Apply dynamic quantization
You've successfully trained a neural network model for deployment, and now you want to optimize it using dynamic quantization. This step is crucial for deploying your model efficiently in environments with limited resources.
The model has been pre-loaded.
Deze oefening maakt deel uit van de cursus
Scalable AI Models with PyTorch Lightning
Oefeninstructies
- Import the necessary quantization module from PyTorch.
- Apply dynamic quantization targeting linear layers, using 8-bit integer precision.
Praktische interactieve oefening
Probeer deze oefening eens door deze voorbeeldcode in te vullen.
import torch
# Import the necessary quantization module
from torch.quantization import ____
# Apply dynamic quantization targeting linear layers
model_quantized = torch.quantization.____(
____, {torch.nn.____}, dtype=torch.____
)