Update predictors.py

predictors.py

@@ -60,16 +60,16 @@ text_bc_model = BetterTransformer.transform(text_bc_model)
 text_mc_model = BetterTransformer.transform(text_mc_model)
 quillbot_model = BetterTransformer.transform(quillbot_model)
 
-bias_model_checker = AutoModelForSequenceClassification.from_pretrained(bias_checker_model_name)
-tokenizer = AutoTokenizer.from_pretrained(bias_checker_model_name)
-bias_model_checker = BetterTransformer.transform(bias_model_checker, keep_original_model=False)
-bias_checker = pipeline(
-    "text-classification",
-    model=bias_checker_model_name,
-    tokenizer=bias_checker_model_name,
-)
-gc.collect()
-bias_corrector = pipeline( "text2text-generation", model=bias_corrector_model_name, accelerator="ort")
+# bias_model_checker = AutoModelForSequenceClassification.from_pretrained(bias_checker_model_name)
+# tokenizer = AutoTokenizer.from_pretrained(bias_checker_model_name)
+# bias_model_checker = BetterTransformer.transform(bias_model_checker, keep_original_model=False)
+# bias_checker = pipeline(
+#     "text-classification",
+#     model=bias_checker_model_name,
+#     tokenizer=bias_checker_model_name,
+# )
+# gc.collect()
+# bias_corrector = pipeline( "text2text-generation", model=bias_corrector_model_name, accelerator="ort")
 
 # model score calibration
 iso_reg = joblib.load("isotonic_regression_model.joblib")
@@ -105,10 +105,11 @@ def update(text: str):
     return corrections_display
 
 def update_main(text: str):
-    text = clean(text, lower=False)
-    corrected_text, corrections = correct_text(text, bias_checker, bias_corrector)
-    corrections_display = "\n\n".join([f"Original: {orig}\nCorrected: {corr}" for orig, corr in corrections])
-    return corrected_text, corrections_display
+    # text = clean(text, lower=False)
+    # corrected_text, corrections = correct_text(text, bias_checker, bias_corrector)
+    # corrections_display = "\n\n".join([f"Original: {orig}\nCorrected: {corr}" for orig, corr in corrections])
+    # return corrected_text, corrections_display
+    return text, "Unavailable"
 
 def split_text(text: str) -> list:
     sentences = sent_tokenize(text)