raina678

I'm encountering the same error: **CUDADriverError: CUDA_ERROR_INVALID_VALUE: invalid argument** when working with a large dataset (7.5 million). I also attempted to run UMAP and HDBSCAN on entire dataset and then...

Thank you for your reply! I am performing dimensionality reduction to 8 dimensions using UMAP. I am utilizing the ml.g5.2xlarge instance with the PyTorch 2.0.0 Python 3.10 GPU Optimized image....

Thank you for your response! A single iteration of UMAP using CPU implementation took about 4.5 hours. Would you recommend performing UMAP with GPU implementation followed by fast_hdbscan, and then...

Thank you so much again! I'll try GPU UMAP and fast_hdbscan! After checking, I found that fast_hdbscan does not support relative_validity_. Moreover, calculating metrics like DBCV and Davies-Bouldin for a...

Hello, I worked on a smaller dataset using GPU UMAP and Fast HDBSCAN, evaluating the results with the Davies-Bouldin metric. After performing Bayesian optimization, I found hyperparameters that resulted in...

Thank you for your reply! I'm working with around 1 million data points. My objective function constraints include limiting the cluster count to between 5 and 20, with outliers kept...

Hi, I used `cluster_extraction_method="leaf"` for a dataset of around 1 million data points and applied Bayesian optimization. However, the results were not satisfactory: - Cluster Distribution: Approximately 80-90% of the...