Merge pull request #2608 from komaljoshi-SSK/ocbl169-notebook-updates

OCBL169 - Notebook updates
GoogleCloudPlatform · May 10, 2024 · bd8362a · bd8362a
2 parents 7cec0f8 + 31b94d7
commit bd8362a
Show file tree

Hide file tree

Showing 2 changed files with 12 additions and 12 deletions.
diff --git a/...s/machine_learning/deepdive2/text_classification/labs/keras_for_text_classification.ipynb b/...s/machine_learning/deepdive2/text_classification/labs/keras_for_text_classification.ipynb
@@ -74,10 +74,10 @@
    "source": [
     "## Create a Dataset from BigQuery \n",
     "\n",
-    "Hacker news headlines are available as a BigQuery public dataset. The [dataset](https://console.cloud.google.com/bigquery?project=bigquery-public-data&page=table&t=stories&d=hacker_news&p=bigquery-public-data&redirect_from_classic=true) contains all headlines from the sites inception in October 2006 until October 2015. \n",
+    "Hacker news headlines are available as a BigQuery public dataset. The [dataset](https://console.cloud.google.com/bigquery?project=bigquery-public-data&page=table&t=full&d=hacker_news&p=bigquery-public-data&redirect_from_classic=true) contains all headlines from the sites inception in October 2006 until October 2015. \n",
     "\n",
     "### Lab Task 1a: \n",
-    "Complete the query below to create a sample dataset containing the `url`, `title`, and `score` of articles from the public dataset `bigquery-public-data.hacker_news.stories`. Use a WHERE clause to restrict to only those articles with\n",
+    "Complete the query below to create a sample dataset containing the `url`, `title`, and `score` of articles from the public dataset `bigquery-public-data.hacker_news.full`. Use a WHERE clause to restrict to only those articles with\n",
     "* title length greater than 10 characters\n",
     "* score greater than 10\n",
     "* url length greater than 0 characters"
@@ -126,10 +126,10 @@
     "%%bigquery --project $PROJECT\n",
     "\n",
     "SELECT\n",
-    "    ARRAY_REVERSE(SPLIT(REGEXP_EXTRACT(url, '.*://(.[^/]+)/'), '.'))[OFFSET(1)] AS source,\n",
+    "    ARRAY_REVERSE(SPLIT(REGEXP_EXTRACT(url, '.*://(.[^/]+)/'), '.'))[safe_offset (1)] AS source,\n",
     "    # TODO: Your code goes here.\n",
     "FROM\n",
-    "    `bigquery-public-data.hacker_news.stories`\n",
+    "    `bigquery-public-data.hacker_news.full`\n",
     "WHERE\n",
     "    REGEXP_CONTAINS(REGEXP_EXTRACT(url, '.*://(.[^/]+)/'), '.com$')\n",
     "    # TODO: Your code goes here.\n",
@@ -158,10 +158,10 @@
     "sub_query = \"\"\"\n",
     "SELECT\n",
     "    title,\n",
-    "    ARRAY_REVERSE(SPLIT(REGEXP_EXTRACT(url, '{0}'), '.'))[OFFSET(1)] AS source\n",
+    "    ARRAY_REVERSE(SPLIT(REGEXP_EXTRACT(url, '{0}'), '.'))[safe_offset (1)] AS source\n",
     "    \n",
     "FROM\n",
-    "    `bigquery-public-data.hacker_news.stories`\n",
+    "    `bigquery-public-data.hacker_news.full`\n",
     "WHERE\n",
     "    REGEXP_CONTAINS(REGEXP_EXTRACT(url, '{0}'), '.com$')\n",
     "    AND LENGTH(title) > 10\n",

diff --git a/...hine_learning/deepdive2/text_classification/solutions/keras_for_text_classification.ipynb b/...hine_learning/deepdive2/text_classification/solutions/keras_for_text_classification.ipynb
@@ -74,7 +74,7 @@
    "source": [
     "## Create a Dataset from BigQuery \n",
     "\n",
-    "Hacker news headlines are available as a BigQuery public dataset. The [dataset](https://console.cloud.google.com/bigquery?project=bigquery-public-data&page=table&t=stories&d=hacker_news&p=bigquery-public-data&redirect_from_classic=true) contains all headlines from the sites inception in October 2006 until October 2015. \n",
+    "Hacker news headlines are available as a BigQuery public dataset. The [dataset](https://console.cloud.google.com/bigquery?project=bigquery-public-data&page=table&t=full&d=hacker_news&p=bigquery-public-data&redirect_from_classic=true) contains all headlines from the sites inception in October 2006 until October 2015. \n",
     "\n",
     "Here is a sample of the dataset:"
    ]
@@ -90,7 +90,7 @@
     "SELECT\n",
     "    url, title, score\n",
     "FROM\n",
-    "    `bigquery-public-data.hacker_news.stories`\n",
+    "    `bigquery-public-data.hacker_news.full`\n",
     "WHERE\n",
     "    LENGTH(title) > 10\n",
     "    AND score > 10\n",
@@ -114,10 +114,10 @@
     "%%bigquery --project $PROJECT\n",
     "\n",
     "SELECT\n",
-    "    ARRAY_REVERSE(SPLIT(REGEXP_EXTRACT(url, '.*://(.[^/]+)/'), '.'))[OFFSET(1)] AS source,\n",
+    "    ARRAY_REVERSE(SPLIT(REGEXP_EXTRACT(url, '.*://(.[^/]+)/'), '.'))[safe_offset (1)] AS source,\n",
     "    COUNT(title) AS num_articles\n",
     "FROM\n",
-    "    `bigquery-public-data.hacker_news.stories`\n",
+    "    `bigquery-public-data.hacker_news.full`\n",
     "WHERE\n",
     "    REGEXP_CONTAINS(REGEXP_EXTRACT(url, '.*://(.[^/]+)/'), '.com$')\n",
     "    AND LENGTH(title) > 10\n",
@@ -146,10 +146,10 @@
     "sub_query = \"\"\"\n",
     "SELECT\n",
     "    title,\n",
-    "    ARRAY_REVERSE(SPLIT(REGEXP_EXTRACT(url, '{0}'), '.'))[OFFSET(1)] AS source\n",
+    "    ARRAY_REVERSE(SPLIT(REGEXP_EXTRACT(url, '{0}'), '.'))[safe_offset (1)] AS source\n",
     "    \n",
     "FROM\n",
-    "    `bigquery-public-data.hacker_news.stories`\n",
+    "    `bigquery-public-data.hacker_news.full`\n",
     "WHERE\n",
     "    REGEXP_CONTAINS(REGEXP_EXTRACT(url, '{0}'), '.com$')\n",
     "    AND LENGTH(title) > 10\n",