dammy commited on
Commit
03e01d3
·
1 Parent(s): fad633c

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +5 -28
app.py CHANGED
@@ -37,8 +37,13 @@ def local_query(query, context):
37
  Context: {}
38
  Question: {}
39
  """.format(context, query)
 
 
 
40
 
41
  inputs = tokenizer(t5query, return_tensors="pt")
 
 
42
  outputs = model.generate(**inputs, max_new_tokens=20)
43
 
44
  return tokenizer.batch_decode(outputs, skip_special_tokens=True)
@@ -85,34 +90,6 @@ def run_query(file, history, query):
85
  return history, result
86
 
87
 
88
- # def load_document(pdf_filename):
89
-
90
-
91
- # loader = PDFMinerLoader(pdf_filename)
92
- # doc = loader.load()
93
-
94
- # text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=0)
95
- # texts = text_splitter.split_documents(doc)
96
-
97
- # texts = [i.page_content for i in texts]
98
-
99
- # doc_emb = st_model.encode(texts)
100
- # doc_emb = doc_emb.tolist()
101
-
102
- # ids = [str(uuid.uuid1()) for _ in doc_emb]
103
-
104
- # client = chromadb.Client()
105
- # collection = client.create_collection("test_db")
106
-
107
- # collection.add(
108
- # embeddings=doc_emb,
109
- # documents=texts,
110
- # ids=ids
111
- # )
112
-
113
- # return 'Success'
114
-
115
-
116
 
117
 
118
  def upload_pdf(file):
 
37
  Context: {}
38
  Question: {}
39
  """.format(context, query)
40
+
41
+ print('t5 query is')
42
+ primt(t5query)
43
 
44
  inputs = tokenizer(t5query, return_tensors="pt")
45
+
46
+ print('done with tokenizer')
47
  outputs = model.generate(**inputs, max_new_tokens=20)
48
 
49
  return tokenizer.batch_decode(outputs, skip_special_tokens=True)
 
90
  return history, result
91
 
92
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
93
 
94
 
95
  def upload_pdf(file):