Kaggle_COVID-19 Detection
Published:
- Who: the solo- & side-project
- When: June 2021 - Aug 2021 (completed)
- Where: Kaggle Image Detection Competition SIIM-FISABIO-RSNA COVID-19 Detection
- Dataset: Kaggle dataset
- What: Identify and localize COVID-19 abnormalities on chest radiographs. In particular, categorize the radiographs as negative for pneumonia or typical, indeterminate, or atypical for COVID-19.
- How: PyTorch. EfficientNet7(classification) + Yolo-v5(Detection). EfficientNet2(classification) + CascadeRCNN(Detection). EfficientNetv2 + DCN(FasterRCNN)
Learn: Image detection.
- Result_1: It’s done (Aug. 9, 2021). I finally ranked at 97/1324 (Top 8% - Bronze medal). I know it’s all my fault, however, the bronze medal was retracted and my account on this competition was deleted because I made two accounts on this competition(the 2nd account’s name was ‘Running_on_Datasets’ which was ranked at 100/1324), in order to overcome the computing power and submission limits. I admit I violated the competition rule which I was not aware of that.
- Result_2:
- Result_3:
