update .md and .ipynb (#2139)

* Update transformer_asr.md

* Update transformer_asr.ipynb

* Update examples/audio/md/transformer_asr.md

Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>

* Update examples/audio/md/transformer_asr.md

Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>

---------

Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>

Author: divyashreepathihalli

examples/audio/ipynb/transformer_asr.ipynb

@@ -321,7 +321,7 @@
     "            preds = self([source, dec_input])\n",
     "            one_hot = tf.one_hot(dec_target, depth=self.num_classes)\n",
     "            mask = tf.math.logical_not(tf.math.equal(dec_target, 0))\n",
-    "            loss = model.compute_loss(None, one_hot, preds, sample_weight=mask)\n",
+    "            loss = self.compute_loss(None, one_hot, preds, sample_weight=mask)\n",
     "        trainable_vars = self.trainable_variables\n",
     "        gradients = tape.gradient(loss, trainable_vars)\n",
     "        self.optimizer.apply_gradients(zip(gradients, trainable_vars))\n",
@@ -336,7 +336,7 @@
     "        preds = self([source, dec_input])\n",
     "        one_hot = tf.one_hot(dec_target, depth=self.num_classes)\n",
     "        mask = tf.math.logical_not(tf.math.equal(dec_target, 0))\n",
-    "        loss = model.compute_loss(None, one_hot, preds, sample_weight=mask)\n",
+    "        loss = self.compute_loss(None, one_hot, preds, sample_weight=mask)\n",
     "        self.loss_metric.update_state(loss)\n",
     "        return {\"loss\": self.loss_metric.result()}\n",
     "\n",

examples/audio/md/transformer_asr.md

@@ -262,7 +262,7 @@ class Transformer(keras.Model):
             preds = self([source, dec_input])
             one_hot = tf.one_hot(dec_target, depth=self.num_classes)
             mask = tf.math.logical_not(tf.math.equal(dec_target, 0))
-            loss = model.compute_loss(None, one_hot, preds, sample_weight=mask)
+            loss = self.compute_loss(y=one_hot, y_pred=preds, sample_weight=mask)
         trainable_vars = self.trainable_variables
         gradients = tape.gradient(loss, trainable_vars)
         self.optimizer.apply_gradients(zip(gradients, trainable_vars))
@@ -277,7 +277,7 @@ class Transformer(keras.Model):
         preds = self([source, dec_input])
         one_hot = tf.one_hot(dec_target, depth=self.num_classes)
         mask = tf.math.logical_not(tf.math.equal(dec_target, 0))
-        loss = model.compute_loss(None, one_hot, preds, sample_weight=mask)
+        loss = self.compute_loss(y=one_hot, y_pred=preds, sample_weight=mask)
         self.loss_metric.update_state(loss)
         return {"loss": self.loss_metric.result()}