Hyperspectral Image Classification on Pavia University (test)
100Overall Accuracy (OA)JigsawHSI
Evaluation results
| Method | Links | |||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| JigsawHSI2022.06 | 100 | — | — | — | — | — | — | — | — | — | 100 | 100 | — | — | — | — | — | — | — | — | — | |
| LGCNet2025.04 | 99.99 | 100 | 100 | 99.88 | 100 | 100 | 100 | 100 | 100 | 100 | 99.99 | 99.99 | — | — | — | — | — | — | — | — | — | |
| SG-DSCNet2025.04 | 99.99 | 99.95 | 100 | 100 | 100 | 100 | 100 | 100 | 100 | 100 | 99.99 | 99.99 | — | — | — | — | — | — | — | — | — | |
| LGCNet2025.04 | 99.99 | 100 | 100 | 99.88 | 100 | 100 | 100 | 100 | 100 | 100 | 99.99 | 99.99 | — | — | — | — | — | — | — | — | — | |
| KANet2025.04 | 99.99 | 100 | 100 | 100 | 100 | 100 | 99.97 | 100 | 100 | 100 | 99.99 | 99.99 | — | — | — | — | — | — | — | — | — | |
| MVNet2025.07 | 99.98 | 99.95 | 100 | 100 | 100 | 100 | 100 | 100 | 99.91 | 100 | 99.98 | 99.98 | — | — | — | — | — | — | — | — | — | |
| HybridSN2022.06 | 99.98 | — | — | — | — | — | — | — | — | — | 99.97 | 99.98 | — | — | — | — | — | — | — | — | — | |
| EKGNet2025.04 | 99.98 | 99.88 | 100 | 100 | 100 | 100 | 100 | 100 | 100 | 100 | 99.99 | 99.98 | — | — | — | — | — | — | — | — | — | |
| SCT-NetTesting Time(s)=8.02, Params(k)=167.492026.04 | 99.92 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| FreeNetProcessing framework=Patch-free, Training samples per class=2002020.11 | 99.81 | 99.58 | 99.88 | 99.95 | 99.27 | 100 | 100 | 100 | 99.83 | 100 | 99.83 | 99.74 | — | — | — | — | — | — | — | — | — | |
| 3D-CNN2025.04 | 99.79 | 99.96 | 99.99 | 99.64 | 99.83 | 99.81 | 99.98 | 97.97 | 99.56 | 100 | 99.75 | 99.87 | — | — | — | — | — | — | — | — | — | |
| 3D-CNN2025.04 | 99.79 | 99.96 | 99.99 | 99.64 | 99.83 | 99.81 | 99.98 | 97.97 | 99.56 | 100 | 99.75 | 99.87 | — | — | — | — | — | — | — | — | — | |
| SSFTTTesting Time(s)=7.24, Params(k)=148.492026.04 | 99.7 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| EdgeLDR-4#Params=81105, Size (MB)=0.34, GPU Latency (ms/batch)=11.11, CPU Latency (ms/batch)=125.112026.01 | 99.62 | — | — | — | — | — | — | — | — | — | 99.48 | 99.48 | — | — | — | — | — | — | — | — | — | |
| PHM-8#Params=99217, Size (MB)=0.42, GPU Latency (ms/batch)=3.81, CPU Latency (ms/batch)=80.542026.01 | 99.53 | — | — | — | — | — | — | — | — | — | 99.32 | 99.36 | — | — | — | — | — | — | — | — | — | |
| 3D-SE-DenseNet2025.04 | 99.48 | 99.32 | 99.87 | 96.76 | 99.23 | 99.64 | 99.8 | 99.47 | 99.32 | 100 | 99.16 | 99.31 | — | — | — | — | — | — | — | — | — | |
| 3D-SE-DenseNet2025.04 | 99.48 | 99.32 | 99.87 | 96.76 | 99.23 | 99.64 | 99.8 | 99.47 | 99.32 | 100 | 99.16 | 99.31 | — | — | — | — | — | — | — | — | — | |
| MASSFormerTesting Time(s)=10.73, Params(k)=314.412026.04 | 99.48 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| BC-8#Params=87505, Size (MB)=0.37, GPU Latency (ms/batch)=5.58, CPU Latency (ms/batch)=124.332026.01 | 99.41 | — | — | — | — | — | — | — | — | — | 99.21 | 99.19 | — | — | — | — | — | — | — | — | — | |
| EdgeLDR-2#Params=87505, Size (MB)=0.37, GPU Latency (ms/batch)=11.89, CPU Latency (ms/batch)=139.492026.01 | 99.41 | — | — | — | — | — | — | — | — | — | 99.24 | 99.19 | — | — | — | — | — | — | — | — | — | |
| Real#Params=177105, Size (MB)=0.72, GPU Latency (ms/batch)=2.41, CPU Latency (ms/batch)=75.652026.01 | 99.37 | — | — | — | — | — | — | — | — | — | 99.11 | 99.14 | — | — | — | — | — | — | — | — | — | |
| BC-4#Params=100305, Size (MB)=0.41, GPU Latency (ms/batch)=5.67, CPU Latency (ms/batch)=135.552026.01 | 99.37 | — | — | — | — | — | — | — | — | — | 99.26 | 99.14 | — | — | — | — | — | — | — | — | — | |
| Real quantized#Params=177105, Size (MB)=0.42, CPU Latency (ms/batch)=71.692026.01 | 99.35 | — | — | — | — | — | — | — | — | — | 99.1 | 99.11 | — | — | — | — | — | — | — | — | — | |
| PHM-4#Params=102609, Size (MB)=0.43, GPU Latency (ms/batch)=3.81, CPU Latency (ms/batch)=79.082026.01 | 99.32 | — | — | — | — | — | — | — | — | — | 99.17 | 99.08 | — | — | — | — | — | — | — | — | — | |
| BC-2#Params=125905, Size (MB)=0.51, GPU Latency (ms/batch)=5.61, CPU Latency (ms/batch)=177.992026.01 | 99.3 | — | — | — | — | — | — | — | — | — | 99.07 | 99.04 | — | — | — | — | — | — | — | — | — | |
| Real pruned 25%#Params=151361, Size (MB)=0.7, GPU Latency (ms/batch)=2.45, CPU Latency (ms/batch)=75.922026.01 | 99.19 | — | — | — | — | — | — | — | — | — | 99.03 | 98.88 | — | — | — | — | — | — | — | — | — | |
| PHM-2#Params=127033, Size (MB)=0.53, GPU Latency (ms/batch)=3.84, CPU Latency (ms/batch)=79.072026.01 | 99.15 | — | — | — | — | — | — | — | — | — | 98.95 | 98.84 | — | — | — | — | — | — | — | — | — | |
| Real pruned 50%#Params=125617, Size (MB)=0.7, GPU Latency (ms/batch)=2.43, CPU Latency (ms/batch)=75.692026.01 | 99.11 | — | — | — | — | — | — | — | — | — | 99.05 | 98.78 | — | — | — | — | — | — | — | — | — | |
| DML-CRFNumber of runs=5, Testing time (s)=8.682019.03 | 99.1 | — | — | — | — | — | — | — | — | — | 98.72 | 98.8 | — | — | — | — | — | — | — | — | — | |
| Quaternion#Params=100305, Size (MB)=0.43, GPU Latency (ms/batch)=5.09, CPU Latency (ms/batch)=78.082026.01 | 99.03 | — | — | — | — | — | — | — | — | — | 98.66 | 98.67 | — | — | — | — | — | — | — | — | — | |
| CSFFNumber of runs=5, Testing time (s)=1779.452019.03 | 98.9 | — | — | — | — | — | — | — | — | — | 98.49 | 98.52 | — | — | — | — | — | — | — | — | — | |
| NN-CRFNumber of runs=52019.03 | 98.78 | — | — | — | — | — | — | — | — | — | 97.53 | 9,838 | — | — | — | — | — | — | — | — | — | |
| DFFNProcessing framework=Patch-based, Training samples per class=2002020.11 | 98.57 | 99.53 | 97.71 | 99.89 | 97.88 | 99.48 | 99.69 | 100 | 98.59 | 99.61 | 99.16 | 98.08 | — | — | — | — | — | — | — | — | — | |
| HybridSNTesting Time(s)=7.29, Params(k)=797.572026.04 | 98.16 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| 3D-CNNNumber of runs=5, Testing time (s)=50.942019.03 | 98.14 | — | — | — | — | — | — | — | — | — | 97.32 | 96.87 | — | — | — | — | — | — | — | — | — | |
| S-CNNProcessing framework=Patch-based, Training samples per class=2002020.11 | 97.93 | 95.47 | 98.71 | 97.32 | 97.72 | 100 | 97.67 | 98.36 | 95.56 | 100 | 97.88 | 97.43 | — | — | — | — | — | — | — | — | — | |
| 3D-GANProcessing framework=Patch-based, Training samples per class=2002020.11 | 97.81 | 99.18 | 98.86 | 94.94 | 90.15 | 99.49 | 98.56 | 92.74 | 97.18 | 98.51 | 96.65 | 96.97 | — | — | — | — | — | — | — | — | — | |
| CSSMamba2026.01 | 97.55 | — | — | — | — | — | — | — | — | — | 97.51 | 98.75 | 97.16 | 97.99 | 90.43 | 97.45 | 100 | 99.01 | 99.76 | 95.88 | 99.88 | |
| BASS Net2016.12 | 97.48 | 97.71 | 97.93 | 94.95 | 97.8 | 100 | 96.6 | 98.14 | 95.46 | 100 | — | — | — | — | — | — | — | — | — | — | — | |
| Gabor-CNNProcessing framework=Patch-based, Training samples per class=2002020.11 | 97.33 | 99.53 | 98.21 | 89.74 | 93.02 | 99.42 | 98.77 | 98.82 | 94.12 | 97.91 | 96.62 | 96.62 | — | — | — | — | — | — | — | — | — | |
| DisentangleFormervariant=2-stage2025.12 | 96.51 | — | — | — | — | — | — | — | — | — | 95.81 | 95.3 | — | — | — | — | — | — | — | — | — | |
| PPF2016.12 | 96.48 | 97.42 | 95.76 | 94.05 | 97.52 | 100 | 99.13 | 96.19 | 93.62 | 99.6 | — | — | — | — | — | — | — | — | — | — | — | |
| GAHTTesting Time(s)=13.69, Params(k)=946.832026.04 | 96.4 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| CNN-MRFAlgorithm Category=Regularized classification algorithms2017.05 | 96.18 | 98.02 | 97.78 | 88.47 | 99.17 | 99.9 | 93 | 87.47 | 91.66 | 98.03 | 94.83 | 94.62 | — | — | — | — | — | — | — | — | — | |
| SS-MambaTesting Time(s)=10.41, Params(k)=470.02026.04 | 95.79 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| RPNet-RF2025.12 | 95.6 | — | — | — | — | — | — | — | — | — | 94.96 | 94.27 | — | — | — | — | — | — | — | — | — | |
| OursLabeled samples per class=102026.01 | 95.21 | 88.28 | 97.47 | 94.04 | 91.63 | 98.15 | 97.66 | 98.36 | 95.57 | 91.47 | 94.46 | 93.67 | — | — | — | — | — | — | — | — | — | |
| CNNAlgorithm Category=Classification algorithms2017.05 | 94.82 | 96.82 | 96.8 | 86.16 | 98.54 | 99.88 | 90.4 | 86.92 | 90.94 | 97.53 | 93.77 | 93.89 | — | — | — | — | — | — | — | — | — | |
| MambaHSI2026.01 | 94.69 | — | — | — | — | — | — | — | — | — | 96.21 | 98.75 | 92.94 | 93.57 | 96.84 | 90.57 | 100 | 99.27 | 97.67 | 95.33 | 99.69 | |
| SVM-3DDWT-GCAlgorithm Category=Regularized classification algorithms2017.05 | 94.12 | 92.03 | 97.33 | 86.89 | 97.09 | 98.7 | 98.12 | 98.84 | 84.46 | 100 | 94.83 | 93.55 | — | — | — | — | — | — | — | — | — | |
| ViT2026.01 | 93.91 | — | — | — | — | — | — | — | — | — | 93.15 | 95 | 81.79 | 98.22 | 94.94 | 82.7 | 97.47 | 99.01 | 98.75 | 92.94 | 92.5 | |
| DMLNumber of runs=52019.03 | 93.67 | — | — | — | — | — | — | — | — | — | 93.64 | 91.53 | — | — | — | — | — | — | — | — | — | |
| SpectralFormerTesting Time(s)=18.52, Params(k)=128.82026.04 | 93.17 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| SSRN2025.04 | 92.99 | 89.93 | 86.48 | 99.95 | 95.78 | 97.69 | 95.44 | 84.4 | 100 | 87.24 | 87.21 | 90.58 | — | — | — | — | — | — | — | — | — | |
| SSRN2025.04 | 92.99 | 89.93 | 86.48 | 99.95 | 95.78 | 97.69 | 95.44 | 84.4 | 100 | 87.24 | 87.21 | 90.58 | — | — | — | — | — | — | — | — | — | |
| RMAELabeled samples per class=102026.01 | 92.68 | 79.18 | 93.76 | 90.37 | 98.72 | 99.42 | 99.94 | 99.97 | 92.48 | 94.13 | 94.22 | 90.46 | — | — | — | — | — | — | — | — | — | |
| MLP2016.12 | 92.54 | 91.73 | 94.79 | 85.41 | 94.13 | 99.65 | 90.87 | 92.56 | 83.19 | 99.73 | — | — | — | — | — | — | — | — | — | — | — | |
| SwinT2025.12 | 92.41 | — | — | — | — | — | — | — | — | — | 93 | 89.75 | — | — | — | — | — | — | — | — | — | |
| CNN2016.12 | 92.27 | 88.38 | 91.27 | 85.88 | 97.24 | 99.91 | 96.41 | 93.62 | 87.45 | 99.57 | — | — | — | — | — | — | — | — | — | — | — | |
| 3D-CNNTesting Time(s)=6.39, Params(k)=462.4862026.04 | 92.01 | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | — | |
| Hit2025.07 | 92 | 96.19 | 92.79 | 93.21 | 97.33 | 99.96 | 99.91 | 98.22 | 99.15 | 99.77 | 93.24 | 89.77 | — | — | — | — | — | — | — | — | — | |
| SVM-3DDWTAlgorithm Category=Classification algorithms2017.05 | 91.27 | 84.78 | 93.62 | 85.04 | 95.9 | 98.77 | 94.29 | 96.59 | 79.82 | 100 | 92.09 | 88.55 | — | — | — | — | — | — | — | — | — | |
| MLRsubMLLAlgorithm Category=Regularized classification algorithms2017.05 | 91.13 | 87.07 | 96.97 | 77.27 | 83.9 | 99.54 | 99.4 | 94.5 | 64.83 | 99.78 | 89.25 | 88.19 | — | — | — | — | — | — | — | — | — | |
| SVM2016.12 | 91.1 | 87.95 | 91.17 | 86.99 | 95.5 | 99.85 | 94.31 | 94.74 | 85.89 | 99.89 | — | — | — | — | — | — | — | — | — | — | — | |
| Spectralformer2025.07 | 91.07 | 82.73 | 94.03 | 73.66 | 93.75 | 99.28 | 90.75 | 87.56 | 95.81 | 94.21 | 90.2 | 88.05 | — | — | — | — | — | — | — | — | — | |
| Spectralformer2025.04 | 91.07 | 82.73 | 94.03 | 73.66 | 93.75 | 99.28 | 90.75 | 87.56 | 95.81 | 94.21 | 90.2 | 88.05 | — | — | — | — | — | — | — | — | — | |
| Spectralformer2025.04 | 91.07 | 82.73 | 94.03 | 73.66 | 93.75 | 99.28 | 90.75 | 87.56 | 95.81 | 94.21 | 90.2 | 88.05 | — | — | — | — | — | — | — | — | — | |
| SpectralFormer2025.04 | 91.07 | 82.73 | 94.03 | 73.66 | 93.75 | 99.28 | 90.75 | 87.56 | 95.81 | 94.21 | 90.2 | 88.05 | — | — | — | — | — | — | — | — | — | |
| SpectraFormer2025.12 | 91.07 | — | — | — | — | — | — | — | — | — | 90.2 | 88.05 | — | — | — | — | — | — | — | — | — | |
| DEMAELabeled samples per class=102026.01 | 90.98 | 87.03 | 90.98 | 83.32 | 94.55 | 99.75 | 95.33 | 99.2 | 85.53 | 98.09 | 92.64 | 88.32 | — | — | — | — | — | — | — | — | — | |
| SVMProcessing framework=Patch-based, Training samples per class=2002020.11 | 90.78 | 85.49 | 92.12 | 85.77 | 96.41 | 98.6 | 92.52 | 93.79 | 86.56 | 97.97 | 92.14 | 88.13 | — | — | — | — | — | — | — | — | — | |
| ELM2016.12 | 89.86 | 81.32 | 90.91 | 85.09 | 96.61 | 99.63 | 94.33 | 95.94 | 82.65 | 99.79 | — | — | — | — | — | — | — | — | — | — | — | |
| SSFTT2026.01 | 88.81 | — | — | — | — | — | — | — | — | — | 89.6 | 84 | 90.53 | 95.56 | 94.27 | 94.31 | 99.46 | 37.18 | 99.77 | 95.39 | 99.89 | |
| SSRN2026.01 | 88.51 | — | — | — | — | — | — | — | — | — | 94.08 | 85.31 | 91.06 | 80.87 | 86.11 | 99.07 | 100 | 97.45 | 100 | 92.17 | 100 | |
| ADGKTtransfer_source=IndianPine2025.12 | 87.52 | — | — | — | — | — | — | — | — | — | 83.93 | 83.01 | 86.45 | 94.44 | 76.64 | 84.68 | 96.05 | 64.52 | 64.63 | 94.2 | 93.71 | |
| Two-stage pixel-wise label propagation schemetraining samples per class=52026.01 | 87.21 | 73.99 | 98.5 | 76.24 | 70.28 | 76.4 | 99.52 | 95.03 | 55.91 | 97.15 | 82.56 | 82.81 | — | — | — | — | — | — | — | — | — | |
| ConvNeXt2026.01 | 87.11 | — | — | — | — | — | — | — | — | — | 89.74 | 91.25 | 86.04 | 89.56 | 87.27 | 99.63 | 100 | 74.1 | 99.37 | 71.69 | 100 | |
| DMSGerLabeled samples per class=102026.01 | 87 | 69.71 | 87.27 | 91.46 | 84.92 | 99.8 | 95.75 | 98.88 | 92.47 | 98.85 | 90.9 | 83.3 | — | — | — | — | — | — | — | — | — | |
| A2S2KLabeled samples per class=102026.01 | 86.94 | 73.06 | 88.29 | 81.58 | 96.8 | 100 | 91.27 | 98.98 | 60.79 | 99.59 | 87.26 | 85.3 | — | — | — | — | — | — | — | — | — | |
| DSSPLtraining samples per class=52026.01 | 86.74 | 80.72 | 92.73 | 95.24 | 82.83 | 96.31 | 99.29 | 88.06 | 83.29 | 99.89 | 89.37 | 84.05 | — | — | — | — | — | — | — | — | — | |
| CTF-SSCLLabeled samples per class=102026.01 | 86.47 | 91.95 | 80.52 | 84.32 | 92.69 | 99.81 | 88.29 | 97.66 | 88.42 | 98.32 | 91.33 | 82.74 | — | — | — | — | — | — | — | — | — | |
| SDMamba2026.01 | 86.34 | — | — | — | — | — | — | — | — | — | 92.46 | 82.57 | 86.24 | 78.44 | 83.68 | 99.07 | 100 | 96.85 | 96.2 | 91.63 | 100 | |
| SSTNLabeled samples per class=102026.01 | 86.15 | 91.83 | 79.59 | 91.87 | 79.13 | 99.98 | 97.36 | 98.14 | 83.91 | 99.53 | 91.26 | 82.27 | — | — | — | — | — | — | — | — | — | |
| SGLtraining samples per class=52026.01 | 85.69 | 80.13 | 80.8 | 99.86 | 64.78 | 96.8 | 77.75 | 100 | 77.73 | 99.89 | 86.53 | 76.08 | — | — | — | — | — | — | — | — | — | |
| SSTN2026.01 | 85.31 | — | — | — | — | — | — | — | — | — | 86.91 | 87.5 | 82.6 | 84.82 | 79.55 | 79.7 | 99.92 | 86.99 | 79.92 | 91.95 | 97.24 | |
| A-SPN2025.12 | 83.87 | — | — | — | — | — | — | — | — | — | 79.29 | 80 | — | — | — | — | — | — | — | — | — | |
| STSE-DWLRtraining samples per class=52026.01 | 83.02 | 77.69 | 82.25 | 98.12 | 59.44 | 98.54 | 97.74 | 88.42 | 75.12 | 99.89 | 87.69 | 78.84 | — | — | — | — | — | — | — | — | — | |
| Adaptortransfer_source=IndianPine2025.12 | 82.97 | — | — | — | — | — | — | — | — | — | 83.37 | 77.61 | 85.42 | 81.57 | 87.99 | 86.44 | 96.77 | 79.18 | 56.57 | 86.95 | 89.43 | |
| Finetunetransfer_source=IndianPine2025.12 | 82.67 | — | — | — | — | — | — | — | — | — | 82.15 | 76.91 | 79.74 | 86.1 | 68.32 | 87.81 | 99.1 | 65.46 | 68.6 | 93.01 | 91.19 | |
| UANtransfer_source=IndianPine2025.12 | 80.77 | — | — | — | — | — | — | — | — | — | 80.67 | 74.61 | 78.11 | 82.34 | 66.89 | 82.49 | 96.32 | 69.79 | 69.42 | 95.3 | 85.41 | |
| FSStraining samples per class=52026.01 | 80.39 | 70 | 79.47 | 95.06 | 32.85 | 95.47 | 97.38 | 99.67 | 87.72 | 30.41 | 83.78 | 77.74 | — | — | — | — | — | — | — | — | — | |
| SVM-GCAlgorithm Category=Regularized classification algorithms2017.05 | 80.21 | 97.74 | 67.92 | 90.38 | 90.34 | 99.85 | 94.75 | 71.16 | 68.51 | 99.67 | 86.7 | 75.09 | — | — | — | — | — | — | — | — | — | |
| k-NN2016.12 | 79.45 | 77.7 | 75.3 | 77.27 | 92.46 | 99.63 | 79.5 | 92.86 | 76.45 | 99.62 | — | — | — | — | — | — | — | — | — | — | — | |
| ViT2025.12 | 76.99 | — | — | — | — | — | — | — | — | — | 80.22 | 70.1 | — | — | — | — | — | — | — | — | — | |
| MTLtransfer_source=IndianPine2025.12 | 76.26 | — | — | — | — | — | — | — | — | — | 81.05 | 69.62 | 78.66 | 69.5 | 73.72 | 81.73 | 94.34 | 79.31 | 70.03 | 93.61 | 88.55 | |
| FFLtransfer_source=IndianPine2025.12 | 74.89 | — | — | — | — | — | — | — | — | — | 79.63 | 67.52 | 73.65 | 73.39 | 71.46 | 91.14 | 97.21 | 50.79 | 72.17 | 95.3 | 91.57 | |
| Baselinetransfer_mode=Target-only training2025.12 | 74.27 | — | — | — | — | — | — | — | — | — | 79.01 | 66.93 | 83.68 | 69.02 | 57.25 | 90.01 | 99.01 | 55.51 | 68.5 | 95.04 | 93.08 | |
| ONEtransfer_source=IndianPine2025.12 | 73.9 | — | — | — | — | — | — | — | — | — | 78.88 | 66.91 | 79.7 | 63.72 | 73.28 | 84.51 | 87.78 | 85.74 | 58 | 89.8 | 87.42 | |
| SVMAlgorithm Category=Classification algorithms2017.05 | 73.41 | 72.17 | 64.85 | 75.57 | 89.62 | 97.78 | 72.42 | 86.82 | 67.13 | 97.57 | 79.36 | 66.23 | — | — | — | — | — | — | — | — | — | |
| GFHFtraining samples per class=52026.01 | 72.52 | 70.78 | 79.07 | 97.95 | 51.92 | 88.71 | 96.76 | 87.54 | 84.63 | 98.6 | 77.44 | 77.74 | — | — | — | — | — | — | — | — | — |