I’ve just launched a Streamlit app ( Insightful Data Explorer) that serves as a comprehensive solution for data analysis. This app provides tools for analysis, visualization, feature engineering, and AutoML tailored to your data needs.
Click the link below to explore it, and I’m looking forward to hearing your feedback!
Hi @shawngiese, thanks for the feedback. I have already provided 10 sample datasets for users who don’t have their own data to explore the app. I hope you are referring to the need for test data to evaluate a trained AutoML model using these sample datasets. Currently, the training and validation results are displayed, and testing with new data is optional. However, I believe adding this feature would be beneficial. I will prioritize its implementation. Thank you for bringing this up!
Update: A new feature has been added to the app: Chat with Data. Please feel free to explore the app, and I’m looking forward to hearing your feedback!
@Dini , Thank you! I’m glad you love the app. No, I’m not paying for the tokens used in the chat; I’m using Google Gemini-1.5-Flash-Latest, which is a free version.
Hello @Archana,
In my job i did a dashboard for analaysis, it look like we did the same architecture in the code
It’s a very nice job, i will spend more time to check everything ! It’s a nice idea you did with google-gemini ! It’s very straightforward, we can’t really go wrong, very easy to use
(i can’t do the same because of the data leaking :/)
My feedback for now : Change the base colors green/grey (it’s not making your app look great)
I have my own data, 600 product reviews, and asked for extractions based on specific criteria. I have a particular set of criteria I am seeking, and in the end, I am using the data to create product-specific buy personas in the buyer’s own words. I’ve tried numerous variations of code written by Openai, more times than I like to admit. I’ve also used other tools. I have never once come within miles of the output I got from your tool. Impressive!
I’ve also switched to Gemini and got better results than anything like the lies and bad code from Openai, but never anything like this on the first shot. Great job!