Spaces:

derek-thomas
/

arabic-RAG

Build error

App Files Files Community

derek-thomas HF staff commited on Nov 4, 2023

Commit

3cb8374

•

1 Parent(s): 9a4e478

Making updates to be more Arabic

Browse files

Files changed (3) hide show

app.py +9 -13
templates/template.j2 +2 -2
templates/template_html.j2 +9 -9

app.py CHANGED Viewed

@@ -1,5 +1,5 @@
-from functools import partial
 import logging
 from pathlib import Path
 from time import perf_counter
@@ -26,9 +26,9 @@ template_html = env.get_template('template_html.j2')
 tokenizer = AutoTokenizer.from_pretrained('derek-thomas/jais-13b-chat-hf')
 # Examples
-examples = ['What is the capital of China?',
-            'Why is the sky blue?',
-            'Who won the mens world cup in 2014?', ]
 def add_text(history, text):
@@ -45,10 +45,7 @@ def bot(history, hyde=False):
     # Retrieve documents relevant to query
     document_start = perf_counter()
     if hyde:
-        hyde_document = ""
-        generator = generate(f"Write a wikipedia article intro paragraph to answer this query: {query}")
-        for output_chunk in generator:
-            hyde_document = output_chunk
         logger.warning(hyde_document)
         documents = retriever(hyde_document, top_k=top_k)
@@ -57,14 +54,13 @@ def bot(history, hyde=False):
     document_time = perf_counter() - document_start
     logger.warning(f'Finished Retrieving documents in {round(document_time, 2)} seconds...')
     # Function to count tokens
     def count_tokens(text):
         return len(tokenizer.encode(text))
     # Create Prompt
     prompt = template.render(documents=documents, query=query)
     # Check if the prompt is too long
     token_count = count_tokens(prompt)
     while token_count > 2048:
@@ -119,7 +115,6 @@ with gr.Blocks() as demo:
         # Turn it back on
         txt_msg.then(lambda: gr.Textbox(interactive=True), None, [txt], queue=False)
     with gr.Tab("Arabic-RAG + HyDE"):
         hyde_chatbot = gr.Chatbot(
                 [],
@@ -143,7 +138,8 @@ with gr.Blocks() as demo:
         gr.Examples(examples, hyde_txt)
         hyde_prompt_html = gr.HTML()
         # Turn off interactivity while generating if you click
-        hyde_txt_msg = hyde_txt_btn.click(add_text, [hyde_chatbot, hyde_txt], [hyde_chatbot, hyde_txt], queue=False).then(
                 partial(bot, hyde=True), [hyde_chatbot], [hyde_chatbot, hyde_prompt_html])
         # Turn it back on

 import logging
+from functools import partial
 from pathlib import Path
 from time import perf_counter
 tokenizer = AutoTokenizer.from_pretrained('derek-thomas/jais-13b-chat-hf')
 # Examples
+examples = ['ما هي عاصمة الصين؟',
+            'لم السماء زرقاء؟',
+            "من فاز بكأس العالم للرجال في عام 2014؟",]
 def add_text(history, text):
     # Retrieve documents relevant to query
     document_start = perf_counter()
     if hyde:
+        hyde_document = generate(f"Write a wikipedia article intro paragraph to answer this query: {query}")[-1]
         logger.warning(hyde_document)
         documents = retriever(hyde_document, top_k=top_k)
     document_time = perf_counter() - document_start
     logger.warning(f'Finished Retrieving documents in {round(document_time, 2)} seconds...')
     # Function to count tokens
     def count_tokens(text):
         return len(tokenizer.encode(text))
     # Create Prompt
     prompt = template.render(documents=documents, query=query)
     # Check if the prompt is too long
     token_count = count_tokens(prompt)
     while token_count > 2048:
         # Turn it back on
         txt_msg.then(lambda: gr.Textbox(interactive=True), None, [txt], queue=False)
     with gr.Tab("Arabic-RAG + HyDE"):
         hyde_chatbot = gr.Chatbot(
                 [],
         gr.Examples(examples, hyde_txt)
         hyde_prompt_html = gr.HTML()
         # Turn off interactivity while generating if you click
+        hyde_txt_msg = hyde_txt_btn.click(add_text, [hyde_chatbot, hyde_txt], [hyde_chatbot, hyde_txt],
+                                          queue=False).then(
                 partial(bot, hyde=True), [hyde_chatbot], [hyde_chatbot, hyde_prompt_html])
         # Turn it back on

templates/template.j2 CHANGED Viewed

@@ -1,5 +1,5 @@
-### Instruction: Use the following unique documents in the Context section to answer the Query at the end. If you don't know the answer, just say that you don't know, don't try to make up an answer.
-### Context
 {% for doc in documents %}
 ---
     {{ doc.content }}

+### Instruction: استخدم المستندات الفريدة التالية في قسم السياق للإجابة على الاستعلام في النهاية. إذا كنت لا تعرف الإجابة، قل فقط أنك لا تعرف، ولا تحاول اختلاق إجابة.
+### Context
 {% for doc in documents %}
 ---
     {{ doc.content }}

templates/template_html.j2 CHANGED Viewed

@@ -70,25 +70,25 @@
 </head>
 <body>
 <div class="prose svelte-1ybaih5" id="component-6">
-    <h2>Prompt</h2>
-    Below is the prompt that is given to the model.
     <hr>
-    <h2>Instructions</h2>
-    <span class="instructions">Use the following pieces of context to answer the question at the end.<br>If you don't know the answer, just say that you don't know, <span
-            style="font-weight: bold;">don't try to make up an answer.</span></span><br>
-    <h2>Context</h2>
     {% for doc in documents %}
         <details class="doc-box" dir="rtl">
             <summary>
                 <b>وثيقة_{{ loop.index }} | <a href="{{ doc.meta.url }}">{{ doc.meta.title }}</a>:</b> <span
                     class="doc-short">{{ doc.content[:50] }}...</span>
             </summary>
-            <div class="doc-full"><a href="{{ doc.meta.url }}">{{ doc.meta.title }}</a>: {{ doc.content }}</div>
         </details>
     {% endfor %}
-    <h2>Query</h2>
-    <span class="query">{{ query }}</span>
 </div>
 <script>

 </head>
 <body>
 <div class="prose svelte-1ybaih5" id="component-6">
+    <h2>اِسْتَدْعَى</h2>
+    فيما يلي الموجه الذي تم تقديمه للنموذج.
     <hr>
+    <h2>تعليمات</h2>
+    <span class="instructions" dir="rtl">استخدم الأجزاء التالية من السياق للإجابة على السؤال في النهاية. <brإذا كنت لا تعرف الإجابة، فقط قل أنك لا تعرف،<span
+            style="font-weight: bold;">لا تحاول اختلاق إجابة.</span></span><br>
+    <h2>سياق</h2>
     {% for doc in documents %}
         <details class="doc-box" dir="rtl">
             <summary>
                 <b>وثيقة_{{ loop.index }} | <a href="{{ doc.meta.url }}">{{ doc.meta.title }}</a>:</b> <span
                     class="doc-short">{{ doc.content[:50] }}...</span>
             </summary>
+            <div class="doc-full">{{ doc.content }}</div>
         </details>
     {% endfor %}
+    <h2>استفسار</h2>
+    <span class="query" dir="rtl">{{ query }}</span>
 </div>
 <script>