Update handwritten-english-recognition-0001 (#2996)

* Update reference metric

Signed-off-by: Junze Wu <[email protected]>

* Update inaccurate help message

Signed-off-by: Junze Wu <[email protected]>

Author: jnzw

demos/handwritten_text_recognition_demo/python/README.md

@@ -77,7 +77,10 @@ Options:
                         Optional. Top k steps in looking up the decoded
                         character, until a designated one is found
   -ob OUTPUT_BLOB, --output_blob OUTPUT_BLOB
-                        Optional. Name of the output layer of the model. Default is 'output'
+                        Optional. Name of the output layer of the model.
+                        Default is None, in which case the demo will read
+                        the output name from the model, assuming there is
+                        only 1 output layer
 ```
 
 The decoding char list files provided within Open Model Zoo and for Japanese it is the `<omz_dir>/data/dataset_classes/kondate_nakayosi.txt` file, while for Simplified Chinese it is the `<omz_dir>/data/dataset_classes/scut_ept.txt` file, and for English it is the `<omz_dir>/data/dataset_classes/gnhk.txt` file. For example, to do inference on a CPU with the OpenVINO&trade; toolkit pre-trained `handwritten-japanese-recognition-0001` model, run the following command:

models/intel/handwritten-english-recognition-0001/README.md

@@ -15,7 +15,7 @@ The network is able to recognize English text consisting of characters in the [G
 | ------------------------- | --------- |
 | GFlops                    | 1.3182    |
 | MParams                   | 0.1413    |
-| Accuracy on GNHK test subset (excluding images wider than 2000px after resized to height 96px with aspect ratio) | 81.5%     |
+| Accuracy on GNHK test subset (excluding images wider than 2000px after resized to height 96px with aspect ratio) | 82.0%     |
 | Source framework          | PyTorch\* |
 
 > **Note:** to achieve the accuracy, images from the GNHK test set should be binarized using adaptive thresholding, and preprocessed into single-line text images, using the coordinates from the accompanying JSON annotation files in the GNHK dataset. See `<omz_dir>/models/intel/handwritten-english-recognition-0001/preprocess_gnhk.py`.

models/intel/handwritten-english-recognition-0001/accuracy-check.yml

@@ -27,4 +27,4 @@ models:
 
         metrics:
           - type: label_level_recognition_accuracy
-            reference: 0.815
+            reference: 0.820