Update app.py

app.py

@@ -11,7 +11,7 @@ import csv
 import datetime
 
 from huggingface_hub import hf_hub_download
-file_path = hf_hub_download(repo_id="PierreHanna/TextRetrieval", repo_type="space", filename="encoder_text_retrievaltext_bmg_221022_54.h5",
+encoder_text_path = hf_hub_download(repo_id="PierreHanna/TextRetrieval", repo_type="space", filename="encoder_text_retrievaltext_bmg_221022_54.h5",
                             use_auth_token=os.environ['TOKEN'])
                                 
 #use_auth_token="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")
@@ -42,6 +42,8 @@ def make_bert_preprocess_model(sentence_features, tfhub_handle_preprocess, seq_l
   model_inputs = packer(truncated_segments)
   return tf.keras.Model(input_segments, model_inputs)
 
+python_path = hf_hub_download(repo_id="PierreHanna/TextRetrieval", repo_type="space", filename="models.py",
+                            use_auth_token=os.environ['TOKEN'])
 from models import *
 
 def process(prompt, lang):
@@ -76,8 +78,7 @@ def process(prompt, lang):
 
     # Embed text
     #from models import *
-    #encoder_text = tf.keras.models.load_model('encoder_text_retrievaltext_bmg_221022_54.h5')
-    encoder_text = tf.keras.models.load_model(file_path)
+    encoder_text = tf.keras.models.load_model(encoder_text_path)
     embed_query  = encoder_text.predict(embed_prompt["pooled_output"])
     faiss.normalize_L2(embed_query)
     print("   text embed computed.")