Game push image / table /cv/nlp, multi-threaded start

• H&M Personalized Fashion Recommendations

The mission of the competition ： Provide product recommendations based on previous purchases
Match Links ：https://www.kaggle.com/c/h-and-m-personalized-fashion-recommendations

• Happywhale - Whale and Dolphin Identification

The mission of the competition ： Identify whales and dolphins based on features
Match Links ：https://www.kaggle.com/c/happy-whale-and-dolphin

• Ubiquant Market Prediction

The mission of the competition ： Make predictions based on future market data
Match Links ：https://www.kaggle.com/c/ubiquant-market-prediction

• Feedback Prize - Evaluating Student Writing

The mission of the competition ： extract 6-12 Writing elements of grade students
Match Links ：https://www.kaggle.com/c/feedback-prize-2021

• TensorFlow - Help Protect the Great Barrier Reef

The mission of the competition ： Detect crown of thorns starfish in underwater image data
Match Links ：https://www.kaggle.com/c/tensorflow-great-barrier-reef

• NBME - Score Clinical Patient Notes

The mission of the competition ： Identify key phrases in patient notes from the medical license exam
Match Links ：https://www.kaggle.com/c/nbme-score-clinical-patient-notes

• Energy big data sub track —— Virtual currency “ dig ” Behavior recognition

The mission of the competition ： Given the electricity consumption of enterprises in a certain region during holidays in recent two years 、 Monthly power consumption and other characteristic data , It is required to build relevant algorithm models according to the data provided by the samples , Identify and screen the suspected virtual currency among the sample users “ dig ” Behavioral users .
Match Links ：https://www.dcic-china.com/competitions/10025

• Energy big data sub track —— Research and judgment on over limit emission of pollution sources

The mission of the competition ： Given the daily energy consumption data of users in a sample area in recent one year 、 Peak valley characteristics and other characteristic data , It requires deep mining of energy consumption data , Build relevant algorithm model , Identify and screen users with over limit emission behavior in the sample , Realize intelligent problem diagnosis of energy using behavior , And draw scientific conclusions from the perspective of energy specialty , For decision making 、 Regulatory support .
Match Links ：https://www.dcic-china.com/competitions/10024

• Satellite application competition —— Intelligent detection of marine ships

The mission of the competition ： Detecting the vertical frame of a ship quickly and accurately is the basic requirement of intelligent ship detection
Match Links ：https://www.dcic-china.com/competitions/10022

• Smart agriculture competition —— Cattle image segmentation competition

The mission of the competition ： This competition takes the segmentation image data of cattle as the training sample , Contestants need to build models based on training samples , Carry out instance segmentation detection on the cattle images in the provided test set .
Match Links ：https://www.dcic-china.com/competitions/10021

• Technology finance sub track —— Transaction verification code recognition based on text characters

The mission of the competition ： This competition takes the example character verification code image data of marked character information as the training sample , Contestants need to build models based on the samples provided , Recognize the character verification code image in the test set , Extract valid character information . The training data set is not limited to the data provided , You can add public data sets .
Match Links ：https://www.dcic-china.com/competitions/10023

• Digital medical track —— Development of a new algorithm for accurately identifying human cell types

The mission of the competition ： In this task , We focus on the hierarchical classification of cell types , The box scRNA-seq The data predicts that each cell is in uHAF The cell type of each layer of the tree . The preliminary stage only involves general classification problems , That is, you only need to predict that the cell is uHAF Cell type of the first layer of the tree ; In the semi-finals, it is necessary to predict the cells in uHAF Cell types on all four layers of the tree .
Match Links ：https://www.dcic-china.com/competitions/10029