Accessing Remote Jupyter at USC ISI · cho.sh

Accessing Remote Jupyter at USC ISI · cho.sh

ssh -L 8080:localhost:8080 scho@bdnf.isi.edu

Check GPU Availability:
- In Jupyter, run:
  python
```
!nvidia-smi
```

Configure GPU in Notebook:

After loading packages, set GPU:

python

import os
import tensorflow as tf

os.environ["CUDA_DEVICE_ORDER"] = "PCI_BUS_ID"
os.environ["CUDA_VISIBLE_DEVICES"] = "0"  # Choose an available GPU

physical_devices = tf.config.list_physical_devices('GPU')
print("Num GPUs Available: ", len(physical_devices))

if physical_devices:
    tf.config.experimental.set_memory_growth(physical_devices[0], True)

Check GPU Availability:
- Same as for TensorFlow, use !nvidia-smi.

Configure GPU in Notebook:

PyTorch automatically uses available GPUs, but you can specify one:

python

import torch

# Check if CUDA is available
if torch.cuda.is_available():
    device = torch.device("cuda:0")  # Replace 0 with your GPU number
    print("Using GPU:", torch.cuda.get_device_name(0))
else:
    device = torch.device("cpu")
    print("Using CPU")

Use device to move tensors or models to the selected device:
python
```
model.to(device)
```