changing to translation

app.py

@@ -4,10 +4,10 @@ import openai
 from ui.app_theme import SoftBlue
 from utilities import constants,api_keys
 from prompts import ui_prompt_builder
-from chat_bot import ui_simple_chat
-''' Disabled 
-from image_gen import ui_image_generation
+#from chat_bot import ui_simple_chat 
+#from image_gen import ui_image_generation
 from summarization import ui_summarize
+''' Disabled#
 from recording_analysis import ui_recording_analysis 
 from slack_sentiment_analysis import ui_sentiment_analysis
 from image_to_text import image_to_text_translator as ui_image_to_text
@@ -16,16 +16,16 @@ from text_analysis import ui_text_analysis'''
 openai.api_key = api_keys.APIKeys().get_key('OPENAI_API_KEY')
 
 prompt_builder=ui_prompt_builder.ui
-usaa_advisor=ui_simple_chat.ui
-''' Disabled
+###usaa_advisor=ui_simple_chat.ui
+
 summarize=ui_summarize.ui
-image_generator=ui_image_generation.ui
+'''image_generator=ui_image_generation.ui
 video_analysis=ui_recording_analysis.ui
 slack_sentiment=ui_sentiment_analysis.ui
 #git_advisor=ui_simple_chat_git.ui
 #image_to_text=ui_image_to_text.ui
 #text_analysis=ui_text_analysis.ui'''
 
-ui = gr.TabbedInterface([prompt_builder, usaa_advisor], (constants.UI_1, "Policy Chat"),theme=SoftBlue())
+ui = gr.TabbedInterface([prompt_builder, summarize], (constants.UI_1, "Summarize"),theme=SoftBlue())
 
 ui.launch()

setup.sh

@@ -1,2 +0,0 @@
-# Install SpaCy modeldock
-python -m spacy download en_core_web_sm

summarization/summarization.py

@@ -7,15 +7,21 @@ def Completion(summary_messages):
         messages=summary_messages)
     return response["choices"][0]["message"]["content"]
 
-def SummarizeCompletion(code):
+def SummarizeCompletion(code, sample):
     truncated_prefix=""
     if len(code) > 14000:
         truncated_prefix=constants.SUMMARIZATION_TRUNCATION
     truncated_code = code[:14000]  
 
     summary_messages = [] 
-    summary_messages.append({"role": "system", "content": prompt_constants.SUMMARY_PROMPT})    
-    summary_messages.append({"role": "user", "content": truncated_code})  
+    if(sample!=""):
+        summary_messages.append({"role": "system", "content": prompt_constants.TRANSLATION_PROMPT_PREFIX+sample+prompt_constants.TRANSLATION_PROMPT_SUFFIX})    
+        summary_messages.append({"role": "user", "content": truncated_code})  
+        print("Sample not '': ",sample, " summary_messages: ",summary_messages)
+    else:
+        summary_messages.append({"role": "system", "content": prompt_constants.SUMMARY_PROMPT})    
+        summary_messages.append({"role": "user", "content": truncated_code})  
+        print("Sample: ",sample, " summary_messages: ",summary_messages)
 
     response_message = Completion(summary_messages)
     summary_messages.append({"role": "assistant", "content": response_message})   

summarization/ui_summarize.py

@@ -2,12 +2,16 @@
 import gradio as gr
 import pandas as pd
 import os
+
+from traitlets import default
 from utilities import constants
 from summarization import summarization
 from translators import translate_pdf_to_text
 
-def Summary(code):
-    sum_messages,sum_text = summarization.SummarizeCompletion(code)
+sample=""
+
+def Summary(code, samples):
+    sum_messages,sum_text = summarization.SummarizeCompletion(code,sample)
     df = pd.DataFrame(sum_messages) 
     return sum_text,df
 
@@ -15,6 +19,11 @@ def InitDF():
     global summaryDF
     summaryDF=pd.DataFrame({"role": [""], "content": [""] })
 
+def UpdateWithSample(newSample):
+    global sample
+    sample=newSample
+    print("Sample: ",sample)
+
 def UpdateWithExample(newExample):
     fileName=newExample.replace(" ","_").lower()
     if os.path.exists(f"{constants.TXT_PREFIX}{fileName}.txt"):
@@ -43,12 +52,16 @@ with gr.Blocks() as ui:
         examples = gr.Radio([constants.EX_1, constants.EX_2, constants.EX_3, constants.EX_4, constants.EX_5, 
                              constants.EX_6, constants.EX_7, constants.EX_8, constants.EX_9, constants.EX_10],
                              show_label=False, info=constants.SUMMARIZE_INFO)
+    gr.Markdown(constants.CODE_LANG_SAMPLES_MD)
+    with gr.Row():
+        samples = gr.Radio([constants.SAMPLE_1, constants.SAMPLE_2, constants.SAMPLE_3],
+                             show_label=False, info=constants.TRANSLATION_INFO)
     with gr.Row():
         summarize=gr.Button(value=constants.SUMMARIZE, variant="primary")
     with gr.Row():
         summaryDF=gr.DataFrame(type="pandas", value=pd.DataFrame({"role": [""], "content": [""] }),
                                wrap=True, show_label=False, label=constants.OPENAI_LOG)
-    summarize.click(Summary, inputs=[code], outputs=[summary,summaryDF])
+    summarize.click(Summary, inputs=[code,samples], outputs=[summary,summaryDF])
     examples.input(UpdateWithExample,inputs=[examples], outputs=[code])
-
+    samples.input(UpdateWithSample,inputs=[samples])
     InitDF()

utilities/constants.py

@@ -39,6 +39,10 @@ VOICE_7="Serena"
 VOICE_8="Sasha"
 VOICE_9="Adam"
 
+SAMPLE_1="C#"
+SAMPLE_2="Python"
+SAMPLE_3="Java"
+
 EX_1="Stored Procedure 1"
 EX_2="Stored Procedure 2"
 EX_3="COBOL 1"
@@ -150,7 +154,10 @@ NLP_POS_MAP_MD="### NLP POS Map"
 FILE_DOES_NOT_EXIST="does not exist."
 ORIGINAL_DOC="Original Document"
 SUMMARIZE_INFO="Pick a document to summarize."
+TRANSLATION_INFO="Pick a language to translate to."
+
 CODE_DOC_EXAMPLES_MD="### Code & Document Examples"
+CODE_LANG_SAMPLES_MD="### Target Language Options"
 
 
 IMAGE_PROMPT="Image Prompt"

utilities/data/csv/policy.csv

@@ -16,10 +16,15 @@ Vehicles 2 Year: 2022
 Vehicles 2 Vin: KNDCD3LD3N5536517
 Vehicles 2 Body Type: Minivan
 Vehicles 3 Make: Ford
-Vehicles 3 Model: Explorer
+Vehicles 3 Model: Fusion
 Vehicles 3 Year: 2012
-Vehicles 3 Vin: 1FM5K8D85CGB069944
-Vehicles 3 Body Type: SUV
+Vehicles 3 Vin: 3FAHP0HA6CR069944
+Vehicles 3 Body Type: Sedan
+Vehicles 4 Make: JRIDE
+Vehicles 4 Model: BHARE
+Vehicles 4 Year: 2024
+Vehicles 4 Vin: 09
+Vehicles 4 Body Type: GAP PROTECTION
 Coverages 0 Type: D
 Coverages 0 Limits Of Liability: COMPREHENSIVE LOSS ACV LESS |D 200} 115.97D 200] 165.74D 200
 Coverages 0 Deductible: None

utilities/data/pkl/policy.pkl

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:18a4d0028f1f0ece07307741e5f0d586a0f77932f8142c52395a71cc3a2a54cf
-size 444516
+oid sha256:4e25b77148be86b5d2fedae6813fba3ed98d87d3ed27017251ac1352d0178269
+size 513809

utilities/prompt_constants.py

@@ -15,7 +15,8 @@ ENGLISH_PROMPT="Provide responses entirely in English."
 
 #summarization preset
 SUMMARY_PROMPT="Identify the language (code) or genre (literature) as Type, summarize what this text does as simply as possible in the Summary. Identify key areas of interest in Key Concepts. Format response as follows:\nType: Python\nSummary: The code performs web scraping, text processing, tokenization, embedding generation, and question-answering tasks using OpenAI models and libraries.\nKey Concepts: Key concepts include the web scraping logic, text processing techniques, and the usage of OpenAI's embedding and question-answering models. Content: "
-
+TRANSLATION_PROMPT_PREFIX="You will translate code from source language into "
+TRANSLATION_PROMPT_SUFFIX=" language. Match data types, functions, and logic as closely as possible. Use comments to explain the code's purpose and functionality."
 #sentiment/emotion presets
 SENTIMENT_PROMPT="At the end of the response, Provide a sentiment evaluation of the user's input as \nSentiment: [Positive, Neutral, Negative]. Include '.' at the end of the sentence."
 EMOTION_PROMPT = "At the end of the response, Provide an emotion evaluation of the user's input as \nEmotion: [Happy, Angry, Sad, etc]. Include '.' at the end of the sentence."