-You can find other examples in the `examples/` folder.
+You can find other examples in the [examples/](./examples/) folder.
## Pre-requisites
To run the library, you need:
- - Python 3.10 or higher.
+ - Python 3.10 or higher. We'll assume you are using [Anaconda](https://docs.anaconda.com/anaconda/install/), but you can use other Python distributions.
- Access to Azure OpenAI Service or Open AI GPT-4 APIs. You can get access to the Azure OpenAI Service [here](https://azure.microsoft.com/en-us/products/ai-services/openai-service), and to the OpenAI API [here](https://platform.openai.com/).
* For Azure OpenAI Service, you will need to set the `AZURE_OPENAI_KEY` and `AZURE_OPENAI_ENDPOINT` environment variables to your API key and endpoint, respectively.
* For OpenAI, you will need to set the `OPENAI_API_KEY` environment variable to your API key.
+ - By default, TinyTroupe `config.ini` is set to use some specific API, model and related parameters. You can customize these values by including your own `config.ini` file in the same folder as the program or notebook you are running. An example of a `config.ini` file is provided in the [examples/](./examples/) folder.
>[!IMPORTANT]
> **Content Filters**: To ensure no harmful content is generated during simulations, it is strongly recommended to use content filters whenever available at the API level. In particular, **if using Azure OpenAI, there's extensive support for content moderation, and we urge you to use it.** For details about how to do so, please consult [the corresponding Azure OpenAI documentation](https://learn.microsoft.com/en-us/azure/ai-services/openai/concepts/content-filter). If content filters are in place, and an API call is rejected by them, the library will raise an exception, as it will be unable to proceed with the simulation at that point.
@@ -131,34 +105,53 @@ To run the library, you need:
## Installation
-Currently, the official recommended way to install the library is directly from this repository, not PyPI.
-
-### From the GitHub repository
-To install the library directly from the GitHub repository::
-
+**Currently, the officially recommended way to install the library is directly from this repository, not PyPI.** You can follow these steps:
+
+1. If Conda is not installed, you can get it from [here](https://docs.anaconda.com/anaconda/install/). You can also use other Python distributions, but we'll assume Conda here for simplicity.
+2. Create a new Python environment:
+ ```bash
+ conda create -n tinytroupe python=3.10
+ ```
+3. Activate the environment:
+ ```bash
+ conda activate tinytroupe
+ ```
+4. Make sure you have eihter Azure OpenAI or OpenAI API keys set as environment variables, as described in the [Pre-requisites](#pre-requisites) section.
+5. Install the library **from this repository, not PyPI**:
+ ```bash
+ pip install git+https://github.com/microsoft/tinytroupe.git
+ ```
+6. You can now use TinyTroupe to create your simulations 🥳. If you want to run the examples in the
+ [examples/](./examples/) folder or modify TinyTroupe itself, however, you should clone the repository as described below.
+
+
+### Cloning the repository: examples and local development
+
+If you want to run the examples in the [examples/](./examples/) folder, you can simply clone the repository and run the examples directly from there:
```bash
-$ pip install git+https://github.com/microsoft/tinytroupe.git
+git clone https://github.com/microsoft/tinytroupe
+cd tinytroupe
```
-### From the local repository
-If you want to make changes to the library and test them locally, you can also of course clone the repository first:
-
+Further, if you want to modify TinyTroupe itself, you can install it from the local clone:
```bash
-$ git clone https://github.com/microsoft/tinytroupe
-$ cd tinytroupe
-$ pip install .
+pip install .
```
+or, in editable mode (i.e., changes to the code will be reflected immediately):
+```bash
+pip install -e .
+```
## Principles
-Recently, we have seen LLMs used to simulate people (such as [this](https://github.com/joonspk-research/generative_agents)), but largely in a “game-like” setting for contemplative or entertainment purposes. What if we try instead to simulate people to support productivity tasks? TinyTroupe is our attempt. To do so, it follows these principles:
+Recently, we have seen LLMs used to simulate people (such as [this](https://github.com/joonspk-research/generative_agents)), but largely in a “game-like” setting for contemplative or entertainment purposes. There are also libraries for building multiagent systems for proble-solving and assitive AI, like [Autogen](https://microsoft.github.io/) and [Crew AI](https://docs.crewai.com/). What if we combine these ideas and simulate people to support productivity tasks? TinyTroupe is our attempt. To do so, it follows these principles:
1. **Programmatic**: agents and environments are defined programmatically (in Python and JSON), allowing very flexible uses. They can also thus underpin other software apps!
- 2. **Analytical**: meant to improve our understanding of people, users and society. Unlike entertainment applications, this is one aspect that is critical for business and productivity use cases.
+ 2. **Analytical**: meant to improve our understanding of people, users and society. Unlike entertainment applications, this is one aspect that is critical for business and productivity use cases. This is also why we recommend using Jupyter notebooks for simulations, just like one uses them for data analysis.
3. **Persona-based**: agents are meant to be archetypical representation of people; for greater realism and control, detailed specification of such personas is encouraged: age, occupation, skills, tastes, opinions, etc.
4. **Multiagent**: allows multiagent interaction under well-defined environmental constraints.
5. **Utilities-heavy**: provides many mechanisms to facilitate specifications, simulations, extractions, reports, validations, etc. This is one area in which dealing with *simulations* differs significantly from *assistance* tools.
- 6. **Experiment-oriented**: simulations are defined, run, analyzed and refined by an *experimenter* iteratively; suitable experimentation tools are thus provided.
+ 6. **Experiment-oriented**: simulations are defined, run, analyzed and refined by an *experimenter* iteratively; suitable experimentation tools are thus provided. *See one of our [previous paper](https://www.microsoft.com/en-us/research/publication/the-case-for-experiment-oriented-computing/) for more on this.*
Together, these are meant to make TinyTroupe a powerful and flexible **imagination enhancement tool** for business and productivity scenarios.
@@ -182,7 +175,6 @@ One common source of confusion is to think all such AI agents are meant for assi
The project is structured as follows:
- `/tinytroupe`: contains the Python library itself. In particular:
* `/tinytroupe/prompts` contains the prompts used to call the LLMs.
- * `/tinytroupe/microsoft` contains elements specific to the _public_ Microsoft ecosystem.
- `/tests`: contains the unit tests for the library. You can use the `test.bat` script to run these.
- `/examples`: contains examples that show how to use the library, mainly using Jupyter notebooks (for greater readability), but also as pure Python scripts.
- `/data`: any data used by the examples or the library.
@@ -243,7 +235,7 @@ lisa.define_several("personality_traits",
`TinyTroupe` also provides a clever way to obtain new agents, using LLMs to generate their specification for you, through the `TinyPersonFactory` class.
```python
-from tinytroupe.personfactory import TinyPersonFactory
+from tinytroupe.factory import TinyPersonFactory
factory = TinyPersonFactory("Create a Brazilian person that is a doctor, like pets and the nature and love heavy metal.")
person = factory.generate_person()
@@ -307,10 +299,11 @@ TinyTroupe provides a number of utilities and conveniences to help you create si
- `TinyPersonFactory`: helps you generate new `TinyPerson`s using LLMs.
- `TinyTool`: simulated tools that can be used by `TinyPerson`s.
- `TinyStory`: helps you create and manage the story told through simulations.
- - `InteractionResultsExtractor` and `InteractionResultsReducer`: extract and reduce the results of interactions between agents.
- - `TinyPersonChecker`: helps you validate the behavior of your `TinyPerson`s.
+ - `TinyPersonValidator`: helps you validate the behavior of your `TinyPerson`s.
+ - `ResultsExtractor` and `ResultsReducer`: extract and reduce the results of interactions between agents.
- ... and more ...
+In general, elements that represent simulated entities or complementary mechanisms are prefixed with `Tiny`, while those that are more infrastructural are not. This is to emphasize the simulated nature of the elements that are part of the simulation itself.
### Caching
Calling LLM APIs can be expensive, thus caching strategies are important to help reduce that cost.
@@ -340,7 +333,7 @@ when a new call comes and is identical to a previous one, the cached value is re
### Config.ini
-The `config.ini` file contains various parameters that can be used to customize the behavior of the library, such as model parameters and logging level. Please pay special attention to `API_TYPE` parameter, which defines whether you are using the Azure OpenAI Service or the OpenAI API.
+The `config.ini` file contains various parameters that can be used to customize the behavior of the library, such as model parameters and logging level. Please pay special attention to `API_TYPE` parameter, which defines whether you are using the Azure OpenAI Service or the OpenAI API. We provide an example of a `config.ini` file, [./examples/config.ini](./examples/config.ini), which you can use as a template for your own, or just modify to run the examples.
## Contributing
@@ -357,9 +350,9 @@ For more information see the [Code of Conduct FAQ](https://opensource.microsoft.
contact [opencode@microsoft.com](mailto:opencode@microsoft.com) with any additional questions or comments.
### What and How to Contribute
-We need all sorts of things, like:
- - New interesting use cases demonstrations, or even just domain-specific application ideas.
- If you are a domain expert in some area that could benefit from TinyTroupe, we'd love to hear from you.
+We need all sorts of things, but we are looking mainly for new interesting use cases demonstrations, or even just domain-specific application ideas. If you are a domain expert in some area that could benefit from TinyTroupe, we'd love to hear from you.
+
+Beyond that, many other aspects can be improved, such as:
- Memory mechanisms.
- Data grounding mechanisms.
- Reasoning mechanisms.
@@ -370,7 +363,7 @@ We need all sorts of things, like:
Please note that anything that you contribute might be released as open-source (under MIT license).
If you would like to make a contribution, please try to follow these general guidelines:
- - **Tiny-everything**: If you are implementing a user-facing element (e.g., an agent or environment type), and it sounds good, call your new _X_ as _TinyX_ :-)
+ - **Tiny naming convention**: If you are implementing a experimenter-facing simulated element (e.g., an agent or environment type) or closely related (e.g., agent factories, or content enrichers), and it sounds good, call your new *XYZ* as *TinyXYZ* :-) On the other hand, auxiliary and infrastructural mechanisms should not start with the "Tiny" prefix. The idea is to emphasize the simulated nature of the elements that are part of the simulation itself.
- **Tests:** If you are writing some new mechanism, please also create at least a unit test `tests/unit/`, and if you can a functional scenario test (`tests/scenarios/`).
- **Demonstrations:** If you'd like to demonstrate a new scenario, please design it preferably as a new Jupyter notebook within `examples/`.
- **Microsoft:** If you are implementing anything that is Microsoft-specific and non-confidential, please put it under a `.../microsoft/` folder.
@@ -399,8 +392,7 @@ Other special contributions were made by:
## Citing TinyTroupe
-We are working in an introductory paper that will be the official academic citation for TinyTroupe. In the meantime, please just cite this repository including the core team members as authors. For instance:
-
+We are working on an introductory paper that will be the official academic citation for TinyTroupe. In the meantime, please just cite this repository including the core team members as authors. For instance:
>Paulo Salem, Christopher Olsen, Paulo Freire, Yi Ding, Prerit Saxena (2024). **TinyTroupe: LLM-powered multiagent persona simulation for imagination enhancement and business insights.** [Computer software]. GitHub repository. https://github.com/microsoft/tinytroupe
diff --git a/data/extractions/appartment_rent_ad.extraction.json b/data/extractions/appartment_rent_ad.extraction.json
index d09db7d..f45b53c 100644
--- a/data/extractions/appartment_rent_ad.extraction.json
+++ b/data/extractions/appartment_rent_ad.extraction.json
@@ -2,7 +2,7 @@
"agent_extractions": {},
"world_extraction": {
"Focus group": {
- "ad_copy": "Discover the perfect blend of modern living and convenience in this beautifully renovated apartment. Located in a quiet spot on a main avenue, this spacious one-bedroom home features an elegant integrated kitchen and living room, complete with stylish 60s-style chairs and high-end appliances. Enjoy the luxury of two tastefully designed bathrooms and a versatile home office space. With wood-like floors and a new washer-dryer, this fully furnished apartment is ready for you to move in. Just steps away from public transport, shops, and restaurants, and near the country's top Medicine School, it's an ideal choice for students and professionals alike. Experience the tranquility and charm of this unique living space with high-quality photos and a virtual tour. Don't miss out on this opportunity to make it your new home!"
+ "ad_copy": "Spacious, Modern Living in a Quiet, Convenient Location! Discover this beautifully renovated apartment with an open design, perfect for students and young professionals. Enjoy the convenience of nearby amenities like supermarkets, gyms, and public transport, while relishing the tranquility of a quiet setting. Ideal for remote work with a home office space, and fully furnished with modern appliances. Don't miss out on this opportunity\u2014contact us today for a virtual tour and special incentives for early renters!"
}
}
}
\ No newline at end of file
diff --git a/data/extractions/synthetic_data_generation.out.csv b/data/extractions/synthetic_data_generation.out.csv
index ed31853..604a6cc 100644
--- a/data/extractions/synthetic_data_generation.out.csv
+++ b/data/extractions/synthetic_data_generation.out.csv
@@ -1,5 +1,6 @@
author,content
USER,Message each other to get work done.
-Liam,"Hi Samantha, let's coordinate on our tasks to ensure we complete them efficiently. How would you like to proceed?"
-Samantha,"Hi Liam, let's start by listing our tasks and prioritizing them. We can then assign responsibilities and set deadlines to ensure everything is completed efficiently. How does that sound?"
-Liam,"That sounds like a great plan, Samantha. Let's start by listing our tasks. Once we have them all, we can prioritize and assign responsibilities."
+Lila,Let's make sure we are all on the same page and communicate effectively to get our tasks done.
+Ethan,"Absolutely, Lila. Clear communication is key to ensuring we all stay aligned and productive."
+Lila,"Thanks, Ethan. Let's set up regular check-ins to keep everyone aligned and address any communication gaps."
+Ethan,"Great idea, Lila. Regular check-ins will definitely help us stay on track and address any issues promptly. Let's schedule them."
diff --git a/data/extractions/wordprocessor/Document/Resume.Lisa.docx b/data/extractions/wordprocessor/Document/Resume.Lisa.docx
new file mode 100644
index 0000000..4b17c6a
Binary files /dev/null and b/data/extractions/wordprocessor/Document/Resume.Lisa.docx differ
diff --git a/data/extractions/wordprocessor/Document/Resume.Lisa.json b/data/extractions/wordprocessor/Document/Resume.Lisa.json
new file mode 100644
index 0000000..052ac5a
--- /dev/null
+++ b/data/extractions/wordprocessor/Document/Resume.Lisa.json
@@ -0,0 +1,5 @@
+{
+ "title": "Resume",
+ "content": "# Lisa's Resume\n\n## Objective\nI am a highly motivated and detail-oriented data science professional seeking a challenging position in the field of data science. My goal is to leverage my extensive skills in machine learning, data analysis, and statistical modeling to contribute to innovative and impactful projects. I am particularly interested in roles that allow me to work on cutting-edge technologies and collaborate with cross-functional teams to drive data-driven decision-making and enhance business outcomes.\n\n## Skills\n\n### Technical Skills\n- **Programming Languages:** Proficient in Python, with extensive experience in using libraries such as pandas for data manipulation, scikit-learn for machine learning, and TensorFlow for deep learning applications. Additionally, I have hands-on experience with Azure ML for deploying and managing machine learning models in the cloud.\n- **Data Analysis Tools:** Familiar with SQL for database querying and data extraction, as well as Power BI for creating interactive data visualizations and dashboards to communicate insights effectively to stakeholders.\n- **Machine Learning and AI:** Strong foundation in machine learning algorithms, including supervised and unsupervised learning, natural language processing, and computer vision. Experienced in building, training, and evaluating models to solve complex business problems.\n- **Analytical and Problem-Solving Skills:** Demonstrated ability to analyze large datasets, identify patterns and trends, and develop actionable insights. Skilled in using statistical techniques to validate hypotheses and support data-driven decision-making.\n\n### Soft Skills\n- **Communication:** Excellent verbal and written communication skills, with the ability to present complex technical concepts to non-technical audiences. Experienced in writing technical reports and documentation.\n- **Collaboration:** Proven track record of working effectively in team environments, collaborating with colleagues from diverse backgrounds to achieve common goals. Strong interpersonal skills and the ability to build positive relationships with stakeholders.\n- **Adaptability:** Quick learner with the ability to adapt to new technologies and methodologies. Open to feedback and committed to continuous professional development.\n\n## Experience\n\n### Data Scientist at Microsoft (June 2018 - Present)\n- **User Behavior Analysis:** Conducted in-depth analysis of user behavior and feedback data to identify areas for improvement in search results. Developed data-driven strategies to enhance user experience and increase engagement.\n- **Machine Learning Model Development:** Built and tested machine learning models for various search scenarios, including personalized recommendations and query understanding. Collaborated with engineering teams to integrate models into production systems.\n- **Privacy and Security Compliance:** Ensured that all data and models adhered to privacy and security policies, working closely with legal and compliance teams to address any potential risks. Implemented best practices for data governance and model transparency.\n\n### Data Analyst Intern at TechCorp (June 2017 - May 2018)\n- Assisted in the development of data pipelines for processing and analyzing large datasets. Conducted exploratory data analysis to uncover insights and support business decision-making.\n- Created interactive dashboards and reports using Power BI to visualize key performance indicators and track progress against business objectives.\n\n## Education\n\n### Bachelor's Degree in Computer Science\n- **University of California, Berkeley (2014 - 2018)**\n - Relevant Coursework: Data Structures and Algorithms, Machine Learning, Database Systems, Artificial Intelligence, Statistical Methods for Data Science\n - Honors: Dean's List (2016, 2017), Member of the Computer Science Honor Society\n\n## Certifications\n- **Certified Data Scientist (CDS)**\n- **Microsoft Certified: Azure Data Scientist Associate**\n\n## Projects\n\n### Predictive Analytics for E-commerce\n- Developed a predictive analytics model to forecast sales trends and optimize inventory management for an e-commerce platform. Utilized time series analysis and regression techniques to improve demand forecasting accuracy.\n\n### Sentiment Analysis of Social Media Data\n- Implemented a sentiment analysis model to analyze customer feedback on social media platforms. Used natural language processing techniques to classify sentiments and provide actionable insights for marketing strategies.\n\n## Interests\n\n- **Artificial Intelligence and Machine Learning:** Passionate about exploring the latest advancements in AI and machine learning, with a particular interest in deep learning and neural networks. Regularly attend industry conferences and workshops to stay updated on emerging trends.\n- **Natural Language Processing and Conversational Agents:** Enthusiastic about the potential of NLP and conversational agents to transform human-computer interactions. Enjoy experimenting with chatbot development and language models.\n- **Cooking and Trying New Recipes:** Avid home cook who loves experimenting with new recipes and cuisines. Enjoys hosting dinner parties and sharing culinary creations with friends and family.\n- **Playing the Piano:** Dedicated pianist with over 10 years of experience. Enjoys playing classical and contemporary pieces, as well as composing original music.\n- **Watching Movies:** Film enthusiast with a particular love for comedies and thrillers. Enjoys analyzing film techniques and storytelling methods.\n\n## References\nAvailable upon request.\n\n---\n\n### Table: Technical Skills Proficiency\n\n| Skill | Proficiency Level | Years of Experience |\n|-------------------------------|-------------------|---------------------|\n| Python | Expert | 5 |\n| pandas | Expert | 4 |\n| scikit-learn | Advanced | 4 |\n| TensorFlow | Advanced | 3 |\n| Azure ML | Intermediate | 2 |\n| SQL | Intermediate | 3 |\n| Power BI | Intermediate | 2 |\n\n### List: Key Achievements\n- Successfully improved search result relevance by 15% through data-driven analysis and model optimization.\n- Developed a machine learning model that reduced customer churn by 10% for a major client.\n- Led a cross-functional team to implement a data governance framework, ensuring compliance with industry standards.\n\nThis enriched resume provides a comprehensive overview of Lisa's qualifications, experience, and interests, making it a compelling document for potential employers.",
+ "author": "Lisa"
+}
\ No newline at end of file
diff --git a/data/extractions/wordprocessor/Document/Resume.Lisa.md b/data/extractions/wordprocessor/Document/Resume.Lisa.md
new file mode 100644
index 0000000..8303d95
--- /dev/null
+++ b/data/extractions/wordprocessor/Document/Resume.Lisa.md
@@ -0,0 +1,79 @@
+# Lisa's Resume
+
+## Objective
+I am a highly motivated and detail-oriented data science professional seeking a challenging position in the field of data science. My goal is to leverage my extensive skills in machine learning, data analysis, and statistical modeling to contribute to innovative and impactful projects. I am particularly interested in roles that allow me to work on cutting-edge technologies and collaborate with cross-functional teams to drive data-driven decision-making and enhance business outcomes.
+
+## Skills
+
+### Technical Skills
+- **Programming Languages:** Proficient in Python, with extensive experience in using libraries such as pandas for data manipulation, scikit-learn for machine learning, and TensorFlow for deep learning applications. Additionally, I have hands-on experience with Azure ML for deploying and managing machine learning models in the cloud.
+- **Data Analysis Tools:** Familiar with SQL for database querying and data extraction, as well as Power BI for creating interactive data visualizations and dashboards to communicate insights effectively to stakeholders.
+- **Machine Learning and AI:** Strong foundation in machine learning algorithms, including supervised and unsupervised learning, natural language processing, and computer vision. Experienced in building, training, and evaluating models to solve complex business problems.
+- **Analytical and Problem-Solving Skills:** Demonstrated ability to analyze large datasets, identify patterns and trends, and develop actionable insights. Skilled in using statistical techniques to validate hypotheses and support data-driven decision-making.
+
+### Soft Skills
+- **Communication:** Excellent verbal and written communication skills, with the ability to present complex technical concepts to non-technical audiences. Experienced in writing technical reports and documentation.
+- **Collaboration:** Proven track record of working effectively in team environments, collaborating with colleagues from diverse backgrounds to achieve common goals. Strong interpersonal skills and the ability to build positive relationships with stakeholders.
+- **Adaptability:** Quick learner with the ability to adapt to new technologies and methodologies. Open to feedback and committed to continuous professional development.
+
+## Experience
+
+### Data Scientist at Microsoft (June 2018 - Present)
+- **User Behavior Analysis:** Conducted in-depth analysis of user behavior and feedback data to identify areas for improvement in search results. Developed data-driven strategies to enhance user experience and increase engagement.
+- **Machine Learning Model Development:** Built and tested machine learning models for various search scenarios, including personalized recommendations and query understanding. Collaborated with engineering teams to integrate models into production systems.
+- **Privacy and Security Compliance:** Ensured that all data and models adhered to privacy and security policies, working closely with legal and compliance teams to address any potential risks. Implemented best practices for data governance and model transparency.
+
+### Data Analyst Intern at TechCorp (June 2017 - May 2018)
+- Assisted in the development of data pipelines for processing and analyzing large datasets. Conducted exploratory data analysis to uncover insights and support business decision-making.
+- Created interactive dashboards and reports using Power BI to visualize key performance indicators and track progress against business objectives.
+
+## Education
+
+### Bachelor's Degree in Computer Science
+- **University of California, Berkeley (2014 - 2018)**
+ - Relevant Coursework: Data Structures and Algorithms, Machine Learning, Database Systems, Artificial Intelligence, Statistical Methods for Data Science
+ - Honors: Dean's List (2016, 2017), Member of the Computer Science Honor Society
+
+## Certifications
+- **Certified Data Scientist (CDS)**
+- **Microsoft Certified: Azure Data Scientist Associate**
+
+## Projects
+
+### Predictive Analytics for E-commerce
+- Developed a predictive analytics model to forecast sales trends and optimize inventory management for an e-commerce platform. Utilized time series analysis and regression techniques to improve demand forecasting accuracy.
+
+### Sentiment Analysis of Social Media Data
+- Implemented a sentiment analysis model to analyze customer feedback on social media platforms. Used natural language processing techniques to classify sentiments and provide actionable insights for marketing strategies.
+
+## Interests
+
+- **Artificial Intelligence and Machine Learning:** Passionate about exploring the latest advancements in AI and machine learning, with a particular interest in deep learning and neural networks. Regularly attend industry conferences and workshops to stay updated on emerging trends.
+- **Natural Language Processing and Conversational Agents:** Enthusiastic about the potential of NLP and conversational agents to transform human-computer interactions. Enjoy experimenting with chatbot development and language models.
+- **Cooking and Trying New Recipes:** Avid home cook who loves experimenting with new recipes and cuisines. Enjoys hosting dinner parties and sharing culinary creations with friends and family.
+- **Playing the Piano:** Dedicated pianist with over 10 years of experience. Enjoys playing classical and contemporary pieces, as well as composing original music.
+- **Watching Movies:** Film enthusiast with a particular love for comedies and thrillers. Enjoys analyzing film techniques and storytelling methods.
+
+## References
+Available upon request.
+
+---
+
+### Table: Technical Skills Proficiency
+
+| Skill | Proficiency Level | Years of Experience |
+|-------------------------------|-------------------|---------------------|
+| Python | Expert | 5 |
+| pandas | Expert | 4 |
+| scikit-learn | Advanced | 4 |
+| TensorFlow | Advanced | 3 |
+| Azure ML | Intermediate | 2 |
+| SQL | Intermediate | 3 |
+| Power BI | Intermediate | 2 |
+
+### List: Key Achievements
+- Successfully improved search result relevance by 15% through data-driven analysis and model optimization.
+- Developed a machine learning model that reduced customer churn by 10% for a major client.
+- Led a cross-functional team to implement a data governance framework, ensuring compliance with industry standards.
+
+This enriched resume provides a comprehensive overview of Lisa's qualifications, experience, and interests, making it a compelling document for potential employers.
\ No newline at end of file
diff --git a/docs/api/tinytroupe/agent.html b/docs/api/tinytroupe/agent.html
index 5cad79a..e30433d 100644
--- a/docs/api/tinytroupe/agent.html
+++ b/docs/api/tinytroupe/agent.html
@@ -1283,9 +1283,9 @@ Module tinytroupe.personfactory
+Module tinytroupe.factory
@@ -147,17 +147,17 @@ Module
logger.info(f"Starting the generation of the {number_of_factories} person factories based on that context: {generic_context_text}")
- person_factories_prompt = open(os.path.join(os.path.dirname(__file__), 'prompts/generate_person_factory.md')).read()
+ system_prompt = open(os.path.join(os.path.dirname(__file__), 'prompts/generate_person_factory.md')).read()
messages = []
- messages.append({"role": "system", "content": person_factories_prompt})
+ messages.append({"role": "system", "content": system_prompt})
- prompt = chevron.render("Please, create {{number_of_factories}} person descriptions based on the following broad context: {{context}}", {
+ user_prompt = chevron.render("Please, create {{number_of_factories}} person descriptions based on the following broad context: {{context}}", {
"number_of_factories": number_of_factories,
"context": generic_context_text
})
- messages.append({"role": "user", "content": prompt})
+ messages.append({"role": "user", "content": user_prompt})
response = openai_utils.client().send_message(messages)
@@ -269,7 +269,7 @@ Module
Module tinytroupe.personfactory
logger.info(f"Starting the generation of the {number_of_factories} person factories based on that context: {generic_context_text}")
- person_factories_prompt = open(os.path.join(os.path.dirname(__file__), 'prompts/generate_person_factory.md')).read()
+ system_prompt = open(os.path.join(os.path.dirname(__file__), 'prompts/generate_person_factory.md')).read()
messages = []
- messages.append({"role": "system", "content": person_factories_prompt})
+ messages.append({"role": "system", "content": system_prompt})
- prompt = chevron.render("Please, create {{number_of_factories}} person descriptions based on the following broad context: {{context}}", {
+ user_prompt = chevron.render("Please, create {{number_of_factories}} person descriptions based on the following broad context: {{context}}", {
"number_of_factories": number_of_factories,
"context": generic_context_text
})
- messages.append({"role": "user", "content": prompt})
+ messages.append({"role": "user", "content": user_prompt})
response = openai_utils.client().send_message(messages)
@@ -269,7 +269,7 @@ Module tinytroupe.personfactory
Classes
-
-
+class TinyFactory (simulation_id: str = None)Args
Subclasses
Class variables
-
-
var all_factories
+var all_factories
Static methods
-
-
+-def add_factory(factory)- @@ -398,7 +398,7 @@
Static methods
TinyFactory.all_factories[factory.name] = factory+-def clear_factories()- @@ -415,7 +415,7 @@
Static methods
TinyFactory.all_factories = {}+def set_simulation_for_free_factories(simulation)- @@ -439,7 +439,7 @@
Static methods
Methods
-
-
+-def decode_complete_state(self, state: dict)- @@ -458,7 +458,7 @@
Methods
return self+def encode_complete_state(self) ‑> dict- @@ -478,7 +478,7 @@
Methods
+
class TinyPersonFactory
(context_text, simulation_id: str = None)
Args
logger.info(f"Starting the generation of the {number_of_factories} person factories based on that context: {generic_context_text}")
- person_factories_prompt = open(os.path.join(os.path.dirname(__file__), 'prompts/generate_person_factory.md')).read()
+ system_prompt = open(os.path.join(os.path.dirname(__file__), 'prompts/generate_person_factory.md')).read()
messages = []
- messages.append({"role": "system", "content": person_factories_prompt})
+ messages.append({"role": "system", "content": system_prompt})
- prompt = chevron.render("Please, create {{number_of_factories}} person descriptions based on the following broad context: {{context}}", {
+ user_prompt = chevron.render("Please, create {{number_of_factories}} person descriptions based on the following broad context: {{context}}", {
"number_of_factories": number_of_factories,
"context": generic_context_text
})
- messages.append({"role": "user", "content": prompt})
+ messages.append({"role": "user", "content": user_prompt})
response = openai_utils.client().send_message(messages)
@@ -641,11 +641,11 @@ Args
Ancestors
Static methods
-
+
def generate_person_factories(number_of_factories, generic_context_text)
-
@@ -681,17 +681,17 @@
Returns
logger.info(f"Starting the generation of the {number_of_factories} person factories based on that context: {generic_context_text}")
- person_factories_prompt = open(os.path.join(os.path.dirname(__file__), 'prompts/generate_person_factory.md')).read()
+ system_prompt = open(os.path.join(os.path.dirname(__file__), 'prompts/generate_person_factory.md')).read()
messages = []
- messages.append({"role": "system", "content": person_factories_prompt})
+ messages.append({"role": "system", "content": system_prompt})
- prompt = chevron.render("Please, create {{number_of_factories}} person descriptions based on the following broad context: {{context}}", {
+ user_prompt = chevron.render("Please, create {{number_of_factories}} person descriptions based on the following broad context: {{context}}", {
"number_of_factories": number_of_factories,
"context": generic_context_text
})
- messages.append({"role": "user", "content": prompt})
+ messages.append({"role": "user", "content": user_prompt})
response = openai_utils.client().send_message(messages)
@@ -711,7 +711,7 @@ Returns
Methods
-
+
def generate_person(self, agent_particularities: str = None, temperature: float = 1.5, attepmpts: int = 5)
-
@@ -798,13 +798,13 @@
Returns
Inherited members
@@ -826,21 +826,21 @@ Index
Classes
diff --git a/docs/api/tinytroupe/index.html b/docs/api/tinytroupe/index.html
index bf7c0da..b1acf0b 100644
--- a/docs/api/tinytroupe/index.html
+++ b/docs/api/tinytroupe/index.html
@@ -37,13 +37,6 @@ Package tinytroupe
sys.path.append('.')
from tinytroupe import utils # now we can import our utils
-config = utils.read_config_file()
-utils.start_logger(config)
-
-# fix an issue in the rich library: we don't want margins in Jupyter!
-rich.jupyter.JUPYTER_HTML_FORMAT = \
- utils.inject_html_css_style_prefix(rich.jupyter.JUPYTER_HTML_FORMAT, "margin:0px;")
-
# AI disclaimers
print(\
"""
@@ -52,7 +45,15 @@ Package tinytroupe
The AI models are not perfect and may produce inappropriate or inacurate results.
For any serious or consequential use, please review the generated content before using it.
!!!!
-""")
+""")
+
+config = utils.read_config_file()
+utils.pretty_print_config(config)
+utils.start_logger(config)
+
+# fix an issue in the rich library: we don't want margins in Jupyter!
+rich.jupyter.JUPYTER_HTML_FORMAT = \
+ utils.inject_html_css_style_prefix(rich.jupyter.JUPYTER_HTML_FORMAT, "margin:0px;")