Fix documentation issues

DOC 

* Update docs for fromPixels
* Improve docs for tidy

src/ops/array_ops.ts

@@ -280,8 +280,7 @@ export class ArrayOps {
     }
     const inferredShape = util.inferShape(values);
     if (inferredShape.length !== 5 && inferredShape.length !== 1) {
-      throw new Error(
-          'tensor5d() requires values to be \
+      throw new Error('tensor5d() requires values to be \
            number[][][][][] or flat/TypedArray');
     }
     if (inferredShape.length === 1 && shape == null) {
@@ -674,10 +673,11 @@ export class ArrayOps {
    * tf.fromPixels(image).print();
    * ```
    *
-   * @param pixels The input image to construct the tensor from.
-   * @param numChannels The number of channels of the output tensor. The
-   * supported image types are all 4-channel by default, a numChannels value
-   * less than 4 allows you to ignore channels.
+   * @param pixels The input image to construct the tensor from. The
+   * supported image types are all 4-channel.
+   * @param numChannels The number of channels of the output tensor. A
+   * numChannels value less than 4 allows you to ignore channels. Defaults to
+   * 3 (ignores alpha channel of input image).
    */
   @doc({heading: 'Tensors', subheading: 'Creation'})
   @operation

src/tracking.ts

@@ -24,8 +24,8 @@ import {extractTensorsFromAny} from './util';
 
 export class Tracking {
   /**
-   * Executes the provided function `f` and after it is executed, cleans up all
-   * intermediate tensors allocated by `f` except those returned by `f`.
+   * Executes the provided function `fn` and after it is executed, cleans up all
+   * intermediate tensors allocated by `fn` except those returned by `fn`.
    * `f` must not return a Promise (async functions not allowed).
    * The returned result can be a complex object, however tidy only walks the
    * top-level properties (depth 1) of that object to search for tensors, or
@@ -61,11 +61,13 @@ export class Tracking {
    *     If debug mode is on, the timing and the memory usage of the function
    *     will be tracked and displayed on the console using the provided name.
    * @param fn The function to execute.
-   * @param gradMode If true, starts a tape and doesn't dispose tensors.
    */
   @doc({heading: 'Performance', subheading: 'Memory'})
   static tidy<T extends TensorContainer>(
       nameOrFn: string|ScopeFn<T>, fn?: ScopeFn<T>, gradMode = false): T {
+    // gradMode Primarily for internal use during backprop
+    //          If true, will start a tape if it is the outermost tidy.
+
     let name = null;
     if (fn == null) {
       // Called with only 1 argument.