stack images and calculate inference time .

[pk-1196]

Hi, can anybody help me with how to stack different image and get prediction on that stacked images and have inference time for the above prediction .

[glenn-jocher]

@pk-1196 see PyTorch Hub Tutorial below:

import torch

# Model
model = torch.hub.load('ultralytics/yolov5', 'yolov5s')

# Image
img = 'https://ultralytics.com/images/zidane.jpg'

# Inference
results = model(img)
results.print()


Using cache found in /root/.cache/torch/hub/ultralytics_yolov5_master
Fusing layers... 
Model Summary: 224 layers, 7266973 parameters, 0 gradients
Adding AutoShape... 
YOLOv5 🚀 2021-5-26 torch 1.8.1+cu101 CUDA:0 (Tesla P100-PCIE-16GB, 16280.875MB)

image 1/1: 720x1280 2 persons, 2 ties
Speed: 270.7ms pre-process, 12.8ms inference, 1.5ms NMS per image at shape (1, 3, 384, 640)

YOLOv5 Tutorials

• Train Custom Data  🚀 RECOMMENDED
• Tips for Best Training Results  ☘️ RECOMMENDED
• Weights & Biases Logging  🌟 NEW
• Supervisely Ecosystem  🌟 NEW
• Multi-GPU Training
• PyTorch Hub  ⭐ NEW
• TorchScript, ONNX, CoreML Export 🚀
• Test-Time Augmentation (TTA)
• Model Ensembling
• Model Pruning/Sparsity
• Hyperparameter Evolution
• Transfer Learning with Frozen Layers  ⭐ NEW
• TensorRT Deployment

[pk-1196]

@glenn-jocher thanks for yr quick response ...

I want to stack 2 images together so that input shape will be (2,3,640,640)
Not able to stack more than 1 image , can you tell where to make changes to stack images so that input will be as (2,3,640,640) if 2 images are stacked with 640*640 resolution

[glenn-jocher]

@pk-1196 see PyTorch Hub Tutorial:

YOLOv5 Tutorials

• Train Custom Data  🚀 RECOMMENDED
• Tips for Best Training Results  ☘️ RECOMMENDED
• Weights & Biases Logging  🌟 NEW
• Supervisely Ecosystem  🌟 NEW
• Multi-GPU Training
• PyTorch Hub  ⭐ NEW
• TorchScript, ONNX, CoreML Export 🚀
• Test-Time Augmentation (TTA)
• Model Ensembling
• Model Pruning/Sparsity
• Hyperparameter Evolution
• Transfer Learning with Frozen Layers  ⭐ NEW
• TensorRT Deployment