Spaces:
Sleeping
Sleeping
import os | |
from typing import Iterator | |
import gradio as gr | |
from model import run | |
HF_PUBLIC = os.environ.get("HF_PUBLIC", False) | |
DEFAULT_SYSTEM_PROMPT = ''' | |
You are a digital assistant for John "LJ" Strenio's Data science portfolio page. Here are some key details about John to keep in mind with your response. | |
John's Resume: | |
John Strenio | |
(802)-734-6892 | |
[email protected] | |
JohnStrenio.com | GitHub | |
WORK EXPERIENCE | |
Scribd - Data Scientist (Jan 2022- Present) | |
- Improved Scribd’s SEO ranking by reducing the index life of 12% of newly uploaded documents at a | |
loss of only 1.2% of attributed signups solely utilizing document metadata collected upon upload | |
- productionized document quality model to perform inference on all newly uploaded documents, | |
processing ~500k docs a week | |
- Modified interaction-based recommendation system training data pipeline, improving user | |
recommendations in all recorded metrics with a projected CTR increase of 5.5% | |
- Identified 200k malicious user-generated documents containing personally identifiable information | |
(1% of corpus) and created a simple heuristic which removed 42k (21%) with a 70% precision rate | |
NASA - Software Engineering Intern (Aug 2019 - Dec 2019) | |
- Ported aircraft structural health monitoring system FOSS (Fiber Optic Sensor System) to cryogenic fuel | |
application using a microcontroller, decreasing program execution time by ~50% using a | |
multithreaded approach | |
Professional Skier (Winter 2007 - Winter 2016) | |
- Competed internationally in freestyle competitions winning an X-Games bronze medal and becoming a | |
finalist in the 2014 Olympic Qualifiers | |
- Coordinated and performed stunts for Vin Diesel in Paramount Pictures’ “The Return of Xander Cage” | |
garnering praise for the stunt team by the New York Times | |
SKILLS | |
Languages: (proficient) Python, SQL/Pyspark (past experience using) C, C++, JavaScript/HTML/CSS | |
Frameworks & Libraries: Pyspark, TensorFlow, Keras, PyTorch, Numpy, Matplotlib, Pandas, Scikit-learn, OpenCV, Huggingface, Verta, Airflow, MLflow | |
Software & Tools: Linux, Databricks, Windows, Git, Jupyter Notebook, Unity, Excel | |
EDUCATION | |
Portland State University, Portland, OR (Graduated Aug 2021) | |
(MS) Computer Science – AI/ML focus | |
GPA: 4.0 | |
Computer Science Grad Prep (Jun 2016 – Aug 2019) | |
University of Utah, Salt Lake City, UT (Graduated Aug 2012) | |
(BA) English Literature (BA) Film & Media Arts | |
John was also a professional freestyle skier for 15 years here are some of his accomplishments: | |
XGames bronze medalist (you can find his video on youtube) | |
he was in a number of ski films including warren miller, level 1 and meatheadfilms | |
he once stunt doubled for vin diesel in "the return of xander cage". | |
Remember you are a professional assistant and you would like to only discuss John and be helpful in answering question about him. | |
''' | |
MAX_MAX_NEW_TOKENS = 4096 | |
DEFAULT_MAX_NEW_TOKENS = 256 | |
MAX_INPUT_TOKEN_LENGTH = 4000 | |
DESCRIPTION = """ | |
# [Mistral-7B](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.1) | |
""" | |
def clear_and_save_textbox(message: str) -> tuple[str, str]: | |
return '', message | |
def display_input(message: str, | |
history: list[tuple[str, str]]) -> list[tuple[str, str]]: | |
history.append((message, '')) | |
return history | |
def delete_prev_fn( | |
history: list[tuple[str, str]]) -> tuple[list[tuple[str, str]], str]: | |
try: | |
message, _ = history.pop() | |
except IndexError: | |
message = '' | |
return history, message or '' | |
def generate( | |
message: str, | |
history_with_input: list[tuple[str, str]], | |
system_prompt: str, | |
max_new_tokens: int, | |
temperature: float, | |
top_p: float, | |
top_k: int, | |
) -> Iterator[list[tuple[str, str]]]: | |
if max_new_tokens > MAX_MAX_NEW_TOKENS: | |
raise ValueError | |
history = history_with_input[:-1] | |
generator = run(message, history, system_prompt, max_new_tokens, temperature, top_p, top_k) | |
try: | |
first_response = next(generator) | |
yield history + [(message, first_response)] | |
except StopIteration: | |
yield history + [(message, '')] | |
for response in generator: | |
yield history + [(message, response)] | |
def process_example(message: str) -> tuple[str, list[tuple[str, str]]]: | |
generator = generate(message, [], DEFAULT_SYSTEM_PROMPT, 1024, 1, 0.95, 50) | |
for x in generator: | |
pass | |
return '', x | |
def check_input_token_length(message: str, chat_history: list[tuple[str, str]], system_prompt: str) -> None: | |
input_token_length = len(message) + len(chat_history) | |
if input_token_length > MAX_INPUT_TOKEN_LENGTH: | |
raise gr.Error(f'The accumulated input is too long ({input_token_length} > {MAX_INPUT_TOKEN_LENGTH}). Clear your chat history and try again.') | |
with gr.Blocks(css='style.css') as demo: | |
gr.Markdown(DESCRIPTION) | |
gr.DuplicateButton(value='Duplicate Space for private use', | |
elem_id='duplicate-button') | |
with gr.Group(): | |
chatbot = gr.Chatbot(label='Playground') | |
with gr.Row(): | |
textbox = gr.Textbox( | |
container=False, | |
show_label=False, | |
placeholder='Hi, Mistral!', | |
scale=10, | |
) | |
submit_button = gr.Button('Submit', | |
variant='primary', | |
scale=1, | |
min_width=0) | |
with gr.Row(): | |
retry_button = gr.Button('🔄 Retry', variant='secondary') | |
undo_button = gr.Button('↩️ Undo', variant='secondary') | |
clear_button = gr.Button('🗑️ Clear', variant='secondary') | |
saved_input = gr.State() | |
with gr.Accordion(label='⚙️ Advanced options', open=False): | |
system_prompt = gr.Textbox(label='System prompt', | |
value=DEFAULT_SYSTEM_PROMPT, | |
lines=5, | |
interactive=False) | |
max_new_tokens = gr.Slider( | |
label='Max new tokens', | |
minimum=1, | |
maximum=MAX_MAX_NEW_TOKENS, | |
step=1, | |
value=DEFAULT_MAX_NEW_TOKENS, | |
) | |
temperature = gr.Slider( | |
label='Temperature', | |
minimum=0.1, | |
maximum=4.0, | |
step=0.1, | |
value=0.1, | |
) | |
top_p = gr.Slider( | |
label='Top-p (nucleus sampling)', | |
minimum=0.05, | |
maximum=1.0, | |
step=0.05, | |
value=0.9, | |
) | |
top_k = gr.Slider( | |
label='Top-k', | |
minimum=1, | |
maximum=1000, | |
step=1, | |
value=10, | |
) | |
textbox.submit( | |
fn=clear_and_save_textbox, | |
inputs=textbox, | |
outputs=[textbox, saved_input], | |
api_name=False, | |
queue=False, | |
).then( | |
fn=display_input, | |
inputs=[saved_input, chatbot], | |
outputs=chatbot, | |
api_name=False, | |
queue=False, | |
).then( | |
fn=check_input_token_length, | |
inputs=[saved_input, chatbot, system_prompt], | |
api_name=False, | |
queue=False, | |
).success( | |
fn=generate, | |
inputs=[ | |
saved_input, | |
chatbot, | |
system_prompt, | |
max_new_tokens, | |
temperature, | |
top_p, | |
top_k, | |
], | |
outputs=chatbot, | |
api_name=False, | |
) | |
button_event_preprocess = submit_button.click( | |
fn=clear_and_save_textbox, | |
inputs=textbox, | |
outputs=[textbox, saved_input], | |
api_name=False, | |
queue=False, | |
).then( | |
fn=display_input, | |
inputs=[saved_input, chatbot], | |
outputs=chatbot, | |
api_name=False, | |
queue=False, | |
).then( | |
fn=check_input_token_length, | |
inputs=[saved_input, chatbot, system_prompt], | |
api_name=False, | |
queue=False, | |
).success( | |
fn=generate, | |
inputs=[ | |
saved_input, | |
chatbot, | |
system_prompt, | |
max_new_tokens, | |
temperature, | |
top_p, | |
top_k, | |
], | |
outputs=chatbot, | |
api_name=False, | |
) | |
retry_button.click( | |
fn=delete_prev_fn, | |
inputs=chatbot, | |
outputs=[chatbot, saved_input], | |
api_name=False, | |
queue=False, | |
).then( | |
fn=display_input, | |
inputs=[saved_input, chatbot], | |
outputs=chatbot, | |
api_name=False, | |
queue=False, | |
).then( | |
fn=generate, | |
inputs=[ | |
saved_input, | |
chatbot, | |
system_prompt, | |
max_new_tokens, | |
temperature, | |
top_p, | |
top_k, | |
], | |
outputs=chatbot, | |
api_name=False, | |
) | |
undo_button.click( | |
fn=delete_prev_fn, | |
inputs=chatbot, | |
outputs=[chatbot, saved_input], | |
api_name=False, | |
queue=False, | |
).then( | |
fn=lambda x: x, | |
inputs=[saved_input], | |
outputs=textbox, | |
api_name=False, | |
queue=False, | |
) | |
clear_button.click( | |
fn=lambda: ([], ''), | |
outputs=[chatbot, saved_input], | |
queue=False, | |
api_name=False, | |
) | |
demo.queue(max_size=32).launch(share=HF_PUBLIC, show_api=False) | |