Revert "fix huggingface#4"

This reverts commit 0b534e6.

src/transformers/models/segformer/modeling_tf_segformer.py

@@ -759,24 +759,7 @@ def hf_compute_loss(self, logits, labels):
 
         upsampled_logits = tf.image.resize(logits, size=label_interp_shape, method="bilinear")
         # compute weighted loss
-
-        if self.config.num_labels > 1:
-            loss_fct = tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True, reduction="none")
-            loss = loss_fct(upsampled_logits, labels)
-        elif self.config.num_labels == 1:
-            valid_mask = tf.cast(
-                (labels >= 0) & (labels != self.config.semantic_loss_ignore_index),
-                dtype=tf.float32
-            )
-            loss_fct = tf.keras.losses.BinaryCrossentropy(
-                from_logits=True,
-                reduction=tf.keras.losses.Reduction.NONE
-            )
-            loss = loss_fct(labels, upsampled_logits[:, 0])  # Assuming channel dimension is last
-            loss = tf.reduce_mean(loss * valid_mask)
-        else:
-            raise ValueError(f"Number of labels should be >=0: {self.config.num_labels}")
-
+        loss_fct = tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True, reduction="none")
 
         def masked_loss(real, pred):
             unmasked_loss = loss_fct(real, pred)
@@ -846,7 +829,10 @@ def call(
 
         loss = None
         if labels is not None:
-            loss = self.hf_compute_loss(logits=logits, labels=labels)
+            if not self.config.num_labels > 1:
+                raise ValueError("The number of labels should be greater than one")
+            else:
+                loss = self.hf_compute_loss(logits=logits, labels=labels)
 
         # make logits of shape (batch_size, num_labels, height, width) to
         # keep them consistent across APIs