Gh140 configure auto build spark

jupyter/SparkOcrDicomDeIdentification.ipynb

@@ -175,8 +175,8 @@
     "spark = start(secret=secret,\n",
     "              nlp_secret=nlp_secret,\n",
     "              jar_path=spark_ocr_jar_path,\n",
-    "              nlp_version=\"2.5.0\",\n",
-    "              nlp_internal=True)\n",
+    "              nlp_version=\"2.5.5\",\n",
+    "              nlp_internal=\"2.5.5\")\n",
     "\n",
     "spark"
    ]
@@ -212,15 +212,18 @@
     }
    ],
    "source": [
-    "from sparknlp.annotator import *\n",
-    "from sparknlp.base import *\n",
-    "import sparknlp_jsl\n",
-    "from sparknlp_jsl.annotator import *\n",
+    "\n",
     "\n",
     "import sparkocr\n",
     "from sparkocr.transformers import *\n",
     "from sparkocr.utils import display_image\n",
     "\n",
+    "import sparknlp\n",
+    "import sparknlp_jsl\n",
+    "from sparknlp.base import *\n",
+    "from sparknlp.annotator import *\n",
+    "from sparknlp_jsl.annotator import *\n",
+    "\n",
     "print(f\"Spark NLP version: {sparknlp.version()}\")\n",
     "print(f\"Spark NLP internal version: {sparknlp_jsl.version()}\")\n",
     "print(f\"Spark OCR version: {sparkocr.version()}\")"
@@ -1251,11 +1254,11 @@
     "dicom_path = '././data/dicom/*.dcm'\n",
     "dicom_df = spark.read.format(\"binaryFile\").load(dicom_path)\n",
     "\n",
+    "\n",
     "# for image in DicomToImage().transform(dicom_df).collect():\n",
     "#     print(image.exception)\n",
     "#     print(image.metadata)\n",
-    "#     display_image(image.image)\n",
-    "   "
+    "#     display_image(image.image)\n"
    ]
   },
   {
@@ -2225,4 +2228,4 @@
  },
  "nbformat": 4,
  "nbformat_minor": 2
-}
+}

jupyter/SparkOcrPositionFinder.ipynb

@@ -70,7 +70,7 @@
    "outputs": [],
    "source": [
     "# install from PYPI using secret\n",
-    "%pip install spark-nlp==2.4.3\n",
+    "%pip install spark-nlp==2.5.5\n",
     "%pip install spark-ocr==$version --user --extra-index-url=https://pypi.johnsnowlabs.com/$secret --upgrade"
    ]
   },
@@ -145,7 +145,7 @@
     "if license:\n",
     "    os.environ['JSL_OCR_LICENSE'] = license\n",
     "\n",
-    "spark = start(secret=secret, jar_path=spark_ocr_jar_path, nlp_version=\"2.4.3\")\n",
+    "spark = start(secret=secret, jar_path=spark_ocr_jar_path, nlp_version=\"2.5.5\")\n",
     "spark"
    ]
   },
@@ -417,4 +417,4 @@
  },
  "nbformat": 4,
  "nbformat_minor": 2
-}
+}

jupyter/SparkOcrUpdateTextPosition.ipynb

@@ -70,7 +70,7 @@
    "outputs": [],
    "source": [
     "# install from PYPI using secret\n",
-    "%pip install spark-nlp==2.4.5\n",
+    "%pip install spark-nlp==2.5.5\n",
     "%pip install spark-ocr==$version --user --extra-index-url=https://pypi.johnsnowlabs.com/$secret --upgrade"
    ]
   },
@@ -144,7 +144,7 @@
     "if license:\n",
     "    os.environ['JSL_OCR_LICENSE'] = license\n",
     "\n",
-    "spark = start(secret=secret, jar_path=spark_ocr_jar_path, nlp_version=\"2.4.5\")\n",
+    "spark = start(secret=secret, jar_path=spark_ocr_jar_path, nlp_version=\"2.5.5\")\n",
     "spark"
    ]
   },
@@ -190,13 +190,14 @@
     "        .setOutputCol(\"tokens\")\n",
     "\n",
     "    spell = NorvigSweetingModel().pretrained(\"spellcheck_norvig\", \"en\") \\\n",
-    "          .setInputCols(\"tokens\") \\\n",
+    "          .setInputCols([\"tokens\"]) \\\n",
     "          .setOutputCol(\"spell\")\n",
     "    \n",
     "    tokenAssem = TokenAssembler() \\\n",
-    "          .setInputCols(\"spell\") \\\n",
+    "          .setInputCols([\"document\",\"spell\"]) \\\n",
     "          .setOutputCol(\"newDocs\")\n",
     "\n",
+    "\n",
     "    updatedText = UpdateTextPosition() \\\n",
     "          .setInputCol(\"positions\") \\\n",
     "          .setOutputCol(\"output_positions\") \\\n",
@@ -346,4 +347,4 @@
  },
  "nbformat": 4,
  "nbformat_minor": 2
-}
+}