Skip to content

docs: remove session and connection in llm notebook #821

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Jul 4, 2024
Merged

docs: remove session and connection in llm notebook #821

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
106 changes: 41 additions & 65 deletions notebooks/generative_ai/large_language_models.ipynb
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -16,8 +16,7 @@
"cell_type": "markdown",
"metadata": {},
"source": [
"## Prerequisites\n",
"Create session and define a BQ connection which we already created and allowlisted. "
"## Define the model"
]
},
{
Expand All @@ -29,33 +28,14 @@
"name": "stderr",
"output_type": "stream",
"text": [
"/usr/local/google/home/garrettwu/src/bigframes/bigframes/session/__init__.py:1762: UserWarning: No explicit location is set, so using location US for the session.\n",
" return Session(context)\n"
"/usr/local/google/home/garrettwu/src/bigframes/bigframes/ml/llm.py:589: DefaultLocationWarning: No explicit location is set, so using location US for the session.\n",
" self.session = session or bpd.get_global_session()\n"
]
}
],
"source": [
"session = bigframes.pandas.get_global_session()\n",
"connection = f\"{session.bqclient.project}.us.bigframes-default-connection\""
]
},
{
"attachments": {},
"cell_type": "markdown",
"metadata": {},
"source": [
"## Define the model"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [
},
{
"data": {
"text/html": [
"Query job 12bcd690-ca99-4001-bf26-032f50e77d62 is DONE. 0 Bytes processed. <a target=\"_blank\" href=\"https://console.cloud.google.com/bigquery?project=bigframes-dev&j=bq:US:12bcd690-ca99-4001-bf26-032f50e77d62&page=queryresults\">Open Job</a>"
"Query job 675a6c8a-213b-496c-9f77-b87bf7cfa5e0 is DONE. 0 Bytes processed. <a target=\"_blank\" href=\"https://console.cloud.google.com/bigquery?project=bigframes-dev&j=bq:US:675a6c8a-213b-496c-9f77-b87bf7cfa5e0&page=queryresults\">Open Job</a>"
],
"text/plain": [
"<IPython.core.display.HTML object>"
Expand All @@ -66,7 +46,7 @@
}
],
"source": [
"model = GeminiTextGenerator(session=session, connection_name=connection)"
"model = GeminiTextGenerator()"
]
},
{
Expand All @@ -81,7 +61,7 @@
},
{
"cell_type": "code",
"execution_count": 4,
"execution_count": 3,
"metadata": {},
"outputs": [],
"source": [
Expand All @@ -102,13 +82,13 @@
},
{
"cell_type": "code",
"execution_count": 5,
"execution_count": 4,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"Query job f8fe31c6-7d8a-4919-9492-8304a0083cca is DONE. 0 Bytes processed. <a target=\"_blank\" href=\"https://console.cloud.google.com/bigquery?project=bigframes-dev&j=bq:US:f8fe31c6-7d8a-4919-9492-8304a0083cca&page=queryresults\">Open Job</a>"
"Query job 7967df2b-9f0f-45c8-a363-15f65891c3bf is DONE. 0 Bytes processed. <a target=\"_blank\" href=\"https://console.cloud.google.com/bigquery?project=bigframes-dev&j=bq:US:7967df2b-9f0f-45c8-a363-15f65891c3bf&page=queryresults\">Open Job</a>"
],
"text/plain": [
"<IPython.core.display.HTML object>"
Expand All @@ -118,21 +98,17 @@
"output_type": "display_data"
},
{
"data": {
"text/html": [
"Query job 28bab71f-e218-4d92-9a50-dab41bb0c71f is DONE. 24 Bytes processed. <a target=\"_blank\" href=\"https://console.cloud.google.com/bigquery?project=bigframes-dev&j=bq:US:28bab71f-e218-4d92-9a50-dab41bb0c71f&page=queryresults\">Open Job</a>"
],
"text/plain": [
"<IPython.core.display.HTML object>"
]
},
"metadata": {},
"output_type": "display_data"
"name": "stderr",
"output_type": "stream",
"text": [
"/usr/local/google/home/garrettwu/src/bigframes/bigframes/core/__init__.py:108: PreviewWarning: Interpreting JSON column(s) as StringDtype. This behavior may change in future versions.\n",
" warnings.warn(\n"
]
},
{
"data": {
"text/html": [
"Query job 01d66b61-459f-474e-9f66-d519f9c2f23d is DONE. 6 Bytes processed. <a target=\"_blank\" href=\"https://console.cloud.google.com/bigquery?project=bigframes-dev&j=bq:US:01d66b61-459f-474e-9f66-d519f9c2f23d&page=queryresults\">Open Job</a>"
"Query job 9a1f57cd-98e1-4eac-a1b3-8f88d61971cd is DONE. 6 Bytes processed. <a target=\"_blank\" href=\"https://console.cloud.google.com/bigquery?project=bigframes-dev&j=bq:US:9a1f57cd-98e1-4eac-a1b3-8f88d61971cd&page=queryresults\">Open Job</a>"
],
"text/plain": [
"<IPython.core.display.HTML object>"
Expand All @@ -144,7 +120,7 @@
{
"data": {
"text/html": [
"Query job af606ca7-4bcf-4bd1-95fd-c516542b5a4f is DONE. 5.3 kB processed. <a target=\"_blank\" href=\"https://console.cloud.google.com/bigquery?project=bigframes-dev&j=bq:US:af606ca7-4bcf-4bd1-95fd-c516542b5a4f&page=queryresults\">Open Job</a>"
"Query job 2a94a2cf-7d4c-4009-a798-d7a5d6d4049d is DONE. 8.5 kB processed. <a target=\"_blank\" href=\"https://console.cloud.google.com/bigquery?project=bigframes-dev&j=bq:US:2a94a2cf-7d4c-4009-a798-d7a5d6d4049d&page=queryresults\">Open Job</a>"
],
"text/plain": [
"<IPython.core.display.HTML object>"
Expand Down Expand Up @@ -183,28 +159,28 @@
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>**BigQuery**\n",
" <td>## BigQuery: A Serverless Data Warehouse\n",
"\n",
"**Definition:**\n",
"\n",
"BigQuery is a s...</td>\n",
" <td>null</td>\n",
"BigQ...</td>\n",
" <td>[{\"category\":1,\"probability\":1,\"probability_sc...</td>\n",
" <td></td>\n",
" <td>What is BigQuery?</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>**BigQuery Machine Learning (BQML)**\n",
" <td>## BigQuery Machine Learning (BQML)\n",
"\n",
"BQML is ...</td>\n",
" <td>null</td>\n",
"BQML is a...</td>\n",
" <td>[{\"category\":1,\"probability\":1,\"probability_sc...</td>\n",
" <td></td>\n",
" <td>What is BQML?</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>BigQuery DataFrame is a Python DataFrame imple...</td>\n",
" <td>null</td>\n",
" <td>## What is BigQuery DataFrame?\n",
"\n",
"**BigQuery Dat...</td>\n",
" <td>[{\"category\":1,\"probability\":1,\"probability_sc...</td>\n",
" <td></td>\n",
" <td>What is BigQuery DataFrame?</td>\n",
" </tr>\n",
Expand All @@ -214,28 +190,28 @@
],
"text/plain": [
" ml_generate_text_llm_result \\\n",
"0 **BigQuery**\n",
"0 ## BigQuery: A Serverless Data Warehouse\n",
"\n",
"**Definition:**\n",
"BigQ... \n",
"1 ## BigQuery Machine Learning (BQML)\n",
"\n",
"BigQuery is a s... \n",
"1 **BigQuery Machine Learning (BQML)**\n",
"BQML is a... \n",
"2 ## What is BigQuery DataFrame?\n",
"\n",
"BQML is ... \n",
"2 BigQuery DataFrame is a Python DataFrame imple... \n",
"**BigQuery Dat... \n",
"\n",
" ml_generate_text_rai_result ml_generate_text_status \\\n",
"0 null \n",
"1 null \n",
"2 null \n",
" ml_generate_text_rai_result ml_generate_text_status \\\n",
"0 [{\"category\":1,\"probability\":1,\"probability_sc... \n",
"1 [{\"category\":1,\"probability\":1,\"probability_sc... \n",
"2 [{\"category\":1,\"probability\":1,\"probability_sc... \n",
"\n",
" prompt \n",
"0 What is BigQuery? \n",
"1 What is BQML? \n",
"2 What is BigQuery DataFrame? "
]
},
"execution_count": 5,
"execution_count": 4,
"metadata": {},
"output_type": "execute_result"
}
Expand All @@ -255,16 +231,16 @@
},
{
"cell_type": "code",
"execution_count": 6,
"execution_count": 5,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"'**BigQuery**\\n\\n**Definition:**\\n\\nBigQuery is a serverless, highly scalable, cloud-based data warehouse and analytics platform offered by Google Cloud.\\n\\n**Key Features:**\\n\\n* **Massive Scalability:** Can handle large datasets (petabytes or more) with fast query execution.\\n* **Elastic:** Automatically scales compute resources based on workload requirements.\\n* **Serverless:** Users do not need to manage infrastructure or provision resources.\\n* **Flexible Data Loading:** Supports a wide range of data sources, including files, databases, and streaming data.\\n* **SQL-Based Querying:** Uses standard SQL syntax for querying and analyzing data.\\n* **Machine Learning Integration:** Provides built-in machine learning capabilities for predictive analytics and data exploration.\\n* **Real-Time Analysis:** Supports streaming data analysis and interactive dashboards.\\n* **Collaboration and Sharing:** Allows multiple users to access and analyze data in a collaborative environment.\\n* **Cost-Effective:** Pay-as-you-go pricing based on data scanned and compute resources used.\\n\\n**Applications:**\\n\\n* Data warehousing and analytics\\n* Business intelligence and reporting\\n* Data science and machine learning\\n* Data exploration and visualization\\n* Marketing analytics\\n* Fraud detection and risk management\\n\\n**Benefits:**\\n\\n* Rapid data analysis on large datasets\\n* Reduced infrastructure management overhead\\n* Increased agility and flexibility\\n* Enhanced collaboration and data sharing\\n* Cost-effective data storage and analytics'"
"\"## BigQuery: A Serverless Data Warehouse\\n\\nBigQuery is a serverless, cloud-based data warehouse that enables scalable analysis of large datasets. It's a popular choice for businesses of all sizes due to its ability to handle petabytes of data and run complex queries quickly and efficiently. Let's delve into its key features:\\n\\n**Serverless Architecture:** BigQuery eliminates the need for server management, allowing you to focus on analyzing data. Google manages the infrastructure, scaling resources up or down automatically based on your needs.\\n\\n**Scalability:** BigQuery can handle massive datasets, scaling seamlessly as your data volume grows. It automatically distributes queries across its infrastructure, ensuring fast and efficient processing.\\n\\n**SQL-like Querying:** BigQuery uses a familiar SQL-like syntax, making it easy for data analysts and developers to learn and use. This allows them to leverage their existing SQL knowledge for data exploration and analysis.\\n\\n**Cost-Effectiveness:** BigQuery offers a pay-as-you-go pricing model, meaning you only pay for the resources you use. This makes it a cost-effective solution for businesses with varying data processing needs.\\n\\n**Integration with Google Cloud:** BigQuery integrates seamlessly with other Google Cloud services like Cloud Storage, Dataflow, and Machine Learning, enabling a comprehensive data processing and analysis workflow within the Google Cloud ecosystem.\\n\\n**Security and Reliability:** BigQuery offers robust security features and high availability, ensuring data protection and reliable access.\\n\\n**Use Cases:** BigQuery finds applications in various scenarios, including:\\n\\n* **Data Warehousing:** Store and analyze large amounts of structured and semi-structured data.\\n* **Business Intelligence:** Generate insights from data for informed decision-making.\\n* **Data Analytics:** Perform complex data analysis and extract valuable patterns.\\n* **Machine Learning:** Train and deploy machine learning models on large datasets.\\n\\n**Getting Started:** To get started with BigQuery, you can create a free trial account on Google Cloud Platform and explore its features. Numerous tutorials and documentation are available to help you learn and use BigQuery effectively.\\n\\n## Additional Resources:\\n\\n* **BigQuery Documentation:** https://cloud.google.com/bigquery/docs/\\n* **BigQuery Quickstart:** https://cloud.google.com/bigquery/docs/quickstarts/quickstart-console\\n* **BigQuery Pricing:** https://cloud.google.com/bigquery/pricing\\n\\nFeel free to ask if you have any further questions about BigQuery!\""
]
},
"execution_count": 6,
"execution_count": 5,
"metadata": {},
"output_type": "execute_result"
}
Expand Down