Table 2 The performance of our deep phenotyping system (CNN + LSTM) compared to other baseline methods (Using handcrafted features and SVM as a classifier, adding the LSTM to consider temporal information, CNN without temporal information and using CRF instead of LSTM to compare their performance)
From: Deep phenotyping: deep learning for temporal phenotype/genotype classification
| Â | Sf-2 | Cvi | Ler-1 | Col-0 | Avg. |
|---|---|---|---|---|---|
Hand-crafted features using SVM | 58.4 | 83.9 | 65.1 | 35.7 | 60.8 |
Hand-crafted features + LSTM | 61.3 | 90.2 | 68.2 | 52.4 | 68.0 |
CNN | 71.0 | 86.0 | 74.4 | 76.0 | 76.8 |
CNNÂ +Â CRF | 84.3 | 96.1 | 90.4 | 89.4 | 87.6 |
CNNÂ +Â LSTM | 89.6 | 93.8 | 94.2 | 94.2 | 93.0 |