nlpjoe / Daguan Classify 2018
2018达观杯长文本分类智能处理挑战赛 18解决方案
Stars: ✭ 154
Programming Languages
python
139335 projects - #7 most used programming language
Labels
Projects that are alternatives of or similar to Daguan Classify 2018
Pyiron
pyiron - an integrated development environment (IDE) for computational materials science.
Stars: ✭ 153 (-0.65%)
Mutual labels: jupyter-notebook
Ipynb Quicklook
A Quick Look generator for Jupyter/IPython notebooks without further dependencies
Stars: ✭ 152 (-1.3%)
Mutual labels: jupyter-notebook
Self Learning
Books Papers, Courses & more I have to learn soon
Stars: ✭ 146 (-5.19%)
Mutual labels: jupyter-notebook
Python Clustering Exercises
Jupyter Notebook exercises for k-means clustering with Python 3 and scikit-learn
Stars: ✭ 153 (-0.65%)
Mutual labels: jupyter-notebook
Hdbscan
A high performance implementation of HDBSCAN clustering.
Stars: ✭ 2,032 (+1219.48%)
Mutual labels: jupyter-notebook
Suite2p
cell detection in calcium imaging recordings
Stars: ✭ 153 (-0.65%)
Mutual labels: jupyter-notebook
Robuststl
Unofficial Implementation of RobustSTL: A Robust Seasonal-Trend Decomposition Algorithm for Long Time Series (AAAI 2019)
Stars: ✭ 154 (+0%)
Mutual labels: jupyter-notebook
Oreilly Intro To Predictive Clv
Repo that contains the supporting material for O'Reilly Webinar "An Intro to Predictive Modeling for Customer Lifetime Value" on Feb 28, 2017
Stars: ✭ 153 (-0.65%)
Mutual labels: jupyter-notebook
Hamiltorch
PyTorch-based library for Riemannian Manifold Hamiltonian Monte Carlo (RMHMC) and inference in Bayesian neural networks
Stars: ✭ 153 (-0.65%)
Mutual labels: jupyter-notebook
Ios Coreml Yolo
Almost Real-time Object Detection using Apple's CoreML and YOLO v1 -
Stars: ✭ 153 (-0.65%)
Mutual labels: jupyter-notebook
Datasciencevm
Tools and Docs on the Azure Data Science Virtual Machine (http://aka.ms/dsvm)
Stars: ✭ 153 (-0.65%)
Mutual labels: jupyter-notebook
Natural Language Processing Specialization
This repo contains my coursework, assignments, and Slides for Natural Language Processing Specialization by deeplearning.ai on Coursera
Stars: ✭ 151 (-1.95%)
Mutual labels: jupyter-notebook
Python For Probability Statistics And Machine Learning 2e
Second edition of Springer Book Python for Probability, Statistics, and Machine Learning
Stars: ✭ 154 (+0%)
Mutual labels: jupyter-notebook
Predict Remaining Useful Life
Predict remaining useful life of a component based on historical sensor observations using automated feature engineering
Stars: ✭ 153 (-0.65%)
Mutual labels: jupyter-notebook
Python Data Visualization Course
Course materials for teaching data visualization in Python.
Stars: ✭ 153 (-0.65%)
Mutual labels: jupyter-notebook
Raster Vision Examples
Examples of using Raster Vision on open datasets
Stars: ✭ 154 (+0%)
Mutual labels: jupyter-notebook
Mastering Python For Finance Second Edition
Mastering Python for Finance – Second Edition, published by Packt
Stars: ✭ 153 (-0.65%)
Mutual labels: jupyter-notebook
Pytorch Question Answering
Important paper implementations for Question Answering using PyTorch
Stars: ✭ 154 (+0%)
Mutual labels: jupyter-notebook
达观杯2018
参数没调好,仓促比赛,单模型线上没测过,线下0.784,最终得分0.791,排名18/3462,排名不高就不多写了,等着前排分享。思路如同代码所写,很简单。
数据请在达观数据处下载,放在data目录下。
一、环境
环境/库 | 版本 |
---|---|
Ubuntu | 14.04.5 LTS |
python | 3.6 |
jupyter notebook | 4.2.3 |
tensorflow-gpu | 1.10.1 |
numpy | 1.14.1 |
pandas | 0.23.0 |
matplotlib | 2.2.2 |
gensim | 3.5.0 |
tqdm | 4.24.0 |
二、数据预处理
都写在jupyter
里了。运行src/preprocess/EDA.ipynb
生成各种文件。
三、baseline模型训练
在src/preprocess/
中运行:
python baseline-x-cv.py
四、深度模型训练
然后直接train模型,单GPU运行,模型自选:
python train_predict.py --gpu 4 --option 5 --model convlstm --feature char
多GPU训练示例:
python train_predict.py --gpu 4,5,6,7 --option 5 --model convlstm --feature char
五、模型融合输出
python stacking.py --gpu 1 --tfidf True --option 5
这里是stacking和伪标签一起做了,请修改代码自选是否用伪标签。
Contributors
This project exists thanks to all the people who contribute. [Contribute].
Backers
Thank you to all our backers! 🙏 [Become a backer]
Sponsors
Support this project by becoming a sponsor. Your logo will show up here with a link to your website. [Become a sponsor]
Note that the project description data, including the texts, logos, images, and/or trademarks,
for each open source project belongs to its rightful owner.
If you wish to add or remove any projects, please contact us at [email protected].