Working with Blobs
Pages 35
- Home
- AWS EC2 GPU enabled Caffe AMI
- Borrowing Weights from a Pretrained Network
- Caffe installing script for ubuntu 16.04 support Cuda 8
- Caffe on EC2 Ubuntu 14.04 Cuda 7
- Caffe Output: .caffemodel .solverstate
- Contributing
- Development
- Excluding Layers: Train and Test Phase
- Faster Caffe Training
- Fine Tuning or Training Certain Layers Exclusively
- GeForce GTX 1080, CUDA 8.0, Ubuntu 16.04, Caffe
- IDE Nvidia’s Eclipse Nsight
- Image Format: BGR not RGB
- Install Caffe on EC2 from scratch (Ubuntu, CUDA 7, cuDNN 3)
- Installation
- Installation (OSX)
- Making Prototxt Nets with Python
- Model Zo
- Model Zoo
- Models accuracy on ImageNet 2012 val
- OpenCV 3.2 Installation Guide on Ubuntu 16.04
- Python Layer Unit Tests
- Related Projects
- Reporting Bugs and Other Issues
- Simple Example: Sin Layer
- Solver Prototxt
- The Data Layer
- The Datum Object
- Training and Resuming
- Ubuntu 14.04 ec2 instance
- Ubuntu 14.04 VirtualBox VM
- Ubuntu 16.04 or 15.10 Installation Guide
- Using a Trained Network: Deploy
- Working with Blobs
- Show 20 more pages…
Clone this wiki locally
Blob
Blob is a variable dimension sized data structure.
The Common Way to Look At It
A Blob is generally used as a four dimensional ordered data structure: number, channels, height, and width.
The channels, height, and width usually describe a piece of data such as an image - RGB by Width by Height - but isn't limited to images. Throughout the network a Blob will often be various sizes as it may represent some high dimensional (large number of channels) set of feature vectors.
Working With the Blob
Getting Data from a Blob
A piece of data can be retrieved from the blob via the data_at(vector& index) or the data_at(int number, int channels, int height, int width) methods.
For example, the following code will check that the label from the first batch element, at the first channel, first height index, and first width index is equal to 1.
CHECK_EQ(blob_top_label_->data_at(0,0,0,0), 1);