Add max input length

app.py

@@ -12,9 +12,9 @@ from pipelines.keyphrase_generation_pipeline import KeyphraseGenerationPipeline
 @st.cache(allow_output_mutation=True, show_spinner=False)
 def load_pipeline(chosen_model):
     if "keyphrase-extraction" in chosen_model:
-        return KeyphraseExtractionPipeline(chosen_model)
+        return KeyphraseExtractionPipeline(chosen_model, truncation=True)
     elif "keyphrase-generation" in chosen_model:
-        return KeyphraseGenerationPipeline(chosen_model)
+        return KeyphraseGenerationPipeline(chosen_model, truncation=True)
 
 
 def extract_keyphrases():
@@ -159,7 +159,12 @@ with st.form("keyphrase-extraction-form"):
     )
 
     st.session_state.input_text = (
-        st.text_area("✍ Input", st.session_state.config.get("example_text"), height=250)
+        st.text_area(
+            "✍ Input",
+            st.session_state.config.get("example_text"),
+            height=250,
+            max_chars=2500,
+        )
         .replace("\n", " ")
         .strip()
     )

pipelines/keyphrase_extraction_pipeline.py

@@ -11,9 +11,7 @@ class KeyphraseExtractionPipeline(TokenClassificationPipeline):
     def __init__(self, model, *args, **kwargs):
         super().__init__(
             model=AutoModelForTokenClassification.from_pretrained(model),
-            tokenizer=AutoTokenizer.from_pretrained(
-                model, truncate=True
-            ),
+            tokenizer=AutoTokenizer.from_pretrained(model),
             *args,
             **kwargs
         )

pipelines/keyphrase_generation_pipeline.py

@@ -1,14 +1,17 @@
 import string
 
-from transformers import (AutoModelForSeq2SeqLM, AutoTokenizer,
-                          Text2TextGenerationPipeline)
+from transformers import (
+    AutoModelForSeq2SeqLM,
+    AutoTokenizer,
+    Text2TextGenerationPipeline,
+)
 
 
 class KeyphraseGenerationPipeline(Text2TextGenerationPipeline):
     def __init__(self, model, keyphrase_sep_token=";", *args, **kwargs):
         super().__init__(
             model=AutoModelForSeq2SeqLM.from_pretrained(model),
-            tokenizer=AutoTokenizer.from_pretrained(model, truncate=True),
+            tokenizer=AutoTokenizer.from_pretrained(model),
             *args,
             **kwargs
         )