usernameauth
diff --git a/‎setup.py
+2-2 b/‎setup.py
+2-2
diff --git a/‎site/en/guide/advanced_autodiff.ipynb
+1-1 b/‎site/en/guide/advanced_autodiff.ipynb
+1-1
diff --git a/‎site/en/guide/basic_training_loops.ipynb
+1-1 b/‎site/en/guide/basic_training_loops.ipynb
+1-1
diff --git a/‎site/en/guide/data.ipynb
+2-2 b/‎site/en/guide/data.ipynb
+2-2
diff --git a/‎site/en/guide/dtensor_overview.ipynb
+3-3 b/‎site/en/guide/dtensor_overview.ipynb
+3-3
diff --git a/‎site/en/guide/graph_optimization.ipynb
+1-1 b/‎site/en/guide/graph_optimization.ipynb
+1-1
diff --git a/‎site/en/guide/intro_to_graphs.ipynb
+1-1 b/‎site/en/guide/intro_to_graphs.ipynb
+1-1
diff --git a/‎site/en/guide/migrate/canned_estimators.ipynb
+1-1 b/‎site/en/guide/migrate/canned_estimators.ipynb
+1-1
diff --git a/‎site/en/guide/migrate/evaluator.ipynb
+1-1 b/‎site/en/guide/migrate/evaluator.ipynb
+1-1
diff --git a/‎site/en/guide/migrate/migrating_feature_columns.ipynb
+1-1 b/‎site/en/guide/migrate/migrating_feature_columns.ipynb
+1-1
diff --git a/‎site/en/guide/migrate/migration_debugging.ipynb
+1-1 b/‎site/en/guide/migrate/migration_debugging.ipynb
+1-1
diff --git a/‎site/en/guide/migrate/tf1_vs_tf2.ipynb
+1-1 b/‎site/en/guide/migrate/tf1_vs_tf2.ipynb
+1-1
diff --git a/‎site/en/guide/migrate/upgrade.ipynb
+1-1 b/‎site/en/guide/migrate/upgrade.ipynb
+1-1
diff --git a/‎site/en/guide/ragged_tensor.ipynb
+1-1 b/‎site/en/guide/ragged_tensor.ipynb
+1-1
diff --git a/‎site/en/guide/random_numbers.ipynb
+2-2 b/‎site/en/guide/random_numbers.ipynb
+2-2
diff --git a/‎site/en/r1/guide/autograph.ipynb
+1-1 b/‎site/en/r1/guide/autograph.ipynb
+1-1
diff --git a/‎site/en/r1/guide/distribute_strategy.ipynb
+2-2 b/‎site/en/r1/guide/distribute_strategy.ipynb
+2-2
diff --git a/‎site/en/r1/guide/eager.ipynb
+1-1 b/‎site/en/r1/guide/eager.ipynb
+1-1
diff --git a/‎site/en/r1/guide/keras.ipynb
+1-1 b/‎site/en/r1/guide/keras.ipynb
+1-1
diff --git a/‎site/en/r1/guide/ragged_tensors.ipynb
+2-2 b/‎site/en/r1/guide/ragged_tensors.ipynb
+2-2
diff --git a/‎site/en/r1/tutorials/_index.ipynb
+1-1 b/‎site/en/r1/tutorials/_index.ipynb
+1-1
diff --git a/‎site/en/r1/tutorials/distribute/keras.ipynb
+2-2 b/‎site/en/r1/tutorials/distribute/keras.ipynb
+2-2
diff --git a/‎site/en/r1/tutorials/distribute/tpu_custom_training.ipynb
+1-1 b/‎site/en/r1/tutorials/distribute/tpu_custom_training.ipynb
+1-1
diff --git a/‎site/en/r1/tutorials/distribute/training_loops.ipynb
+1-1 b/‎site/en/r1/tutorials/distribute/training_loops.ipynb
+1-1
diff --git a/‎site/en/r1/tutorials/eager/automatic_differentiation.ipynb
+1-1 b/‎site/en/r1/tutorials/eager/automatic_differentiation.ipynb
+1-1
diff --git a/‎site/en/r1/tutorials/eager/custom_layers.ipynb
+1-1 b/‎site/en/r1/tutorials/eager/custom_layers.ipynb
+1-1
@@ -44,7 +44,7 @@
 if (sys.version_info.major, sys.version_info.minor) == (3, 6):
   REQUIRED_PKGS.append('dataclasses')
 
-VIS_REQURE = [
+VIS_REQUIRE = [
     'numpy',
     'PILLOW',
     'webp',
@@ -65,7 +65,7 @@
     package_dir={'': 'tools'},
     scripts=[],
     install_requires=REQUIRED_PKGS,
-    extras_require={'vis': VIS_REQURE},
+    extras_require={'vis': VIS_REQUIRE},
     classifiers=[
         'Development Status :: 4 - Beta',
         'Intended Audience :: Developers',
 
@@ -1101,7 +1101,7 @@
         "id": "M_x7ih5sarvG"
       },
       "source": [
-        "In this case, `batch_jacobian` still runs and returns _something_ with the expected shape, but it's contents have an unclear meaning:"
+        "In this case, `batch_jacobian` still runs and returns _something_ with the expected shape, but its contents have an unclear meaning:"
       ]
     },
     {
 
@@ -233,7 +233,7 @@
         "id": "rdpN_3ssG9D5"
       },
       "source": [
-        "The initial variables are set here in a fixed way, but Keras comes with any of a number of [initalizers](https://www.tensorflow.org/api_docs/python/tf/keras/initializers) you could use, with or without the rest of Keras."
+        "The initial variables are set here in a fixed way, but Keras comes with any of a number of [initializers](https://www.tensorflow.org/api_docs/python/tf/keras/initializers) you could use, with or without the rest of Keras."
       ]
     },
     {
 
@@ -472,7 +472,7 @@
         "\n",
         "Another common data source that can easily be ingested as a `tf.data.Dataset` is the python generator.\n",
         "\n",
-        "Caution: While this is a convienient approach it has limited portability and scalibility. It must run in the same python process that created the generator, and is still subject to the Python [GIL](https://en.wikipedia.org/wiki/Global_interpreter_lock)."
+        "Caution: While this is a convenient approach it has limited portability and scalability. It must run in the same python process that created the generator, and is still subject to the Python [GIL](https://en.wikipedia.org/wiki/Global_interpreter_lock)."
       ]
     },
     {
@@ -2631,7 +2631,7 @@
         "\n",
         "`data.Dataset.rejection_resample` takes a `class_func` argument. This `class_func` is applied to each dataset element, and is used to determine which class an example belongs to for the purposes of balancing.\n",
         "\n",
-        "The goal here is to balance the lable distribution, and the elements of `creditcard_ds` are already `(features, label)` pairs. So the `class_func` just needs to return those labels:"
+        "The goal here is to balance the label distribution, and the elements of `creditcard_ds` are already `(features, label)` pairs. So the `class_func` just needs to return those labels:"
       ]
     },
     {
 
@@ -225,7 +225,7 @@
         "\n",
         "Note: In order to avoid confusions between `Mesh` and `Layout`, the term *dimension* is always associated with `Mesh`, and the term *axis* with `Tensor` and `Layout` in this guide.\n",
         "\n",
-        "The rank of `Layout` should be the same as the rank of the `Tensor` where the `Layout` is applied. For each of the `Tensor`'s axes the `Layout` may specifiy a mesh dimension to shard the tensor across, or specify the axis as \"unsharded\".\n",
+        "The rank of `Layout` should be the same as the rank of the `Tensor` where the `Layout` is applied. For each of the `Tensor`'s axes the `Layout` may specify a mesh dimension to shard the tensor across, or specify the axis as \"unsharded\".\n",
         "The tensor is replicated across any mesh dimensions that it is not sharded across.\n",
         "\n",
         "The rank of a `Layout` and the number of dimensions of a `Mesh` do not need to match. The `unsharded` axes of a `Layout` do not need to be associated to a mesh dimension, and `unsharded` mesh dimensions do not need to be associated with a `layout` axis.\n",
@@ -528,7 +528,7 @@
         "\n",
         "So far you've worked with the `my_first_dtensor`, which is a rank-1 DTensor fully replicated across a dim-1 `Mesh`.\n",
         "\n",
-        "Next create and inspect DTensors that are sharded across a dim-2 `Mesh`. The next example does this with a 3x2 `Mesh` on 6 CPU devices, where size of mesh dimension `'x'` is 3 devices, and and size of mesh dimension`'y'` is 2 devices."
+        "Next create and inspect DTensors that are sharded across a dim-2 `Mesh`. The next example does this with a 3x2 `Mesh` on 6 CPU devices, where size of mesh dimension `'x'` is 3 devices, and size of mesh dimension`'y'` is 2 devices."
       ]
     },
     {
@@ -696,7 +696,7 @@
         "To accomplish this, for each [TensorFlow Graph](https://www.tensorflow.org/guide/intro_to_graphs), DTensor produces and executes an equivalent [SPMD](https://en.wikipedia.org/wiki/SPMD) graph in a procedure called *SPMD expansion*. A few critical steps in DTensor SPMD expansion are:\n",
         "\n",
         "  - Propagating the sharding `Layout` of DTensor in the TensorFlow graph\n",
-        "  - Rewriting TensorFlow Ops on the global DTensor with equivalent TensorFlow Ops on the componenent tensors, inserting collective and communication Ops when necessary\n",
+        "  - Rewriting TensorFlow Ops on the global DTensor with equivalent TensorFlow Ops on the component tensors, inserting collective and communication Ops when necessary\n",
         "  - Lowering backend neutral TensorFlow Ops to backend specific TensorFlow Ops.\n",
         "\n",
         "The final result is that **DTensor is a drop-in replacement for Tensor**.\n",
 
@@ -90,7 +90,7 @@
         "* *Constant folding optimizer -* Statically infers the value of tensors when possible by folding constant nodes in the graph and materializes the result using constants.\n",
         "* *Arithmetic optimizer -* Simplifies arithmetic operations by eliminating common subexpressions and simplifying arithmetic statements.  \n",
         "* *Layout optimizer -* Optimizes tensor layouts to execute data format dependent operations such as convolutions more efficiently.\n",
-        "* *Remapper optimizer -* Remaps subgraphs onto more efficient implementations by replacing commonly occuring subgraphs with optimized fused monolithic kernels.\n",
+        "* *Remapper optimizer -* Remaps subgraphs onto more efficient implementations by replacing commonly occurring subgraphs with optimized fused monolithic kernels.\n",
         "* *Memory optimizer -* Analyzes the graph to inspect the peak memory usage for each operation and inserts CPU-GPU memory copy operations for swapping GPU memory to CPU to reduce the peak memory usage.\n",
         "* *Dependency optimizer -* Removes or rearranges control dependencies to shorten the critical path for a model step or enables other\n",
         "optimizations. Also removes nodes that are effectively no-ops such as Identity.\n",
 
@@ -676,7 +676,7 @@
         "  tf.gather(x, [1]) # unused\n",
         "  return x\n",
         "\n",
-        "# Only needed operations are run during graph exection. The error is not raised.\n",
+        "# Only needed operations are run during graph execution. The error is not raised.\n",
         "print(unused_return_graph(tf.constant([0.0])))"
       ]
     },
 
@@ -718,7 +718,7 @@
         "id": "Z0QYolhoZb_k"
       },
       "source": [
-        "Finaly, in the next example, we train and evaluate a CART model."
+        "Finally, in the next example, we train and evaluate a CART model."
       ]
     },
     {
 
@@ -193,7 +193,7 @@
       "source": [
         "## TensorFlow 2: Evaluating a Keras model\n",
         "\n",
-        "In TensorFlow 2, if you use the Keras `Model.fit` API for training, you can evaluate the model with `tf.keras.utils.SidecarEvaluator`. You can also visualize the evaluation metrics in Tensorboard which is not shown in this guide.\n",
+        "In TensorFlow 2, if you use the Keras `Model.fit` API for training, you can evaluate the model with `tf.keras.utils.SidecarEvaluator`. You can also visualize the evaluation metrics in TensorBoard which is not shown in this guide.\n",
         "\n",
         "To help demonstrate this, let's first start by defining and training the model:\n"
       ]
 
@@ -961,7 +961,7 @@
         "\n",
         "\\* `output_mode` can be passed to `layers.CategoryEncoding`, `layers.StringLookup`, `layers.IntegerLookup`, and `layers.TextVectorization`.\n",
         "\n",
-        "† `layers.TextVectorization` can handle freeform text input directly (e.g. entire sentences or paragraphs). This is not one-to-one replacement for categorical sequence handling in TF1, but may offer a convinient replacement for ad-hoc text preprocessing.\n",
+        "† `layers.TextVectorization` can handle freeform text input directly (e.g. entire sentences or paragraphs). This is not one-to-one replacement for categorical sequence handling in TF1, but may offer a convenient replacement for ad-hoc text preprocessing.\n",
         "\n",
         "Note: Linear estimators, such as `tf.estimator.LinearClassifier`, can handle direct categorical input (integer indices) without an `embedding_column` or `indicator_column`. However, integer indices cannot be passed directly to `tf.keras.layers.Dense` or `tf.keras.experimental.LinearModel`. These inputs should be first encoded with `tf.layers.CategoryEncoding` with `output_mode='count'` (and `sparse=True` if the category sizes are large) before calling into `Dense` or `LinearModel`."
       ]
 
@@ -486,7 +486,7 @@
       "source": [
         "Compare numerical equivalence for first few training steps.\n",
         "\n",
-        "You can also check the [Validating correctness & numerical equivalence notebook](./validate_correctness.ipynb) for additonal advice for numerical equivalence."
+        "You can also check the [Validating correctness & numerical equivalence notebook](./validate_correctness.ipynb) for additional advice for numerical equivalence."
       ]
     },
     {
 
@@ -340,7 +340,7 @@
         "  print(e) # <tf.Tensor ...> is out of scope and cannot be used here.\n",
         "```\n",
         "\n",
-        "The most straightfoward solution is ensuring that the variable creation and dataset creation are both outside of the `tf.funciton` call. For example:\n",
+        "The most straightfoward solution is ensuring that the variable creation and dataset creation are both outside of the `tf.function` call. For example:\n",
         "\n",
         "```python\n",
         "class Model(tf.Module):\n",
 
@@ -108,7 +108,7 @@
       "source": [
         "## Recommended upgrade process\n",
         "\n",
-        "The rest of this guide demonstrates how to use the symbol-rewriting script. While the script is easy to use, it is strongly recomended that you use the script as part of the following process: \n",
+        "The rest of this guide demonstrates how to use the symbol-rewriting script. While the script is easy to use, it is strongly recommended that you use the script as part of the following process: \n",
         "\n",
         "1. **Unit Test**: Ensure that the code you’re upgrading has a unit test suite with reasonable coverage. This is Python code, so the language won’t protect you from many classes of mistakes. Also ensure that any dependency you have has already been upgraded to be compatible with TensorFlow 2.x.\n",
         "\n",
 
@@ -1440,7 +1440,7 @@
         "\n",
         "1.  Use `tf.RaggedTensor.to_list` to convert the ragged tensor to a nested Python list.\n",
         "2.  Use `tf.RaggedTensor.numpy` to convert the ragged tensor to a NumPy array whose values are nested NumPy arrays.\n",
-        "3.  Decompose the ragged tensor into its components, using the `tf.RaggedTensor.values` and `tf.RaggedTensor.row_splits` properties, or row-paritioning methods such as `tf.RaggedTensor.row_lengths` and `tf.RaggedTensor.value_rowids`.\n",
+        "3.  Decompose the ragged tensor into its components, using the `tf.RaggedTensor.values` and `tf.RaggedTensor.row_splits` properties, or row-partitioning methods such as `tf.RaggedTensor.row_lengths` and `tf.RaggedTensor.value_rowids`.\n",
         "4.  Use Python indexing to select values from the ragged tensor.\n"
       ]
     },
 
@@ -268,7 +268,7 @@
       "source": [
         "Note: In theory, you can use constructors such as `from_seed` instead of `split` here to obtain a new generator, but by doing so you lose the guarantee that the new generator is independent of the global generator. You will also run the risk that you may accidentally create two generators with the same seed or with seeds that lead to overlapping random-number streams.\n",
         "\n",
-        "You can do splitting recursively, calling `split` on splitted generators. There are no limits (barring integer overflow) on the depth of recursions."
+        "You can do splitting recursively, calling `split` on split generators. There are no limits (barring integer overflow) on the depth of recursions."
       ]
     },
     {
@@ -325,7 +325,7 @@
       "source": [
         "#### Creating generators inside `tf.function` \n",
         "\n",
-        "Creation of generators inside a `tf.function` can only happend during the first run of the function. "
+        "Creation of generators inside a `tf.function` can only happened during the first run of the function. "
       ]
     },
     {
 
@@ -66,7 +66,7 @@
       "source": [
         "> Note: This is an archived TF1 notebook. These are configured\n",
         "to run in TF2's \n",
-        "[compatbility mode](https://www.tensorflow.org/guide/migrate)\n",
+        "[compatibility mode](https://www.tensorflow.org/guide/migrate)\n",
         "but will run in TF1 as well. To use TF1 in Colab, use the\n",
         "[%tensorflow_version 1.x](https://colab.research.google.com/notebooks/tensorflow_version.ipynb)\n",
         "magic."
 
@@ -64,7 +64,7 @@
       "source": [
         "> Note: This is an archived TF1 notebook. These are configured\n",
         "to run in TF2's \n",
-        "[compatbility mode](https://www.tensorflow.org/guide/migrate)\n",
+        "[compatibility mode](https://www.tensorflow.org/guide/migrate)\n",
         "but will run in TF1 as well. To use TF1 in Colab, use the\n",
         "[%tensorflow_version 1.x](https://colab.research.google.com/notebooks/tensorflow_version.ipynb)\n",
         "magic."
@@ -223,7 +223,7 @@
         "id": "KY1nJHNkMl7b"
       },
       "source": [
-        "This will create a `CentralStorageStrategy` instance which will use all visible GPUs and CPU. Update to variables on replicas will be aggragated before being applied to variables."
+        "This will create a `CentralStorageStrategy` instance which will use all visible GPUs and CPU. Update to variables on replicas will be aggregated before being applied to variables."
       ]
     },
     {
 
@@ -64,7 +64,7 @@
       "source": [
         "> Note: This is an archived TF1 notebook. These are configured\n",
         "to run in TF2's \n",
-        "[compatbility mode](https://www.tensorflow.org/guide/migrate)\n",
+        "[compatibility mode](https://www.tensorflow.org/guide/migrate)\n",
         "but will run in TF1 as well. To use TF1 in Colab, use the\n",
         "[%tensorflow_version 1.x](https://colab.research.google.com/notebooks/tensorflow_version.ipynb)\n",
         "magic."
 
@@ -64,7 +64,7 @@
       "source": [
         "> Note: This is an archived TF1 notebook. These are configured\n",
         "to run in TF2's \n",
-        "[compatbility mode](https://www.tensorflow.org/guide/migrate)\n",
+        "[compatibility mode](https://www.tensorflow.org/guide/migrate)\n",
         "but will run in TF1 as well. To use TF1 in Colab, use the\n",
         "[%tensorflow_version 1.x](https://colab.research.google.com/notebooks/tensorflow_version.ipynb)\n",
         "magic."
 
@@ -57,7 +57,7 @@
       "source": [
         "> Note: This is an archived TF1 notebook. These are configured\n",
         "to run in TF2's \n",
-        "[compatbility mode](https://www.tensorflow.org/guide/migrate)\n",
+        "[compatibility mode](https://www.tensorflow.org/guide/migrate)\n",
         "but will run in TF1 as well. To use TF1 in Colab, use the\n",
         "[%tensorflow_version 1.x](https://colab.research.google.com/notebooks/tensorflow_version.ipynb)\n",
         "magic."
@@ -1010,7 +1010,7 @@
         "    `tf.RaggedTensor.values`\n",
         "    and\n",
         "    `tf.RaggedTensor.row_splits`\n",
-        "    properties, or row-paritioning methods such as `tf.RaggedTensor.row_lengths()`\n",
+        "    properties, or row-partitioning methods such as `tf.RaggedTensor.row_lengths()`\n",
         "    and `tf.RaggedTensor.value_rowids()`."
       ]
     },
 
@@ -64,7 +64,7 @@
       "source": [
         "> Note: This is an archived TF1 notebook. These are configured\n",
         "to run in TF2's \n",
-        "[compatbility mode](https://www.tensorflow.org/guide/migrate)\n",
+        "[compatibility mode](https://www.tensorflow.org/guide/migrate)\n",
         "but will run in TF1 as well. To use TF1 in Colab, use the\n",
         "[%tensorflow_version 1.x](https://colab.research.google.com/notebooks/tensorflow_version.ipynb)\n",
         "magic."
 
@@ -64,7 +64,7 @@
       "source": [
         "> Note: This is an archived TF1 notebook. These are configured\n",
         "to run in TF2's \n",
-        "[compatbility mode](https://www.tensorflow.org/guide/migrate)\n",
+        "[compatibility mode](https://www.tensorflow.org/guide/migrate)\n",
         "but will run in TF1 as well. To use TF1 in Colab, use the\n",
         "[%tensorflow_version 1.x](https://colab.research.google.com/notebooks/tensorflow_version.ipynb)\n",
         "magic."
@@ -345,7 +345,7 @@
       "source": [
         "The callbacks used here are:\n",
         "\n",
-        "*   *Tensorboard*: This callback writes a log for Tensorboard which allows you to visualize the graphs.\n",
+        "*   *TensorBoard*: This callback writes a log for TensorBoard which allows you to visualize the graphs.\n",
         "*   *Model Checkpoint*: This callback saves the model after every epoch.\n",
         "*   *Learning Rate Scheduler*: Using this callback, you can schedule the learning rate to change after every epoch/batch.\n",
         "\n",
 
@@ -64,7 +64,7 @@
       "source": [
         "> Note: This is an archived TF1 notebook. These are configured\n",
         "to run in TF2's \n",
-        "[compatbility mode](https://www.tensorflow.org/guide/migrate)\n",
+        "[compatibility mode](https://www.tensorflow.org/guide/migrate)\n",
         "but will run in TF1 as well. To use TF1 in Colab, use the\n",
         "[%tensorflow_version 1.x](https://colab.research.google.com/notebooks/tensorflow_version.ipynb)\n",
         "magic."
 
@@ -64,7 +64,7 @@
       "source": [
         "> Note: This is an archived TF1 notebook. These are configured\n",
         "to run in TF2's \n",
-        "[compatbility mode](https://www.tensorflow.org/guide/migrate)\n",
+        "[compatibility mode](https://www.tensorflow.org/guide/migrate)\n",
         "but will run in TF1 as well. To use TF1 in Colab, use the\n",
         "[%tensorflow_version 1.x](https://colab.research.google.com/notebooks/tensorflow_version.ipynb)\n",
         "magic."
 
@@ -64,7 +64,7 @@
       "source": [
         "> Note: This is an archived TF1 notebook. These are configured\n",
         "to run in TF2's \n",
-        "[compatbility mode](https://www.tensorflow.org/guide/migrate)\n",
+        "[compatibility mode](https://www.tensorflow.org/guide/migrate)\n",
         "but will run in TF1 as well. To use TF1 in Colab, use the\n",
         "[%tensorflow_version 1.x](https://colab.research.google.com/notebooks/tensorflow_version.ipynb)\n",
         "magic."
 
@@ -64,7 +64,7 @@
       "source": [
         "> Note: This is an archived TF1 notebook. These are configured\n",
         "to run in TF2's \n",
-        "[compatbility mode](https://www.tensorflow.org/guide/migrate)\n",
+        "[compatibility mode](https://www.tensorflow.org/guide/migrate)\n",
         "but will run in TF1 as well. To use TF1 in Colab, use the\n",
         "[%tensorflow_version 1.x](https://colab.research.google.com/notebooks/tensorflow_version.ipynb)\n",
         "magic."
Original file line number	Diff line number	Diff line change
`@@ -1101,7 +1101,7 @@`
`1101`	`1101`	`"id": "M_x7ih5sarvG"`
`1102`	`1102`	`},`
`1103`	`1103`	`"source": [`
`1104`		- "In this case, `batch_jacobian` still runs and returns _something_ with the expected shape, but it's contents have an unclear meaning:"
	`1104`	+ "In this case, `batch_jacobian` still runs and returns _something_ with the expected shape, but its contents have an unclear meaning:"
`1105`	`1105`	`]`
`1106`	`1106`	`},`
`1107`	`1107`	`{`
Original file line number	Diff line number	Diff line change
`@@ -233,7 +233,7 @@`
`233`	`233`	`"id": "rdpN_3ssG9D5"`
`234`	`234`	`},`
`235`	`235`	`"source": [`
`236`		`- "The initial variables are set here in a fixed way, but Keras comes with any of a number of [initalizers](https://www.tensorflow.org/api_docs/python/tf/keras/initializers) you could use, with or without the rest of Keras."`
	`236`	`+ "The initial variables are set here in a fixed way, but Keras comes with any of a number of [initializers](https://www.tensorflow.org/api_docs/python/tf/keras/initializers) you could use, with or without the rest of Keras."`
`237`	`237`	`]`
`238`	`238`	`},`
`239`	`239`	`{`
Original file line number	Diff line number	Diff line change
`@@ -472,7 +472,7 @@`
`472`	`472`	`"\n",`
`473`	`473`	"Another common data source that can easily be ingested as a `tf.data.Dataset` is the python generator.\n",
`474`	`474`	`"\n",`
`475`		`- "Caution: While this is a convienient approach it has limited portability and scalibility. It must run in the same python process that created the generator, and is still subject to the Python [GIL](https://en.wikipedia.org/wiki/Global_interpreter_lock)."`
	`475`	`+ "Caution: While this is a convenient approach it has limited portability and scalability. It must run in the same python process that created the generator, and is still subject to the Python [GIL](https://en.wikipedia.org/wiki/Global_interpreter_lock)."`
`476`	`476`	`]`
`477`	`477`	`},`
`478`	`478`	`{`
`@@ -2631,7 +2631,7 @@`
`2631`	`2631`	`"\n",`
`2632`	`2632`	"`data.Dataset.rejection_resample` takes a `class_func` argument. This `class_func` is applied to each dataset element, and is used to determine which class an example belongs to for the purposes of balancing.\n",
`2633`	`2633`	`"\n",`
`2634`		- "The goal here is to balance the lable distribution, and the elements of `creditcard_ds` are already `(features, label)` pairs. So the `class_func` just needs to return those labels:"
	`2634`	+ "The goal here is to balance the label distribution, and the elements of `creditcard_ds` are already `(features, label)` pairs. So the `class_func` just needs to return those labels:"
`2635`	`2635`	`]`
`2636`	`2636`	`},`
`2637`	`2637`	`{`
Original file line number	Diff line number	Diff line change
`@@ -676,7 +676,7 @@`
`676`	`676`	`" tf.gather(x, [1]) # unused\n",`
`677`	`677`	`" return x\n",`
`678`	`678`	`"\n",`
`679`		`- "# Only needed operations are run during graph exection. The error is not raised.\n",`
	`679`	`+ "# Only needed operations are run during graph execution. The error is not raised.\n",`
`680`	`680`	`"print(unused_return_graph(tf.constant([0.0])))"`
`681`	`681`	`]`
`682`	`682`	`},`
Original file line number	Diff line number	Diff line change
`@@ -718,7 +718,7 @@`
`718`	`718`	`"id": "Z0QYolhoZb_k"`
`719`	`719`	`},`
`720`	`720`	`"source": [`
`721`		`- "Finaly, in the next example, we train and evaluate a CART model."`
	`721`	`+ "Finally, in the next example, we train and evaluate a CART model."`
`722`	`722`	`]`
`723`	`723`	`},`
`724`	`724`	`{`
Original file line number	Diff line number	Diff line change
`@@ -193,7 +193,7 @@`
`193`	`193`	`"source": [`
`194`	`194`	`"## TensorFlow 2: Evaluating a Keras model\n",`
`195`	`195`	`"\n",`
`196`		- "In TensorFlow 2, if you use the Keras `Model.fit` API for training, you can evaluate the model with `tf.keras.utils.SidecarEvaluator`. You can also visualize the evaluation metrics in Tensorboard which is not shown in this guide.\n",
	`196`	+ "In TensorFlow 2, if you use the Keras `Model.fit` API for training, you can evaluate the model with `tf.keras.utils.SidecarEvaluator`. You can also visualize the evaluation metrics in TensorBoard which is not shown in this guide.\n",
`197`	`197`	`"\n",`
`198`	`198`	`"To help demonstrate this, let's first start by defining and training the model:\n"`
`199`	`199`	`]`
Original file line number	Diff line number	Diff line change
`@@ -961,7 +961,7 @@`
`961`	`961`	`"\n",`
`962`	`962`	"\\* `output_mode` can be passed to `layers.CategoryEncoding`, `layers.StringLookup`, `layers.IntegerLookup`, and `layers.TextVectorization`.\n",
`963`	`963`	`"\n",`
`964`		- "† `layers.TextVectorization` can handle freeform text input directly (e.g. entire sentences or paragraphs). This is not one-to-one replacement for categorical sequence handling in TF1, but may offer a convinient replacement for ad-hoc text preprocessing.\n",
	`964`	+ "† `layers.TextVectorization` can handle freeform text input directly (e.g. entire sentences or paragraphs). This is not one-to-one replacement for categorical sequence handling in TF1, but may offer a convenient replacement for ad-hoc text preprocessing.\n",
`965`	`965`	`"\n",`
`966`	`966`	"Note: Linear estimators, such as `tf.estimator.LinearClassifier`, can handle direct categorical input (integer indices) without an `embedding_column` or `indicator_column`. However, integer indices cannot be passed directly to `tf.keras.layers.Dense` or `tf.keras.experimental.LinearModel`. These inputs should be first encoded with `tf.layers.CategoryEncoding` with `output_mode='count'` (and `sparse=True` if the category sizes are large) before calling into `Dense` or `LinearModel`."
`967`	`967`	`]`
Original file line number	Diff line number	Diff line change
`@@ -486,7 +486,7 @@`
`486`	`486`	`"source": [`
`487`	`487`	`"Compare numerical equivalence for first few training steps.\n",`
`488`	`488`	`"\n",`
`489`		`- "You can also check the [Validating correctness & numerical equivalence notebook](./validate_correctness.ipynb) for additonal advice for numerical equivalence."`
	`489`	`+ "You can also check the [Validating correctness & numerical equivalence notebook](./validate_correctness.ipynb) for additional advice for numerical equivalence."`
`490`	`490`	`]`
`491`	`491`	`},`
`492`	`492`	`{`
Original file line number	Diff line number	Diff line change
`@@ -1440,7 +1440,7 @@`
`1440`	`1440`	`"\n",`
`1441`	`1441`	"1. Use `tf.RaggedTensor.to_list` to convert the ragged tensor to a nested Python list.\n",
`1442`	`1442`	"2. Use `tf.RaggedTensor.numpy` to convert the ragged tensor to a NumPy array whose values are nested NumPy arrays.\n",
`1443`		- "3. Decompose the ragged tensor into its components, using the `tf.RaggedTensor.values` and `tf.RaggedTensor.row_splits` properties, or row-paritioning methods such as `tf.RaggedTensor.row_lengths` and `tf.RaggedTensor.value_rowids`.\n",
	`1443`	+ "3. Decompose the ragged tensor into its components, using the `tf.RaggedTensor.values` and `tf.RaggedTensor.row_splits` properties, or row-partitioning methods such as `tf.RaggedTensor.row_lengths` and `tf.RaggedTensor.value_rowids`.\n",
`1444`	`1444`	`"4. Use Python indexing to select values from the ragged tensor.\n"`
`1445`	`1445`	`]`
`1446`	`1446`	`},`
Original file line number	Diff line number	Diff line change
`@@ -268,7 +268,7 @@`
`268`	`268`	`"source": [`
`269`	`269`	"Note: In theory, you can use constructors such as `from_seed` instead of `split` here to obtain a new generator, but by doing so you lose the guarantee that the new generator is independent of the global generator. You will also run the risk that you may accidentally create two generators with the same seed or with seeds that lead to overlapping random-number streams.\n",
`270`	`270`	`"\n",`
`271`		- "You can do splitting recursively, calling `split` on splitted generators. There are no limits (barring integer overflow) on the depth of recursions."
	`271`	+ "You can do splitting recursively, calling `split` on split generators. There are no limits (barring integer overflow) on the depth of recursions."
`272`	`272`	`]`
`273`	`273`	`},`
`274`	`274`	`{`
`@@ -325,7 +325,7 @@`
`325`	`325`	`"source": [`
`326`	`326`	"#### Creating generators inside `tf.function` \n",
`327`	`327`	`"\n",`
`328`		- "Creation of generators inside a `tf.function` can only happend during the first run of the function. "
	`328`	+ "Creation of generators inside a `tf.function` can only happened during the first run of the function. "
`329`	`329`	`]`
`330`	`330`	`},`
`331`	`331`	`{`