James Wade is a Research Scientist working in the chemicals and materials science industry where he combines the power of data science with chemistry and materials science. His current projects focus on augmenting materials characterization innovations with statistical analysis, machine learning, and data visualization. James enjoys contributing to a growing data science community both in and outside of work and is a passionate advocate of open-source data science.
it works - i had the same issue at first, add a value/dollar amount of >=5 on your billing - you can't just add a credit card (the free credit thing is prolly from a while ago) i also made sure to uncheck 'autorefill credit' or whatever it's called
Great video, now I ca deploy my model as API Can you make a video like this for plumber API deployment to vercel app project? It would be helfpul since if I using huggingface the space must a public and poeple can access to my R code files.
Awesome video! Have you tried using the stream: True parameter in the openAI api via httr2's req_perform_stream()? I can get partial streamed results, but it always ends with an error. Curious if you had any run-ins with this. Thx
hey men 👋Excellent! could you set up a local docker to store the project images in huggingface. i have been looking for a long time and have not found anything. 💡
Hello James, I have a question. I see that you did EDA first, then split the data into train and test sets. Shouldn't I do EDA after the split to avoid data leakage?
Hi!! thank you for the beautiful video :)! I kindly have a question, I am trying the code on RStudio desktop and I keep getting the error " HTTP 429 Too Many Requests" for req_perform(). I currently have the free ChatGPT API bundle, still it shows 0 out of 18$ used, any thoughts on this please?
auc is an unreliable metric if classes are imbalanced; prediction probabilities need to be adjusted to "undo" the stratified sampling. you should keep a hold out set (randomly sampled) to verify the performance