awesome-deep-learning-finetune-experience
This repository not only contains experience about parameter finetune, but also other in-practice experience such as model ensemble (boosting, bagging and stacking) in Kaggle or other competitions.
Welcome pull request!
General
- My Neural Network isn't working! What should I do?
- williamFalcon/DeepRLHacks: Hacks for training RL systems from John Schulman's lesson at Deep RL Bootcamp (Aug 2017)
- A brief discussion on the code efficiency problem in deep learning
Course
- How to Win a Data Science Competition: Learn from Top Kagglers | Coursera
Stacking & Boosting & Ensemble
- Overview of Model Fusion Methods | Zhihu Column
- How to enter the top 10% in Kaggle's first game | Wille
- Introduction to Ensembling/Stacking in Python | Kaggle
- One article to understand integrated learning (with learning resources) | Data THU
Hyper-parameter search
- Scikit-learn hyperparameter search wrapper | skopt API documentation
GAN
- Generative adversarial network review: From architecture to training skills, it’s enough to read this paper | Heart of Machine [Original English]
Computer Vision
Image Classification
- [CVPR 2017] Squeeze-and-Excitation networks (ILSVRC 2017 winner) at CVPR2017
- Column | Momenta explains the 2017 championship architecture SENet
- [FAIR 2017] Accurate, Large Minibatch SGD: Training ImageNet in 1 Hour
- Kaggle Invasive Species Monitor competition: 3rd place solution overview
multi-label classification
- Kaggle Amazon Championship Interview: Using tag relevance to deal with classification issues | Lei Feng.com
Detection
- Tianchi Medical AI Competition Season 1 Yiyuan Intelligent_HKBU Team Rank7 Solution (including code) [code]
Movie Description
- [LSMDC 2016] 1st place: Video Description by Combining Strong Representation and a Simple Nearest Neighbor Approach. Gil Levi, Dotan Kaufman, Lior Wolf (Tel Aviv University), Tal Hassner (University of Southern California) [slides]
Movie Annotation and Retrieval
- [LSMDC 2016] 1st place: Video Captioning and Retrieval Models with Semantic Attention. YoungJae Yu, Hyungjin Ko, Jongwook Choi, Gunhee Kim (Seoul National University) [slides]
- [LSMDC 2016] 2st place: Video Description by Combining Strong Representation and a Simple Nearest Neighbor Approach. Gil Levi, Dotan Kaufman, Lior Wolf (Tel Aviv University), Tal Hassner (University of Southern California)
Movie Fill-in-the-Blank
[LSMDC 2016] 1st place: Video Captioning and Retrieval Models with Semantic Attention. YoungJae Yu, Hyungjin Ko, Jongwook Choi, Gunhee Kim (Seoul National University)
[LSMDC 2016] 2st place: Video Fill in the Blank with Merging LSTMs. Amir Mazaheri, Dong Zhang, Mubarak Shah (University of Central Florida)
Speech
- 10686 A CTC-RNN parameter adjustment experience | Zhihu column
Feature Engineering
Thousands of hammers and carve out the deep mountains: Let’s talk about the best practices of feature engineering | Big Data Miscellaneous
plantsgo/Rental-Listing-Inquiries: kaggle:Two Sigma Connect: Rental Listing Inquiries--top1
