Fix initial convolution channels not multiplied by width_multiplier.

PiperOrigin-RevId: 327005191

byol/utils/networks.py

@@ -45,7 +45,12 @@ class MLP(hk.Module):
     return out
 
 
-class ResNetTorso(hk.nets.ResNet):
+def check_length(length, value, name):
+  if len(value) != length:
+    raise ValueError(f'`{name}` must be of length 4 not {len(value)}')
+
+
+class ResNetTorso(hk.Module):
   """ResNet model."""
 
   def __init__(
@@ -80,16 +85,49 @@ class ResNetTorso(hk.nets.ResNet):
       width_multiplier: An integer multiplying the number of channels per group.
       name: Name of the module.
     """
-    channels_per_group = [width_multiplier * channel
+    super().__init__(name=name)
-                          for channel in channels_per_group]
+    self.resnet_v2 = resnet_v2
-    super().__init__(
+
-        blocks_per_group=blocks_per_group,
+    bn_config = dict(bn_config or {})
-        num_classes=num_classes,
+    bn_config.setdefault('decay_rate', 0.9)
-        bn_config=bn_config,
+    bn_config.setdefault('eps', 1e-5)
-        resnet_v2=resnet_v2,
+    bn_config.setdefault('create_scale', True)
-        bottleneck=bottleneck,
+    bn_config.setdefault('create_offset', True)
-        channels_per_group=channels_per_group,
+
-        name=name)
+    # Number of blocks in each group for ResNet.
+    check_length(4, blocks_per_group, 'blocks_per_group')
+    check_length(4, channels_per_group, 'channels_per_group')
+
+    self.initial_conv = hk.Conv2D(
+        output_channels=64 * width_multiplier,
+        kernel_shape=7,
+        stride=2,
+        with_bias=False,
+        padding='SAME',
+        name='initial_conv')
+
+    if not self.resnet_v2:
+      self.initial_batchnorm = hk.BatchNorm(name='initial_batchnorm',
+                                            **bn_config)
+
+    self.block_groups = []
+    strides = (1, 2, 2, 2)
+    for i in range(4):
+      self.block_groups.append(
+          hk.nets.ResNet.BlockGroup(
+              channels=width_multiplier * channels_per_group[i],
+              num_blocks=blocks_per_group[i],
+              stride=strides[i],
+              bn_config=bn_config,
+              resnet_v2=resnet_v2,
+              bottleneck=bottleneck,
+              use_projection=use_projection[i],
+              name='block_group_%d' % (i)))
+
+    if self.resnet_v2:
+      self.final_batchnorm = hk.BatchNorm(name='final_batchnorm', **bn_config)
+
+    self.logits = hk.Linear(num_classes, w_init=jnp.zeros, name='logits')
 
   def __call__(self, inputs, is_training, test_local_stats=False):
     out = inputs