mariagrandury commited on
Commit
68aad16
·
1 Parent(s): 4c8df35

add section titles and rescale buttons

Browse files
Files changed (1) hide show
  1. app.py +17 -12
app.py CHANGED
@@ -208,17 +208,18 @@ def demo():
208
 
209
  gr.Markdown(
210
  """
211
- <center><h2>PDF-based chatbot</center></h2>
212
- <h3>Ask any questions about your PDF documents</h3>
213
- """
214
- )
215
- gr.Markdown(
216
- """<b>Note:</b> This AI assistant, using Langchain and open-source LLMs, performs retrieval-augmented generation (RAG) from your PDF documents. \
217
- This chatbot takes past questions into account when generating answers (via conversational memory), and includes document references for clarity purposes.<br>
218
  """
219
  )
 
 
 
 
 
220
 
221
  with gr.Tab("Chatbot configuration"):
 
222
  with gr.Row():
223
  document = gr.Files(
224
  height=100,
@@ -229,6 +230,7 @@ def demo():
229
  )
230
  # upload_btn = gr.UploadButton("Loading document...", height=100, file_count="multiple", file_types=["pdf"], scale=1)
231
 
 
232
  with gr.Row():
233
  with gr.Row():
234
  db_btn = gr.Radio(
@@ -261,13 +263,14 @@ def demo():
261
  info="Chunk overlap",
262
  interactive=True,
263
  )
264
- with gr.Row():
265
- db_btn = gr.Button("Generate vector database")
266
  with gr.Row():
267
  db_progress = gr.Textbox(
268
  label="Vector database initialization", value="0% Configure the DB"
269
  )
270
 
 
271
  with gr.Row():
272
  with gr.Row():
273
  llm_btn = gr.Radio(
@@ -308,11 +311,13 @@ def demo():
308
  info="Model top-k samples",
309
  interactive=True,
310
  )
311
- with gr.Row():
312
- qachain_btn = gr.Button("Initialize Question Answering chain")
 
 
313
  with gr.Row():
314
  llm_progress = gr.Textbox(
315
- value="None", label="0% Configure the QA chain"
316
  )
317
 
318
  with gr.Tab("Chatbot"):
 
208
 
209
  gr.Markdown(
210
  """
211
+ <center><h1>Chat with your PDF!</center></h1>
212
+ <center><h3>Ask any questions about your PDF documents</h3><center>
 
 
 
 
 
213
  """
214
  )
215
+ # gr.Markdown(
216
+ # """<b>Note:</b> This AI assistant, using Langchain and open-source LLMs, performs retrieval-augmented generation (RAG) from your PDF documents. \
217
+ # This chatbot takes past questions into account when generating answers (via conversational memory), and includes document references for clarity purposes.<br>
218
+ # """
219
+ # )
220
 
221
  with gr.Tab("Chatbot configuration"):
222
+ gr.Markdown("1. Upload the PDF(s)")
223
  with gr.Row():
224
  document = gr.Files(
225
  height=100,
 
230
  )
231
  # upload_btn = gr.UploadButton("Loading document...", height=100, file_count="multiple", file_types=["pdf"], scale=1)
232
 
233
+ gr.Markdown("2. Configure the vector database")
234
  with gr.Row():
235
  with gr.Row():
236
  db_btn = gr.Radio(
 
263
  info="Chunk overlap",
264
  interactive=True,
265
  )
266
+ with gr.Row():
267
+ db_btn = gr.Button("Generate vector database", size="sm")
268
  with gr.Row():
269
  db_progress = gr.Textbox(
270
  label="Vector database initialization", value="0% Configure the DB"
271
  )
272
 
273
+ gr.Markdown("3. Configure the LLM model")
274
  with gr.Row():
275
  with gr.Row():
276
  llm_btn = gr.Radio(
 
311
  info="Model top-k samples",
312
  interactive=True,
313
  )
314
+ with gr.Row():
315
+ qachain_btn = gr.Button(
316
+ "Initialize Question Answering chain", size="sm"
317
+ )
318
  with gr.Row():
319
  llm_progress = gr.Textbox(
320
+ value="QA chain initialization", label="0% Configure the QA chain"
321
  )
322
 
323
  with gr.Tab("Chatbot"):