zs219274 Goto Github PK
Name: 吴潇潇
Type: User
Name: 吴潇潇
Type: User
AlexNet keras implementation
CTPN+CRNN bank card number identification(data/test pictures accuary ≈90%)
Classification of Environment Sound using CNN and ImageDataGenerator
Convolutional Neural Network for Text Classification in Tensorflow
Convolutional and LSTM networks to classify human activity
Notebooks for learning deep learning
TensorFlow examples
A python library built to empower developers to build applications and systems with self-contained Computer Vision capabilities
用keras实现OCR定位、识别
Simple TimeDistributed() wrapper Demo in Keras; sums images of MNIST digits
Updated 4 minutes ago This is an assignment from Planetly. The data is taken from Kaggle, two datasets: GlobalLandTemperaturesByCity.csv and GlobalLandTemperaturesByCountry.csv . https://www.kaggle.com/berkeleyearth/climate-change-earth-surface-temperature-data. The dataset contains the observations of the monthly average temperature of cities and of countries. Comments: The data, especially City dataset is quite big and full. In all calculations I used a decadal average. The data is more or less clean, especially after the 1900 year. Both achieves have around 3-4% of missing data. This data I replaced by the mean value calculated over the period 1900-2013. The data was averaged from daily values initially. Obviously that data wasn't manually reconstructed before. This is good, since it gives us the opportunity to work more on the quality control of the data and provide and clean data more carefully. Data cleaning: Nan's have been replaced by the average calculated for the period 1900-2013. It was my "straightforward" approach since there was no time to go deep into this problem. However, with the extra time, I could find other temperature data archives and provide a comparison analysis between the data. If data has all necessary values it might be considered for being for replacing. Data can be: observational, modelling and reconstructional. Thus, we would have a diverse choice of where and which data to take. With the extra time, I would go more away from decadal averaging and do monthly scaling per year or per 11/30 years. During 20th century there've been small and relatively large climate seasons with similar weather conditions. Relatively large (again, I'm talking only on century-scale) last around 30 years, small = 11 year. Thus, I would rather think about doing 11 years averaging or 30 years instead of 10 years. According to climatology, there are certain climatological periods like: 1950-1980, 1981-2010. During those periods, a relatively similar climate was observed. With more time I would do an outlier analysis, this actually can also help to determine the best averaging periods. I would also spend more time on the visualisation part, with providing spatial distribution (maps) of temperature. I have split data into 70/30 ration for obtaining training and testing data. I didn't perform a cross-validation set, however, It worth doing. For the modelling approach I have taken LSTM since I worked only on 4 cities and it calculated relatively quick, with good performance (according to baseline model). I employed a very simple baseline model based on averaging. Of course, since the modelling part is the most interesting, I would probably try to grab other meteorological parameters and would create/generate more features. Then I would probably try some regression approaches (log, lasso) and XGBoosting Regressor. Of course, I would experiment on baseline models as well.
For this project, we are going to detect rice leaf disease using CNN and serve the result via messenger chatbot. We will also implement this to an independent Android app.
Data Set for Rice Diseases with labels
riceleaf dataset
C++高性能分布式服务器框架,webserver,websocket server,自定义tcp_server(包含日志模块,配置模块,线程模块,协程模块,协程调度模块,io协程调度模块,hook模块,socket模块,bytearray序列化,http模块,TcpServer模块,Websocket模块,Https模块等, Smtp邮件模块, MySQL, SQLite3, ORM,Redis,Zookeeper)
用TensorFlow搭建CNN/RNN/LSTM/GRU/BiRNN/BiLSTM/BiGRU/Capsule Network等deep learning模型
TensorFlow2教程 TensorFlow 2.0 Tutorial 入门教程实战案例
CNN BiGRU ensembled-method
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.