'Loading dataset on CUDF (RAPIDS) smaller than GPU memory
I'm using a single NVIDIA T4 to load a big dataset using CUDF, the problem is that the dataset is only 4.8 GB and the GPU has 14 GB of free memory but when I try loading the dataset throught cudf.read_csv I receive a memory error:
MemoryError: std::bad_alloc: CUDA error at: /opt/conda/envs/rapids/include/rmm/mr/device/cuda_memory_resource.hpp
I know that I can use dask_cudf but i can't figure out why I can't load the dataset using only cudf.
Sources
This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.
Source: Stack Overflow
| Solution | Source |
|---|