Feature Extraction
Transformers
PyTorch
hear_canon_vit
audio
medical
embeddings
vision-transformer
distillation
canon
custom_code
Instructions to use matthewagi/HeAR-s with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use matthewagi/HeAR-s with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("feature-extraction", model="matthewagi/HeAR-s", trust_remote_code=True)# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("matthewagi/HeAR-s", trust_remote_code=True, dtype="auto") - Notebooks
- Google Colab
- Kaggle
| { | |
| "checkpoint_source": "/home/matt/Documents/HeAR/trained_model.pt", | |
| "embedding_dimension": 384, | |
| "exported_weights_file": "pytorch_model.bin", | |
| "input_spectrogram_shape": [ | |
| "B", | |
| 1, | |
| 192, | |
| 128 | |
| ], | |
| "input_waveform_shape": [ | |
| "B", | |
| 32000 | |
| ], | |
| "student_parameters": 22140288, | |
| "student_state_dict_shapes": { | |
| "blocks.0.block.attn.canon._fallback.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.0.block.attn.canon._fallback.conv.weight": [ | |
| 384, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.0.block.attn.canon.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.0.block.attn.canon.conv.weight": [ | |
| 384, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.0.block.attn.module.proj.0.bias": [ | |
| 384 | |
| ], | |
| "blocks.0.block.attn.module.proj.0.weight": [ | |
| 384, | |
| 384 | |
| ], | |
| "blocks.0.block.attn.module.proj.1._fallback.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.0.block.attn.module.proj.1._fallback.conv.weight": [ | |
| 384, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.0.block.attn.module.proj.1.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.0.block.attn.module.proj.1.conv.weight": [ | |
| 384, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.0.block.attn.module.qkv.bias": [ | |
| 1152 | |
| ], | |
| "blocks.0.block.attn.module.qkv.weight": [ | |
| 1152, | |
| 384 | |
| ], | |
| "blocks.0.block.mlp.canon._fallback.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.0.block.mlp.canon._fallback.conv.weight": [ | |
| 384, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.0.block.mlp.canon.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.0.block.mlp.canon.conv.weight": [ | |
| 384, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.0.block.mlp.module.fc1.canon._fallback.conv.bias": [ | |
| 1536 | |
| ], | |
| "blocks.0.block.mlp.module.fc1.canon._fallback.conv.weight": [ | |
| 1536, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.0.block.mlp.module.fc1.canon.conv.bias": [ | |
| 1536 | |
| ], | |
| "blocks.0.block.mlp.module.fc1.canon.conv.weight": [ | |
| 1536, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.0.block.mlp.module.fc1.fc1.bias": [ | |
| 1536 | |
| ], | |
| "blocks.0.block.mlp.module.fc1.fc1.weight": [ | |
| 1536, | |
| 384 | |
| ], | |
| "blocks.0.block.mlp.module.fc2.bias": [ | |
| 384 | |
| ], | |
| "blocks.0.block.mlp.module.fc2.weight": [ | |
| 384, | |
| 1536 | |
| ], | |
| "blocks.0.block.norm1.bias": [ | |
| 384 | |
| ], | |
| "blocks.0.block.norm1.weight": [ | |
| 384 | |
| ], | |
| "blocks.0.block.norm2.bias": [ | |
| 384 | |
| ], | |
| "blocks.0.block.norm2.weight": [ | |
| 384 | |
| ], | |
| "blocks.1.block.attn.canon._fallback.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.1.block.attn.canon._fallback.conv.weight": [ | |
| 384, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.1.block.attn.canon.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.1.block.attn.canon.conv.weight": [ | |
| 384, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.1.block.attn.module.proj.0.bias": [ | |
| 384 | |
| ], | |
| "blocks.1.block.attn.module.proj.0.weight": [ | |
| 384, | |
| 384 | |
| ], | |
| "blocks.1.block.attn.module.proj.1._fallback.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.1.block.attn.module.proj.1._fallback.conv.weight": [ | |
| 384, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.1.block.attn.module.proj.1.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.1.block.attn.module.proj.1.conv.weight": [ | |
| 384, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.1.block.attn.module.qkv.bias": [ | |
| 1152 | |
| ], | |
| "blocks.1.block.attn.module.qkv.weight": [ | |
| 1152, | |
| 384 | |
| ], | |
| "blocks.1.block.mlp.canon._fallback.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.1.block.mlp.canon._fallback.conv.weight": [ | |
| 384, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.1.block.mlp.canon.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.1.block.mlp.canon.conv.weight": [ | |
| 384, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.1.block.mlp.module.fc1.canon._fallback.conv.bias": [ | |
| 1536 | |
| ], | |
| "blocks.1.block.mlp.module.fc1.canon._fallback.conv.weight": [ | |
| 1536, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.1.block.mlp.module.fc1.canon.conv.bias": [ | |
| 1536 | |
| ], | |
| "blocks.1.block.mlp.module.fc1.canon.conv.weight": [ | |
| 1536, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.1.block.mlp.module.fc1.fc1.bias": [ | |
| 1536 | |
| ], | |
| "blocks.1.block.mlp.module.fc1.fc1.weight": [ | |
| 1536, | |
| 384 | |
| ], | |
| "blocks.1.block.mlp.module.fc2.bias": [ | |
| 384 | |
| ], | |
| "blocks.1.block.mlp.module.fc2.weight": [ | |
| 384, | |
| 1536 | |
| ], | |
| "blocks.1.block.norm1.bias": [ | |
| 384 | |
| ], | |
| "blocks.1.block.norm1.weight": [ | |
| 384 | |
| ], | |
| "blocks.1.block.norm2.bias": [ | |
| 384 | |
| ], | |
| "blocks.1.block.norm2.weight": [ | |
| 384 | |
| ], | |
| "blocks.10.block.attn.canon._fallback.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.10.block.attn.canon._fallback.conv.weight": [ | |
| 384, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.10.block.attn.canon.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.10.block.attn.canon.conv.weight": [ | |
| 384, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.10.block.attn.module.proj.0.bias": [ | |
| 384 | |
| ], | |
| "blocks.10.block.attn.module.proj.0.weight": [ | |
| 384, | |
| 384 | |
| ], | |
| "blocks.10.block.attn.module.proj.1._fallback.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.10.block.attn.module.proj.1._fallback.conv.weight": [ | |
| 384, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.10.block.attn.module.proj.1.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.10.block.attn.module.proj.1.conv.weight": [ | |
| 384, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.10.block.attn.module.qkv.bias": [ | |
| 1152 | |
| ], | |
| "blocks.10.block.attn.module.qkv.weight": [ | |
| 1152, | |
| 384 | |
| ], | |
| "blocks.10.block.mlp.canon._fallback.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.10.block.mlp.canon._fallback.conv.weight": [ | |
| 384, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.10.block.mlp.canon.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.10.block.mlp.canon.conv.weight": [ | |
| 384, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.10.block.mlp.module.fc1.canon._fallback.conv.bias": [ | |
| 1536 | |
| ], | |
| "blocks.10.block.mlp.module.fc1.canon._fallback.conv.weight": [ | |
| 1536, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.10.block.mlp.module.fc1.canon.conv.bias": [ | |
| 1536 | |
| ], | |
| "blocks.10.block.mlp.module.fc1.canon.conv.weight": [ | |
| 1536, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.10.block.mlp.module.fc1.fc1.bias": [ | |
| 1536 | |
| ], | |
| "blocks.10.block.mlp.module.fc1.fc1.weight": [ | |
| 1536, | |
| 384 | |
| ], | |
| "blocks.10.block.mlp.module.fc2.bias": [ | |
| 384 | |
| ], | |
| "blocks.10.block.mlp.module.fc2.weight": [ | |
| 384, | |
| 1536 | |
| ], | |
| "blocks.10.block.norm1.bias": [ | |
| 384 | |
| ], | |
| "blocks.10.block.norm1.weight": [ | |
| 384 | |
| ], | |
| "blocks.10.block.norm2.bias": [ | |
| 384 | |
| ], | |
| "blocks.10.block.norm2.weight": [ | |
| 384 | |
| ], | |
| "blocks.11.block.attn.canon._fallback.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.11.block.attn.canon._fallback.conv.weight": [ | |
| 384, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.11.block.attn.canon.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.11.block.attn.canon.conv.weight": [ | |
| 384, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.11.block.attn.module.proj.0.bias": [ | |
| 384 | |
| ], | |
| "blocks.11.block.attn.module.proj.0.weight": [ | |
| 384, | |
| 384 | |
| ], | |
| "blocks.11.block.attn.module.proj.1._fallback.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.11.block.attn.module.proj.1._fallback.conv.weight": [ | |
| 384, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.11.block.attn.module.proj.1.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.11.block.attn.module.proj.1.conv.weight": [ | |
| 384, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.11.block.attn.module.qkv.bias": [ | |
| 1152 | |
| ], | |
| "blocks.11.block.attn.module.qkv.weight": [ | |
| 1152, | |
| 384 | |
| ], | |
| "blocks.11.block.mlp.canon._fallback.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.11.block.mlp.canon._fallback.conv.weight": [ | |
| 384, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.11.block.mlp.canon.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.11.block.mlp.canon.conv.weight": [ | |
| 384, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.11.block.mlp.module.fc1.canon._fallback.conv.bias": [ | |
| 1536 | |
| ], | |
| "blocks.11.block.mlp.module.fc1.canon._fallback.conv.weight": [ | |
| 1536, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.11.block.mlp.module.fc1.canon.conv.bias": [ | |
| 1536 | |
| ], | |
| "blocks.11.block.mlp.module.fc1.canon.conv.weight": [ | |
| 1536, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.11.block.mlp.module.fc1.fc1.bias": [ | |
| 1536 | |
| ], | |
| "blocks.11.block.mlp.module.fc1.fc1.weight": [ | |
| 1536, | |
| 384 | |
| ], | |
| "blocks.11.block.mlp.module.fc2.bias": [ | |
| 384 | |
| ], | |
| "blocks.11.block.mlp.module.fc2.weight": [ | |
| 384, | |
| 1536 | |
| ], | |
| "blocks.11.block.norm1.bias": [ | |
| 384 | |
| ], | |
| "blocks.11.block.norm1.weight": [ | |
| 384 | |
| ], | |
| "blocks.11.block.norm2.bias": [ | |
| 384 | |
| ], | |
| "blocks.11.block.norm2.weight": [ | |
| 384 | |
| ], | |
| "blocks.2.block.attn.canon._fallback.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.2.block.attn.canon._fallback.conv.weight": [ | |
| 384, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.2.block.attn.canon.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.2.block.attn.canon.conv.weight": [ | |
| 384, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.2.block.attn.module.proj.0.bias": [ | |
| 384 | |
| ], | |
| "blocks.2.block.attn.module.proj.0.weight": [ | |
| 384, | |
| 384 | |
| ], | |
| "blocks.2.block.attn.module.proj.1._fallback.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.2.block.attn.module.proj.1._fallback.conv.weight": [ | |
| 384, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.2.block.attn.module.proj.1.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.2.block.attn.module.proj.1.conv.weight": [ | |
| 384, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.2.block.attn.module.qkv.bias": [ | |
| 1152 | |
| ], | |
| "blocks.2.block.attn.module.qkv.weight": [ | |
| 1152, | |
| 384 | |
| ], | |
| "blocks.2.block.mlp.canon._fallback.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.2.block.mlp.canon._fallback.conv.weight": [ | |
| 384, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.2.block.mlp.canon.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.2.block.mlp.canon.conv.weight": [ | |
| 384, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.2.block.mlp.module.fc1.canon._fallback.conv.bias": [ | |
| 1536 | |
| ], | |
| "blocks.2.block.mlp.module.fc1.canon._fallback.conv.weight": [ | |
| 1536, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.2.block.mlp.module.fc1.canon.conv.bias": [ | |
| 1536 | |
| ], | |
| "blocks.2.block.mlp.module.fc1.canon.conv.weight": [ | |
| 1536, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.2.block.mlp.module.fc1.fc1.bias": [ | |
| 1536 | |
| ], | |
| "blocks.2.block.mlp.module.fc1.fc1.weight": [ | |
| 1536, | |
| 384 | |
| ], | |
| "blocks.2.block.mlp.module.fc2.bias": [ | |
| 384 | |
| ], | |
| "blocks.2.block.mlp.module.fc2.weight": [ | |
| 384, | |
| 1536 | |
| ], | |
| "blocks.2.block.norm1.bias": [ | |
| 384 | |
| ], | |
| "blocks.2.block.norm1.weight": [ | |
| 384 | |
| ], | |
| "blocks.2.block.norm2.bias": [ | |
| 384 | |
| ], | |
| "blocks.2.block.norm2.weight": [ | |
| 384 | |
| ], | |
| "blocks.3.block.attn.canon._fallback.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.3.block.attn.canon._fallback.conv.weight": [ | |
| 384, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.3.block.attn.canon.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.3.block.attn.canon.conv.weight": [ | |
| 384, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.3.block.attn.module.proj.0.bias": [ | |
| 384 | |
| ], | |
| "blocks.3.block.attn.module.proj.0.weight": [ | |
| 384, | |
| 384 | |
| ], | |
| "blocks.3.block.attn.module.proj.1._fallback.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.3.block.attn.module.proj.1._fallback.conv.weight": [ | |
| 384, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.3.block.attn.module.proj.1.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.3.block.attn.module.proj.1.conv.weight": [ | |
| 384, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.3.block.attn.module.qkv.bias": [ | |
| 1152 | |
| ], | |
| "blocks.3.block.attn.module.qkv.weight": [ | |
| 1152, | |
| 384 | |
| ], | |
| "blocks.3.block.mlp.canon._fallback.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.3.block.mlp.canon._fallback.conv.weight": [ | |
| 384, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.3.block.mlp.canon.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.3.block.mlp.canon.conv.weight": [ | |
| 384, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.3.block.mlp.module.fc1.canon._fallback.conv.bias": [ | |
| 1536 | |
| ], | |
| "blocks.3.block.mlp.module.fc1.canon._fallback.conv.weight": [ | |
| 1536, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.3.block.mlp.module.fc1.canon.conv.bias": [ | |
| 1536 | |
| ], | |
| "blocks.3.block.mlp.module.fc1.canon.conv.weight": [ | |
| 1536, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.3.block.mlp.module.fc1.fc1.bias": [ | |
| 1536 | |
| ], | |
| "blocks.3.block.mlp.module.fc1.fc1.weight": [ | |
| 1536, | |
| 384 | |
| ], | |
| "blocks.3.block.mlp.module.fc2.bias": [ | |
| 384 | |
| ], | |
| "blocks.3.block.mlp.module.fc2.weight": [ | |
| 384, | |
| 1536 | |
| ], | |
| "blocks.3.block.norm1.bias": [ | |
| 384 | |
| ], | |
| "blocks.3.block.norm1.weight": [ | |
| 384 | |
| ], | |
| "blocks.3.block.norm2.bias": [ | |
| 384 | |
| ], | |
| "blocks.3.block.norm2.weight": [ | |
| 384 | |
| ], | |
| "blocks.4.block.attn.canon._fallback.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.4.block.attn.canon._fallback.conv.weight": [ | |
| 384, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.4.block.attn.canon.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.4.block.attn.canon.conv.weight": [ | |
| 384, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.4.block.attn.module.proj.0.bias": [ | |
| 384 | |
| ], | |
| "blocks.4.block.attn.module.proj.0.weight": [ | |
| 384, | |
| 384 | |
| ], | |
| "blocks.4.block.attn.module.proj.1._fallback.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.4.block.attn.module.proj.1._fallback.conv.weight": [ | |
| 384, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.4.block.attn.module.proj.1.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.4.block.attn.module.proj.1.conv.weight": [ | |
| 384, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.4.block.attn.module.qkv.bias": [ | |
| 1152 | |
| ], | |
| "blocks.4.block.attn.module.qkv.weight": [ | |
| 1152, | |
| 384 | |
| ], | |
| "blocks.4.block.mlp.canon._fallback.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.4.block.mlp.canon._fallback.conv.weight": [ | |
| 384, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.4.block.mlp.canon.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.4.block.mlp.canon.conv.weight": [ | |
| 384, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.4.block.mlp.module.fc1.canon._fallback.conv.bias": [ | |
| 1536 | |
| ], | |
| "blocks.4.block.mlp.module.fc1.canon._fallback.conv.weight": [ | |
| 1536, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.4.block.mlp.module.fc1.canon.conv.bias": [ | |
| 1536 | |
| ], | |
| "blocks.4.block.mlp.module.fc1.canon.conv.weight": [ | |
| 1536, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.4.block.mlp.module.fc1.fc1.bias": [ | |
| 1536 | |
| ], | |
| "blocks.4.block.mlp.module.fc1.fc1.weight": [ | |
| 1536, | |
| 384 | |
| ], | |
| "blocks.4.block.mlp.module.fc2.bias": [ | |
| 384 | |
| ], | |
| "blocks.4.block.mlp.module.fc2.weight": [ | |
| 384, | |
| 1536 | |
| ], | |
| "blocks.4.block.norm1.bias": [ | |
| 384 | |
| ], | |
| "blocks.4.block.norm1.weight": [ | |
| 384 | |
| ], | |
| "blocks.4.block.norm2.bias": [ | |
| 384 | |
| ], | |
| "blocks.4.block.norm2.weight": [ | |
| 384 | |
| ], | |
| "blocks.5.block.attn.canon._fallback.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.5.block.attn.canon._fallback.conv.weight": [ | |
| 384, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.5.block.attn.canon.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.5.block.attn.canon.conv.weight": [ | |
| 384, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.5.block.attn.module.proj.0.bias": [ | |
| 384 | |
| ], | |
| "blocks.5.block.attn.module.proj.0.weight": [ | |
| 384, | |
| 384 | |
| ], | |
| "blocks.5.block.attn.module.proj.1._fallback.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.5.block.attn.module.proj.1._fallback.conv.weight": [ | |
| 384, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.5.block.attn.module.proj.1.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.5.block.attn.module.proj.1.conv.weight": [ | |
| 384, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.5.block.attn.module.qkv.bias": [ | |
| 1152 | |
| ], | |
| "blocks.5.block.attn.module.qkv.weight": [ | |
| 1152, | |
| 384 | |
| ], | |
| "blocks.5.block.mlp.canon._fallback.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.5.block.mlp.canon._fallback.conv.weight": [ | |
| 384, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.5.block.mlp.canon.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.5.block.mlp.canon.conv.weight": [ | |
| 384, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.5.block.mlp.module.fc1.canon._fallback.conv.bias": [ | |
| 1536 | |
| ], | |
| "blocks.5.block.mlp.module.fc1.canon._fallback.conv.weight": [ | |
| 1536, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.5.block.mlp.module.fc1.canon.conv.bias": [ | |
| 1536 | |
| ], | |
| "blocks.5.block.mlp.module.fc1.canon.conv.weight": [ | |
| 1536, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.5.block.mlp.module.fc1.fc1.bias": [ | |
| 1536 | |
| ], | |
| "blocks.5.block.mlp.module.fc1.fc1.weight": [ | |
| 1536, | |
| 384 | |
| ], | |
| "blocks.5.block.mlp.module.fc2.bias": [ | |
| 384 | |
| ], | |
| "blocks.5.block.mlp.module.fc2.weight": [ | |
| 384, | |
| 1536 | |
| ], | |
| "blocks.5.block.norm1.bias": [ | |
| 384 | |
| ], | |
| "blocks.5.block.norm1.weight": [ | |
| 384 | |
| ], | |
| "blocks.5.block.norm2.bias": [ | |
| 384 | |
| ], | |
| "blocks.5.block.norm2.weight": [ | |
| 384 | |
| ], | |
| "blocks.6.block.attn.canon._fallback.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.6.block.attn.canon._fallback.conv.weight": [ | |
| 384, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.6.block.attn.canon.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.6.block.attn.canon.conv.weight": [ | |
| 384, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.6.block.attn.module.proj.0.bias": [ | |
| 384 | |
| ], | |
| "blocks.6.block.attn.module.proj.0.weight": [ | |
| 384, | |
| 384 | |
| ], | |
| "blocks.6.block.attn.module.proj.1._fallback.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.6.block.attn.module.proj.1._fallback.conv.weight": [ | |
| 384, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.6.block.attn.module.proj.1.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.6.block.attn.module.proj.1.conv.weight": [ | |
| 384, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.6.block.attn.module.qkv.bias": [ | |
| 1152 | |
| ], | |
| "blocks.6.block.attn.module.qkv.weight": [ | |
| 1152, | |
| 384 | |
| ], | |
| "blocks.6.block.mlp.canon._fallback.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.6.block.mlp.canon._fallback.conv.weight": [ | |
| 384, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.6.block.mlp.canon.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.6.block.mlp.canon.conv.weight": [ | |
| 384, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.6.block.mlp.module.fc1.canon._fallback.conv.bias": [ | |
| 1536 | |
| ], | |
| "blocks.6.block.mlp.module.fc1.canon._fallback.conv.weight": [ | |
| 1536, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.6.block.mlp.module.fc1.canon.conv.bias": [ | |
| 1536 | |
| ], | |
| "blocks.6.block.mlp.module.fc1.canon.conv.weight": [ | |
| 1536, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.6.block.mlp.module.fc1.fc1.bias": [ | |
| 1536 | |
| ], | |
| "blocks.6.block.mlp.module.fc1.fc1.weight": [ | |
| 1536, | |
| 384 | |
| ], | |
| "blocks.6.block.mlp.module.fc2.bias": [ | |
| 384 | |
| ], | |
| "blocks.6.block.mlp.module.fc2.weight": [ | |
| 384, | |
| 1536 | |
| ], | |
| "blocks.6.block.norm1.bias": [ | |
| 384 | |
| ], | |
| "blocks.6.block.norm1.weight": [ | |
| 384 | |
| ], | |
| "blocks.6.block.norm2.bias": [ | |
| 384 | |
| ], | |
| "blocks.6.block.norm2.weight": [ | |
| 384 | |
| ], | |
| "blocks.7.block.attn.canon._fallback.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.7.block.attn.canon._fallback.conv.weight": [ | |
| 384, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.7.block.attn.canon.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.7.block.attn.canon.conv.weight": [ | |
| 384, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.7.block.attn.module.proj.0.bias": [ | |
| 384 | |
| ], | |
| "blocks.7.block.attn.module.proj.0.weight": [ | |
| 384, | |
| 384 | |
| ], | |
| "blocks.7.block.attn.module.proj.1._fallback.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.7.block.attn.module.proj.1._fallback.conv.weight": [ | |
| 384, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.7.block.attn.module.proj.1.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.7.block.attn.module.proj.1.conv.weight": [ | |
| 384, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.7.block.attn.module.qkv.bias": [ | |
| 1152 | |
| ], | |
| "blocks.7.block.attn.module.qkv.weight": [ | |
| 1152, | |
| 384 | |
| ], | |
| "blocks.7.block.mlp.canon._fallback.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.7.block.mlp.canon._fallback.conv.weight": [ | |
| 384, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.7.block.mlp.canon.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.7.block.mlp.canon.conv.weight": [ | |
| 384, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.7.block.mlp.module.fc1.canon._fallback.conv.bias": [ | |
| 1536 | |
| ], | |
| "blocks.7.block.mlp.module.fc1.canon._fallback.conv.weight": [ | |
| 1536, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.7.block.mlp.module.fc1.canon.conv.bias": [ | |
| 1536 | |
| ], | |
| "blocks.7.block.mlp.module.fc1.canon.conv.weight": [ | |
| 1536, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.7.block.mlp.module.fc1.fc1.bias": [ | |
| 1536 | |
| ], | |
| "blocks.7.block.mlp.module.fc1.fc1.weight": [ | |
| 1536, | |
| 384 | |
| ], | |
| "blocks.7.block.mlp.module.fc2.bias": [ | |
| 384 | |
| ], | |
| "blocks.7.block.mlp.module.fc2.weight": [ | |
| 384, | |
| 1536 | |
| ], | |
| "blocks.7.block.norm1.bias": [ | |
| 384 | |
| ], | |
| "blocks.7.block.norm1.weight": [ | |
| 384 | |
| ], | |
| "blocks.7.block.norm2.bias": [ | |
| 384 | |
| ], | |
| "blocks.7.block.norm2.weight": [ | |
| 384 | |
| ], | |
| "blocks.8.block.attn.canon._fallback.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.8.block.attn.canon._fallback.conv.weight": [ | |
| 384, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.8.block.attn.canon.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.8.block.attn.canon.conv.weight": [ | |
| 384, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.8.block.attn.module.proj.0.bias": [ | |
| 384 | |
| ], | |
| "blocks.8.block.attn.module.proj.0.weight": [ | |
| 384, | |
| 384 | |
| ], | |
| "blocks.8.block.attn.module.proj.1._fallback.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.8.block.attn.module.proj.1._fallback.conv.weight": [ | |
| 384, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.8.block.attn.module.proj.1.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.8.block.attn.module.proj.1.conv.weight": [ | |
| 384, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.8.block.attn.module.qkv.bias": [ | |
| 1152 | |
| ], | |
| "blocks.8.block.attn.module.qkv.weight": [ | |
| 1152, | |
| 384 | |
| ], | |
| "blocks.8.block.mlp.canon._fallback.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.8.block.mlp.canon._fallback.conv.weight": [ | |
| 384, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.8.block.mlp.canon.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.8.block.mlp.canon.conv.weight": [ | |
| 384, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.8.block.mlp.module.fc1.canon._fallback.conv.bias": [ | |
| 1536 | |
| ], | |
| "blocks.8.block.mlp.module.fc1.canon._fallback.conv.weight": [ | |
| 1536, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.8.block.mlp.module.fc1.canon.conv.bias": [ | |
| 1536 | |
| ], | |
| "blocks.8.block.mlp.module.fc1.canon.conv.weight": [ | |
| 1536, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.8.block.mlp.module.fc1.fc1.bias": [ | |
| 1536 | |
| ], | |
| "blocks.8.block.mlp.module.fc1.fc1.weight": [ | |
| 1536, | |
| 384 | |
| ], | |
| "blocks.8.block.mlp.module.fc2.bias": [ | |
| 384 | |
| ], | |
| "blocks.8.block.mlp.module.fc2.weight": [ | |
| 384, | |
| 1536 | |
| ], | |
| "blocks.8.block.norm1.bias": [ | |
| 384 | |
| ], | |
| "blocks.8.block.norm1.weight": [ | |
| 384 | |
| ], | |
| "blocks.8.block.norm2.bias": [ | |
| 384 | |
| ], | |
| "blocks.8.block.norm2.weight": [ | |
| 384 | |
| ], | |
| "blocks.9.block.attn.canon._fallback.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.9.block.attn.canon._fallback.conv.weight": [ | |
| 384, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.9.block.attn.canon.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.9.block.attn.canon.conv.weight": [ | |
| 384, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.9.block.attn.module.proj.0.bias": [ | |
| 384 | |
| ], | |
| "blocks.9.block.attn.module.proj.0.weight": [ | |
| 384, | |
| 384 | |
| ], | |
| "blocks.9.block.attn.module.proj.1._fallback.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.9.block.attn.module.proj.1._fallback.conv.weight": [ | |
| 384, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.9.block.attn.module.proj.1.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.9.block.attn.module.proj.1.conv.weight": [ | |
| 384, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.9.block.attn.module.qkv.bias": [ | |
| 1152 | |
| ], | |
| "blocks.9.block.attn.module.qkv.weight": [ | |
| 1152, | |
| 384 | |
| ], | |
| "blocks.9.block.mlp.canon._fallback.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.9.block.mlp.canon._fallback.conv.weight": [ | |
| 384, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.9.block.mlp.canon.conv.bias": [ | |
| 384 | |
| ], | |
| "blocks.9.block.mlp.canon.conv.weight": [ | |
| 384, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.9.block.mlp.module.fc1.canon._fallback.conv.bias": [ | |
| 1536 | |
| ], | |
| "blocks.9.block.mlp.module.fc1.canon._fallback.conv.weight": [ | |
| 1536, | |
| 1, | |
| 4 | |
| ], | |
| "blocks.9.block.mlp.module.fc1.canon.conv.bias": [ | |
| 1536 | |
| ], | |
| "blocks.9.block.mlp.module.fc1.canon.conv.weight": [ | |
| 1536, | |
| 1, | |
| 4, | |
| 4 | |
| ], | |
| "blocks.9.block.mlp.module.fc1.fc1.bias": [ | |
| 1536 | |
| ], | |
| "blocks.9.block.mlp.module.fc1.fc1.weight": [ | |
| 1536, | |
| 384 | |
| ], | |
| "blocks.9.block.mlp.module.fc2.bias": [ | |
| 384 | |
| ], | |
| "blocks.9.block.mlp.module.fc2.weight": [ | |
| 384, | |
| 1536 | |
| ], | |
| "blocks.9.block.norm1.bias": [ | |
| 384 | |
| ], | |
| "blocks.9.block.norm1.weight": [ | |
| 384 | |
| ], | |
| "blocks.9.block.norm2.bias": [ | |
| 384 | |
| ], | |
| "blocks.9.block.norm2.weight": [ | |
| 384 | |
| ], | |
| "cls_token": [ | |
| 1, | |
| 1, | |
| 384 | |
| ], | |
| "fc_norm.bias": [ | |
| 384 | |
| ], | |
| "fc_norm.weight": [ | |
| 384 | |
| ], | |
| "patch_embed.proj.bias": [ | |
| 384 | |
| ], | |
| "patch_embed.proj.weight": [ | |
| 384, | |
| 1, | |
| 16, | |
| 16 | |
| ], | |
| "pos_embed": [ | |
| 1, | |
| 97, | |
| 384 | |
| ] | |
| }, | |
| "student_state_dict_size": 342, | |
| "total_parameters": 22140288 | |
| } |