MCPcopy Create free account
hub / github.com/WeChatCV/WeVisionOne / __init__

Method __init__

WeVisionOne/pixel_decoder/groundingdino/utils.py:140–146  ·  view source on GitHub ↗
(self, input_dim, hidden_dim, output_dim, num_layers)

Source from the content-addressed store, hash-verified

138 """Very simple multi-layer perceptron (also called FFN)"""
139
140 def __init__(self, input_dim, hidden_dim, output_dim, num_layers):
141 super().__init__()
142 self.num_layers = num_layers
143 h = [hidden_dim] * (num_layers - 1)
144 self.layers = nn.ModuleList(
145 nn.Linear(n, k) for n, k in zip([input_dim] + h, h + [output_dim])
146 )
147
148 def forward(self, x):
149 for i, layer in enumerate(self.layers):

Callers

nothing calls this directly

Calls

no outgoing calls

Tested by

no test coverage detected