| ## 📊 Hybrid Model Results (HMDB51) |
|
|
| The hybrid model (TimeSformer + RetNet) was also trained on the **HMDB51 dataset**. |
|
|
| Due to Kaggle’s runtime limitation, training was interrupted at **Epoch 12**, so results are reported up to **Epoch 11**. Training will be resumed in a later stage. |
|
|
| --- |
|
|
| ### 🔹 Training Results (Epoch 1–11) |
|
|
| | Epoch | Train Loss | Train Acc | Val Loss | Val Acc | F1 | |
| |------|------------|-----------|----------|---------|-----| |
| | 1 | 3.9312 | 0.0350 | 3.8099 | 0.0967 | 0.0855 | |
| | 2 | 3.6330 | 0.1791 | 3.2948 | 0.3654 | 0.3149 | |
| | 3 | 3.0989 | 0.3691 | 2.6927 | 0.5150 | 0.4579 | |
| | 4 | 2.6278 | 0.5048 | 2.2879 | 0.5869 | 0.5503 | |
| | 5 | 2.3198 | 0.5782 | 2.0438 | 0.6255 | 0.5961 | |
| | 6 | 2.1387 | 0.6194 | 1.9152 | 0.6242 | 0.6074 | |
| | 7 | 1.9876 | 0.6657 | 1.8369 | 0.6418 | 0.6308 | |
| | 8 | 1.9140 | 0.6936 | 1.7966 | 0.6359 | 0.6188 | |
| | 9 | 1.8539 | 0.7041 | 1.7619 | 0.6556 | 0.6426 | |
| | 10 | 1.8149 | 0.7244 | 1.7523 | **0.6614** | **0.6512** | |
| | 11 | 1.7270 | 0.7561 | 1.7543 | 0.6556 | 0.6472 | |
|
|
| --- |
|
|
| ## 🏆 Best Performance (Current) |
|
|
| - **Validation Accuracy:** **66.14%** |
| - **F1 Score:** 0.6512 |
| - Achieved at **Epoch 10** |
|
|
| --- |
|
|
| ## ⚠️ Training Status |
|
|
| - Training **interrupted at Epoch 12** due to runtime limit |
| - Model will be **resumed from best checkpoint** |
| - Final performance may improve after full training |
|
|
| --- |
|
|
| ## ⚡ Efficiency |
|
|
| - Peak GPU Memory: **~7.2 GB** |
| - ~25% lower than standard TimeSformer |
| - Faster training per epoch |
|
|
| --- |
|
|
| ## 📌 Observations |
|
|
| - Steady improvement until Epoch 10 |
| - Slight plateau after that (possible early convergence) |
| - Lower accuracy compared to UCF101 (expected due to dataset complexity) |
|
|
| --- |
|
|
| ## 🔄 Next Steps |
|
|
| - Resume training from Epoch 11 checkpoint |
| - Complete remaining epochs |
| - Compare final performance with baseline model |
