relatively low accuracy
Opened this issue · 1 comments
Thank you for your message. The file I uploaded contains 51 correct results, but it's worth noting that these results were obtained 3-4 years ago, and changes and updates in the libraries during this period may account for the discrepancies. Additionally, it's important to acknowledge the inherent randomness in machine learning applications. When I downloaded and ran the code, I obtained a result of 0.4852. Given the high number of classes and the complexity of classification tasks with numerous classes, such variation can be considered within the expected range of outcomes.
In this context, it's essential to understand that you haven't done anything wrong when executing the code. If you wish, you can experiment with different parameters and random seeds to explore the results further.