update readme

README.md

@@ -1,15 +1,14 @@
 ---
-title: cvchatbot
-emoji: 💬
-colorFrom: yellow
-colorTo: purple
+title: CV Chatbot
+emoji: 📝
+colorFrom: green
+colorTo: indigo
 sdk: gradio
 sdk_version: 6.3.0
 python_version: "3.12"
 app_file: app.py
+short_description: "RAG Chatbot with knowledge of your CV"
 pinned: false
-hf_oauth: true
-hf_oauth_scopes:
-- inference-api
+# hf_oauth: true
 ---
 

app.py

@@ -1,3 +1,4 @@
+import os
 import argparse
 import logging
 import gradio as gr
@@ -7,9 +8,13 @@ import spaces
 
 parser = argparse.ArgumentParser("cvchatbot")
 parser.add_argument("--debug", help="Enable debug logging", action='store_true')
+parser.add_argument("--prod", help="Use production environment", action='store_true')
 parser.add_argument("--generate", dest="generate_cache", help="Generate new cache", action='store_true')
 parser.add_argument("--download-cv-data", dest="download_cv_data", help="Download CV data from Cloud Storage", action='store_true')
 args = parser.parse_args()
+if args.prod:
+    os.environ["ENV"] = "PROD"
+
 logger = logging.getLogger(__name__)
 logging.basicConfig(level=logging.DEBUG if args.debug else logging.INFO)
 
@@ -54,8 +59,8 @@ def respond(
 
 with gr.Blocks() as demo:
     # login button
-    with gr.Sidebar():
-        gr.LoginButton()
+    # with gr.Sidebar():
+    #     gr.LoginButton()
 
     # chat context
     context_state = gr.State(ChatContext())

config.py

@@ -3,11 +3,20 @@ from string import Template
 
 
 HF_TOKEN = os.getenv('HF_TOKEN')
-# LLM_ID = "google/gemma-3-1b-it"
-LLM_ID = "google/gemma-3-4b-it"
+
+# model configs
+LLM_ID = "google/gemma-3-4b-it" if os.getenv("ENV") == "PROD" else "google/gemma-3-1b-it"
 EMBEDDING_MODEL_ID = "google/embeddinggemma-300M"
-PERSON_NAME = "William"
-PERSON_PRONOUN = "he"
+
+# embedding parameters
+DOMAIN_SELECTION_THRESHOLD = 0.5
+DOCUMENT_SELECTION_THRESHOLD = 0.2
+DOCUMENT_FOLLOWUP_THRESHOLD = 0.5
+
+# generation parameters
+RESPONSE_MAX_NEW_TOKENS = 512
+RESPONSE_SAMPLING_TOP_P = 0.95
+RESPONSE_SAMPLING_TEMPERATURE = 0.7
 
 # cv document data files
 CV_DATA_GCS_BUCKET = "wlee-jsons"
@@ -17,7 +26,12 @@ CV_DATA_FILE_JOBS = "jobs.json"
 CV_DATA_FILE_PAPERS = "papers.json"
 CV_DATA_FILE_SKILLS = "skills.json"
 RESET_QUERY = "restart chat"
-WELCOME_MESSAGE_TEMPLATE = Template(f"Hi! I'm William's Career Assistant. I can answer questions you might have regarding William's resume. "
+
+# text configs
+PERSON_NAME = "William"
+PERSON_PRONOUN = "he"
+PERSON_PRONOUN_POSSESSIVE = "his"
+WELCOME_MESSAGE_TEMPLATE = Template(f"Hi! I'm {PERSON_NAME}'s Career Assistant. I can answer questions you might have regarding {PERSON_PRONOUN_POSSESSIVE} resume. "
                    "To start, tell me which one of the following domains you would like to focus on:\n"
                    "$domains\n"
                    f"At any point in chat, say \'{RESET_QUERY}\' to reset chat and return to current state.")
@@ -26,29 +40,10 @@ EDUCATION_DESCRIPTION = "Academic degrees and relevant coursework"
 JOBS_DESCRIPTION = "Work experience"
 PAPERS_DESCRIPTION = "Papers and publications"
 SKILLS_DESCRIPTION = "Technical skills, such as programming language, framework, and tools"
-
-# cache files
-KNOWLEDGE_CACHE_FILE = "knowledge.pkl"
-
-# embedding parameters
-DOMAIN_SELECTION_THRESHOLD = 0.5
-DOCUMENT_SELECTION_THRESHOLD = 0.2
-DOCUMENT_FOLLOWUP_THRESHOLD = 0.5
-
-# generation parameters
-RESPONSE_MAX_NEW_TOKENS = 512
-RESPONSE_SAMPLING_TOP_P = 0.95
-RESPONSE_SAMPLING_TEMPERATURE = 0.7
-
-# helper configs
-SKILL_COMPETENCY_LEVELS = {
-    5: "Expert",
-    4: "Proficient",
-    3: "Competent",
-    2: "Advanced Beginner",
-    1: "Novice",
-}
 JOB_TYPES = [
     "Software Engineer",
     "Research Intern",
-]
+]
+
+# cache files
+KNOWLEDGE_CACHE_FILE = "knowledge.pkl"

helpers.py

@@ -4,8 +4,7 @@ import json
 from typing import Tuple
 from google.cloud.storage import Client, transfer_manager
 
-from config import JOB_TYPES, PERSON_NAME, PERSON_PRONOUN, RESET_QUERY, SKILL_COMPETENCY_LEVELS, \
-    CV_DATA_DIRNAME, CV_DATA_GCS_BUCKET
+from config import JOB_TYPES, PERSON_NAME, PERSON_PRONOUN, RESET_QUERY, CV_DATA_DIRNAME, CV_DATA_GCS_BUCKET
 
 
 def download_cv_data() -> None: