Update app.py

app.py

@@ -1,12 +1,13 @@
 import gradio as gr
-from transformers import pipeline
-import pandas as pd
 from transformers import (
     AutoModelForSeq2SeqLM,
     AutoModelForTableQuestionAnswering,
     AutoTokenizer,
     pipeline,
+    TapexTokenizer, 
+    BartForConditionalGeneration
 )
+import pandas as pd
 
 # model_tapex = "microsoft/tapex-large-finetuned-wtq"
 # tokenizer_tapex = AutoTokenizer.from_pretrained(model_tapex)
@@ -16,7 +17,8 @@ from transformers import (
 # )
 
 #new
-pipe_tapex = pipeline(task="table-question-answering", model="microsoft/tapex-large-finetuned-wtq")
+tokenizer = TapexTokenizer.from_pretrained("microsoft/tapex-base")
+model = BartForConditionalGeneration.from_pretrained("microsoft/tapex-base")
 
 
 # model_tapas = "google/tapas-large-finetuned-wtq"
@@ -30,16 +32,16 @@ pipe_tapex = pipeline(task="table-question-answering", model="microsoft/tapex-la
 pipe_tapas = pipeline(task="table-question-answering", model="google/tapas-large-finetuned-wtq")
 
 
-def process(query, file, correct_answer, rows=20):
-    table = pd.read_csv(file.name, header=0).astype(str)
-    table = table[:rows]
-    result_tapex = pipe_tapex(table=table, query=query)
-    return result_tapex["answer"], result_tapas["answer"], correct_answer
+
 
 def process2(query, csv_data):
     csv_data={"Actors": ["Brad Pitt", "Leonardo Di Caprio", "George Clooney"], "Number of movies": ["87", "53", "69"]}
     table = pd.DataFrame.from_dict(csv_data)
-    result_tapex = pipe_tapex(table=table, query=query)['cells'][0]
+    #microsoft
+    encoding = tokenizer(table=table, query=query, return_tensors="pt")
+    outputs = model.generate(**encoding)
+    result_tapex=tokenizer.batch_decode(outputs, skip_special_tokens=True)
+    #google
     result_tapas = pipe_tapas(table=table, query=query)['cells'][0]
     return result_tapex, result_tapas