| ---
|
| license: mit
|
| tags:
|
| - approx-gelu
|
| - bilinear
|
| - classification
|
| - dilated
|
| - kaiming
|
| - linear-warmup
|
| - perceiver
|
| - scalenorm
|
| - sgd
|
| - tiny
|
| ---
|
|
|
| # model_083400871_perceiver_tiny.py
|
|
|
| ## Model Overview
|
|
|
| A **tiny**-scale implementation of the **perceiver** architecture, built for **classification** tasks.
|
|
|
| ## Architecture
|
|
|
| - **Architecture**: perceiver
|
| - **Scale**: tiny
|
| - **Attention**: dilated
|
| - **Fusion strategy**: bilinear
|
| - **Task head**: classification
|
| - **Activation**: approx gelu
|
| - **Normalization**: scalenorm
|
| - **Initialization**: kaiming
|
|
|
| ## Training
|
|
|
| - **Optimizer**: sgd
|
| - **LR scheduler**: linear warmup
|
|
|
| ## Files
|
|
|
| - `model_083400871_perceiver_tiny.py` — main artifact of this repository
|
|
|
| ## License
|
|
|
| See the license field above.
|
|
|
