hi @adcps
welcome to the community. I had a fun time interacting with your app and training a model on a csv sheet I fed it.
Nicely done!
1 Like
Hi everyone, here is my submission for the challenge!
Introducing ChartGPT, an application allowing you to visualize your own CSV files from a text prompt. It is built on top of the Python library ChartGPT of my creation. After adding your OpenAI API key and uploading your CSV file (with a header), you can write what you want to see in a descriptive way (bar chart where x is the job type and y is the average salary , filtered on the top 5 biggest average salaries and year 2023) or broader way (top 5 most paying jobs in 2023), and the model will do its best to understand your request and plot it using Plotly. You can also export the code generated by the model.
Try it out here!
Additional notes:
- No data is shared with OpenAI, only the header column names
- Built using Dash Mantine Components
- CSV file is cached
GitHub links:
6 Likes
Greetings!
I am thrilled to share the success of my latest endeavor at Right Information! As a data scientist and information designer, I had the opportunity to contribute to an impactful project called ‘AI-based gene expression profiling’ within our expanding AI department in biomedicine research. 
The project involved several key parts. First, I trained predictive models using gene expression datasets for various cancer types. Next, I analyzed these models to identify the most crucial genes. Lastly, I integrated AI-driven scientific literature analysis to extract essential insights, which greatly aided our domain experts. This step, in particular, was significant for the challenge. 
I crafted a series of plots that vividly present insights from the dataset and thousands of analyzed publications. One particularly engaging plot is showcased on our page, allowing interactive exploration and utilizing our contextual chatbot to provide deeper insights. 
Additional details about the study can be found on our page and in our published paper. 
Notable features of our web application include:
- Information Retrieval Animation: An animation with graphs showcasing gene regulation and correlations extracted from the literature. Each graph line represents a scientific article connecting two nodes, enabling users to retrieve information by clicking on a line.
- Enhanced GPT 3.5 Chatbot: Our chatbot, powered by an enhanced GPT 3.5 model, is a remarkable asset. We split research papers into smaller text chunks and calculated their semantic vectors using the ‘text-embedding-ada-002’ model. All this valuable data resides in the Chroma database, allowing the chatbot to respond with the most relevant content when users inquire about a specific publication. We reduced model hallucination using the novel Langchain library, though we faced some challenges working with this cutting-edge tool.
- Knowledge Base Summarization Table: To provide comprehensive problem-related insights, we generated a table of questions and answers based on GPT 3.5 analysis of 10k research papers.
You can explore my work on the GitHub repository, where you’ll find the code and additional details in the readme: GitHub Repo 
Discover the web application I submitted for the challenge here: Genomic Analysis Web App
I hope my project inspires fellow enthusiasts to explore the potential of Dash and similar technologies in AI-driven research. Your feedback is invaluable to me, as I am always striving to learn and improve. Though I am relatively new to this framework, any guidance would be highly appreciated. 
Major problems I couldn’t overcome:
- Asynchronous callbacks
- Handling long callbacks in a multipage setup
Thank you for taking the time to engage with my work. I look forward to hearing from you and contributing further to the growing community. 
Best regards,
Antoni Dąbrowski
Data Scientist and Information Designer
at Right Information 
4 Likes
Hello everyone,
Our take at this challenge - PropagandaBot 
. This app uses chatGPT API to discover propaganda techniques in a given text. It can also generate propaganda articles based on user-provided topics.
Unfortunately, we had some troubles with deployment so as of now the app is not available online yet. We will try to fix this asap. Until then, you can run the app locally on your machine using docker.
The app: propagandabot.online:8000
GitHub repo: https://github.com/vsisl/dash-chatgpt-challenge
GIF demo: https://github.com/vsisl/dash-chatgpt-challenge/blob/main/demo.gif
3 Likes
Hello everyone
Let the prompts in the background. Introducing https://preguntagpt.com/ - Your Ultimate Survival Kit for Non-Coders!
Tired of being dependent on others for your tech-related tasks? Do you wish to have a versatile app that empowers you to handle various digital tasks without the need for coding knowledge? Look no further! PreguntaGPT is the progressive developing tool you’ve been waiting for.
PreguntaGPT is an all-in-one application that provides you with a wide array of hands-on tools to simplify your digital life. Whether you’re a professional seeking to draft proper emails, a student aiming to improve your grammar, or just someone who needs a little help organizing and planning, PreguntaGPT has got your back!
Here’s a glimpse of what PreguntaGPT has to offer:
- Draft Proper Emails: Say goodbye to email anxiety with PreguntaGPT’s intuitive email drafting feature. Craft compelling and well-structured emails effortlessly.
- Check Your Grammar: Avoid embarrassing mistakes and ensure your writing is top-notch with the grammar-checking tool. Receive instant suggestions to enhance your writing style.
- Text Improver: Take your writing to the next level. PreguntaGPT helps you enhance the clarity and coherence of your text, making your content stand out.
- Planner Assistant: Stay organized and on top of your tasks with the planner assistant. Never miss a deadline or appointment again.
- YouTube Transcriber: Extract text from YouTube videos for note-taking or transcription purposes. Easily access valuable information from video content.
- Audio Meeting Summarizer: Busy with back-to-back meetings? Let PreguntaGPT summarize the key points and action items from your audio meetings, saving you time and effort.
- Chat with URL Site: Interact with websites and gather information using natural language. PreguntaGPT makes browsing and researching a breeze.
- Explainer of Complex Stuff: Confused by complicated concepts? PreguntaGPT simplifies complex ideas and provides easy-to-understand explanations.
- Location Awareness Planner Developer: Plan your day efficiently based on your location and schedule. PreguntaGPT optimizes your tasks to suit your environment.
…and much more!
PreguntaGPT’s user-friendly interface and advanced AI capabilities make it the go-to survival kit for non-coders. Empowering you to handle various digital tasks with ease, PreguntaGPT ensures that you no longer have to rely on others for tech-related assistance.
Step into the world of seamless productivity and unlock your full potential with PreguntaGPT today!
2 Likes
Great submissions so far! Kudos to everyone for the creativity.
For my app, I used Fast Dash, a Dash-based framework I built, along with Embedchain to build an app to chat over various documents (PDFs, web pages, YouTube videos, and any input text).
- The app is hosted here.
- Source code: GitHub - dkedar7/embedchain-fastdash
- Colab demo: Google Colab
3 Likes
Thank you everyone for all these amazing Dash apps. The app submission period is now over.
A group of Plotly judges will be reviewing the apps over the next 2 weeks. And the challenge top 3 winners will be announced at the next Dash Club newsletter published shortly thereafter.
For those interested in reading more about Dash Club, you can use this link.
Thank you @AntoniDabrowski for the app submission. Is this part of your company?
Never mind, just saw your submission above. Great work. It’s always exciting to see Dash apps in the world of medicine and research.
Beautiful app, @dkedar7 . And I remember when I first discovered FastDash – definitely makes it easier to build web apps.
I like how this app summarizes information from specific sources, as opposed to reading general content on the web. Is that the main reason you built this app?
hi @marcoV
Thank you for your app submission. I tried registering a new account with username adam but after doing that, it wouldn’t allow me to log in. Do you know what might be happening? I also created adamtest but couldn’t log in with that one either.
hi @sislvacl
Thank you for the gif demo and sharing your code open source. I’ll try to run it using the GitHub code, but if you manage to deploy it, that would be great. Let me know.
1 Like
hi @youplala
Great app. I can see how this can be helpful, especially if data sets are big, making them hard to visualize certain trends in the data. How did you get the idea to build this type of app?
hi @ceyhun
I cloned your app and ran it with my own openai api key, but I get this error:
And when I type the text in the chat, I can’t see it.
An easy communication with GPT-4 or 3.5 (first UI tab)
Hi @LeRobert
I tried using the text proofreader in the first Answering tab, but I get this error with my api key.
: OpenAI API error: The model gpt-4 does not exist or you do not have access to it. Learn more: How can I access GPT-4? | OpenAI Help Center.
Hi @adamschroeder,
I didn’t faced this RateLimitError, but I think you can change the model name to text-davinci-002. This is not a app error normally.
Hi Adam
Is all good now, tiny bug in checking credentials 
please try again
thanks
Marco
Thanks @adamschroeder ! That’s correct. My intention is to get ChatGPT to return more relevant and specific answers. Without that relevant context, ChatGPT has either zero or massively outdated knowledge of a narrow, rapdily-evolving subject.
Hi @adamschroeder ,
Sorry for the late answer, I was on vacation. GPT-4 access via API needs to be granted by OpenAI. At first they had a waitlist and on July 6 they gave access to all API users that spent at least 1 USD. Others still do not have access. Is this maybe the case with your API key?
I just tested with my key again and the app works fine with models 3.5 and 4.