FishNet: A Versatile Backbone for Image, Region, and Pixel Level Prediction

Authors: Shuyang Sun, Jiangmiao Pang, Jianping Shi, Shuai Yi, Wanli Ouyang

NeurIPS 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments have been conducted to demonstrate the remarkable performance of the Fish Net.
Researcher Affiliation Collaboration 1The University of Sydney 2Sense Time Research 3Zhejiang University
Pseudocode No The paper describes the network architecture and components using text and diagrams (Figure 2, Figure 3), but it does not include any explicitly labeled pseudocode or algorithm blocks.
Open Source Code Yes The code is available at https://github.com/kevin-ssy/Fish Net.
Open Datasets Yes For image classification, we evaluate our network on the Image Net 2012 classification dataset [25]
Dataset Splits Yes This dataset has 1.2 million images for training, and 50,000 images for validation (denoted by Image Net-1k val).
Hardware Specification No The paper mentions training on '16 GPUs' but does not provide specific hardware details such as GPU models, CPU models, or memory specifications.
Software Dependencies No The paper mentions using 'Py Torch [23]' but does not specify a version number for this or any other software dependency.
Experiment Setup Yes For training, we randomly crop the images into the resolution of 224 224 with batch size 256, and choose stochastic gradient descent (SGD) as the training optimizer with the base learning rate set to 0.1. The weight decay and momentum are 10 4 and 0.9 respectively. We train the network for 100 epochs, and the learning rate is decreased by 10 times every 30 epochs.