Skip to content

Commit

Permalink
Blueprint updates 0124 (#339)
Browse files Browse the repository at this point in the history
* update console blueprints

- add synthetic card details page
- new gretel tuner card
- update gpt card and add details page

* added icon back

* blueprint updates 01162024

- Added images to the new details pages
- Minor copy updates in gretel.json

* Fixed mistakes

Added image to relational details instead of synthetic details. Fixed.

* Updated relational-db.md

Added copy and cleaned up an incomplete sentence.
  • Loading branch information
yamini authored Jan 17, 2024
1 parent 3cab1f1 commit d1c5b1c
Show file tree
Hide file tree
Showing 8 changed files with 19 additions and 8 deletions.
4 changes: 3 additions & 1 deletion use_cases/details/gpt-natural-language.md
Original file line number Diff line number Diff line change
@@ -1,4 +1,6 @@
You can streamline your natural language processing projects with our GPT-driven blueprint. It supports the creation of labeled examples for various language tasks. From sentiment analysis to text summarization, Gretel GPT makes it simple to prompt or fine-tune LLMs to enhance the quality of your datasets with precision and efficiency.
![Generate synthetic text data](https://blueprints.gretel.cloud/use_cases/images/natural-language-gpt.png "Generate synthetic text data")

Streamline your natural language processing projects with our GPT-driven blueprint. It supports the creation of labeled examples for various language tasks. From sentiment analysis to text summarization, Gretel GPT makes it simple to prompt or fine-tune LLMs to enhance the quality of your datasets with precision and efficiency.

For Gretel GPT, training and prompt data needs to be a single column CSV file containing natural language text. If working with a multi-column dataset, specify the text input column in the config.

Expand Down
2 changes: 2 additions & 0 deletions use_cases/details/gretel-tuner.md
Original file line number Diff line number Diff line change
@@ -1,3 +1,5 @@
![Gretel Tuner - Autotune synthetic data params](https://blueprints.gretel.cloud/use_cases/images/auto-tune-params.png "Gretel Tuner - Autotune synthetic data params")

Gretel Tuner streamlines the hyperparameter tuning process, making it easier and more efficient to generate high-quality synthetic data. Integrated into our [Python SDK](https://github.com/gretelai/gretel-python-client), Gretel Tuner offers a straightforward and efficient way to optimize your data models. It’s an essential tool for both experienced data scientists and newcomers to the field, simplifying complex tasks and improving outcomes.

### Notebook Details
Expand Down
7 changes: 6 additions & 1 deletion use_cases/details/relational-db.md
Original file line number Diff line number Diff line change
@@ -1,4 +1,9 @@
Gretel Relational takes you beyond synthesizing and transforming one table at a time, and enables generating entire databases, all while maintaining referential integrity. This early preview will give you a taste of what's coming in the Gretel Console and CLI/SDK. [We'd love to hear your feedback!](https://dqq4jigtkl1.typeform.com/to/Gibb8awJ)
Gretel Relational takes you beyond synthesizing and transforming one table at a time, and enables generating entire databases, all while maintaining referential integrity. We support connections to several relational databases such as MS SQL Server, Snowflake, BigQuery, Oracle, MySQL, PostgreSQL, with more connectors coming soon.

To get started in the Console, go to "Workflows" in the sidebar, and select the "New Workflow" button. Choose your project and the model you want to use (ACGTAN for synthesizing or Transform v2 for redacting PII). Next, use the Connetion Creation
Wizard to create a connection to your remote database. And then follow the easy steps to finish creating your workflow.

Want to see how Gretel Relational works? Check out one of our notebooks below to go through the process step by step and compare the data before and after synthesis.

To connect to your remote database, just

Expand Down
2 changes: 2 additions & 0 deletions use_cases/details/synthetic.md
Original file line number Diff line number Diff line change
@@ -1,3 +1,5 @@
![Generate synthetic tabular data](https://blueprints.gretel.cloud/use_cases/images/synthetic-tabular-generation.png "Generate synthetic tabular data")

If you’re new to Gretel, our synthetic data blueprint is a great place to start. This gentle introduction to synthetic data generation automatically selects our popular [ACTGAN model](https://gretel.ai/blog/scale-synthetic-data-to-millions-of-rows-with-actgan) and provides a sample healthcare dataset. Just answer a few questions, review the model configuration and hit **Run**.

Prefer coding? Check out the [Gretel 101 notebook](https://colab.research.google.com/github/gretelai/gretel-blueprints/blob/main/sdk_blueprints/Gretel_101_Blueprint.ipynb) example. Synthesize data in just 4 lines of code!
Expand Down
12 changes: 6 additions & 6 deletions use_cases/gretel.json
Original file line number Diff line number Diff line change
Expand Up @@ -2,8 +2,8 @@
"cards": [
{
"gtmId": "use-case-synthetic",
"title": "Generate synthetic data",
"description": "Create highly accurate synthetic training data for ML models.",
"title": "Generate synthetic tabular data",
"description": "Create highly accurate synthetic tabular training data for ML models.",
"cardType": "Console",
"imageName": "synthetics.png",
"icon": "synthetics.png",
Expand All @@ -30,8 +30,8 @@
},
{
"gtmId": "use-case-relational-db",
"title": "Connect to any normalized SQL database ",
"description": "Synthesize and transform data in a relational database with Gretel Relational.",
"title": "Synthesize any normalized SQL database ",
"description": "Use Gretel Relational to synthesize or transform data in a relational database without losing foreign key relationships.",
"cardType": "Notebook",
"imageName": "relational-db.png",
"icon": "relational-db.png",
Expand Down Expand Up @@ -68,7 +68,7 @@
{
"gtmId": "use-case-tabular-dp",
"title": "Create provably private versions of sensitive data",
"description": "Use Gretel Tabular DP, our new graph-based generative model, to generate synthetic data with strong differential privacy guarantees.",
"description": "Use Gretel Tabular DP, our graph-based generative model, to generate synthetic data with strong differential privacy guarantees.",
"cardType": "Console",
"imageName": "tabular-dp.png",
"icon": "tabular-dp.png",
Expand All @@ -86,7 +86,7 @@
},
{
"gtmId": "use-case-natural-lang-gpt",
"title": "Generate natural language text using GPT",
"title": "Generate synthetic text data using GPT",
"description": "Quickly generate diverse text examples for natural language tasks, simplifying the generation of labeled datasets.",
"cardType": "Console",
"imageName": "natural-lang-gpt.png",
Expand Down
Binary file added use_cases/images/auto-tune-params.png
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added use_cases/images/natural-language-gpt.png
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.

0 comments on commit d1c5b1c

Please sign in to comment.