ling_ents_btn")
 
-            ling_ents_btn.click(lambda _, 
-                                message=ling_ents_prompt: linguistic_features_textbox.update(linguistic_features(ling_ents_prompt.value)),
-                                inputs=[ling_ents_prompt], 
-                                outputs=[linguistic_features_textbox])
+            ling_ents_btn.click(linguistic_features_fn, inputs=[ling_ents_prompt], outputs=[linguistic_features_textbox])
 
             # Event Handler for GPT 3.5 Chatbot
             ling_ents_btn.click(gpt_respond, inputs=[have_key2, linguistic_entities, ling_ents_prompt, gpt_ling_ents_chatbot],