MCPcopy Create free account
hub / github.com/alinlab/SelfPatch / vit_small

Function vit_small

selfpatch_vision_transformer.py:350–354  ·  view source on GitHub ↗
(patch_size=16, **kwargs)

Source from the content-addressed store, hash-verified

348 return model
349
350def vit_small(patch_size=16, **kwargs):
351 model = VisionTransformer(
352 patch_size=patch_size, embed_dim=384, depth=12, num_heads=6, mlp_ratio=4,
353 qkv_bias=True, norm_layer=partial(nn.LayerNorm, eps=1e-6), **kwargs)
354 return model
355
356def vit_base(patch_size=16, **kwargs):
357 model = VisionTransformer(

Callers

nothing calls this directly

Calls 1

VisionTransformerClass · 0.85

Tested by

no test coverage detected