Skip to content
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.

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

Merges latest development version into main. #90

Merged
merged 34 commits into from
Jan 29, 2025
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
34 commits
Select commit Hold shift + click to select a range
88fc4d1
Fix spelling and grammatical errors
terrchen Nov 12, 2024
72c1f8e
Rename .env.local to .env.example
webysther Nov 13, 2024
d146506
docs: update README.md
eltociear Nov 14, 2024
54d3d87
variable typo #18 fixed
rahimbaig28 Nov 14, 2024
ca34193
add matplotlib to dependencies
RektPunk Nov 15, 2024
fe8e224
fix typo, call method of agent not self
ewheeler Nov 25, 2024
bd20904
also retry aux_act_once upon TypeError
ewheeler Nov 25, 2024
62c8daf
allow configuration of Azure OpenAI embedding model
ewheeler Nov 25, 2024
37529c4
Adds missing comma
paulosalem Nov 28, 2024
1622a1a
Marks the embedding model API version as Azure-specific.
paulosalem Nov 28, 2024
0d626bf
Adds the "azure_" prefix to embedding model API version specific to A…
paulosalem Nov 28, 2024
665b536
Merge branch 'development' into azure_openai_embeddings_conf
paulosalem Nov 28, 2024
96d7e62
Merge pull request #59 from ewheeler/azure_openai_embeddings_conf
paulosalem Nov 28, 2024
cd057a9
Merge pull request #19 from terrchen/main
paulosalem Nov 28, 2024
5ae2659
Renames demo .env file
paulosalem Nov 28, 2024
abe3fc6
Merge branch 'development' into patch-1
paulosalem Nov 28, 2024
4c211d8
Merge pull request #22 from webysther/patch-1
paulosalem Nov 28, 2024
12a8342
Merge pull request #57 from ewheeler/call_on_agent_not_self
paulosalem Nov 28, 2024
982a67a
Merge pull request #24 from eltociear/patch-1
paulosalem Nov 28, 2024
bb0e17c
Merge branch 'development' into chore/add-matplotlib
paulosalem Nov 28, 2024
3b7fb00
Merge pull request #33 from RektPunk/chore/add-matplotlib
paulosalem Nov 28, 2024
ff84048
Merge branch 'development' into typo
paulosalem Nov 28, 2024
edc65b9
Merge pull request #25 from rahimbaig28/typo
paulosalem Nov 28, 2024
67ba2bd
Merge pull request #58 from ewheeler/retry_aux_act_once
paulosalem Nov 28, 2024
b3e9070
Adds some seasoning to the gazpacho example.
paulosalem Nov 29, 2024
a22da4a
Refactors grounding data access mechanisms, tweaks other details.
paulosalem Dec 7, 2024
dfee478
Refactors agent and environment modules in multiple files for better …
paulosalem Dec 12, 2024
585a498
Refactors more modules as multiple files, for better organization
paulosalem Dec 17, 2024
d417d3f
Adds deeper persona structure, including fragments.
paulosalem Jan 3, 2025
df00bdf
Merge pull request #86 from microsoft/users/paulosalem/denser_persona…
paulosalem Jan 3, 2025
62d85be
Implements logical propositions and interventions.
paulosalem Jan 29, 2025
166b53b
Adds news to the README
paulosalem Jan 29, 2025
86d8a6a
Merge branch 'main' into development
paulosalem Jan 29, 2025
10cc8b3
Fixes test.
paulosalem Jan 29, 2025
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 6 additions & 0 deletions .env.example
Original file line number Diff line number Diff line change
@@ -0,0 +1,6 @@
# Define the OpenAI API key here if you want to use a local .env file for that.
OPENAI_API_KEY=....

# Or the Azure OpenAI API details
AZURE_OPENAI_API_KEY=....
AZURE_OPENAI_ENDPOINT=....
2 changes: 0 additions & 2 deletions .env.local

This file was deleted.

248 changes: 196 additions & 52 deletions README.md

Large diffs are not rendered by default.

2 changes: 1 addition & 1 deletion data/extractions/appartment_rent_ad.extraction.json
Original file line number Diff line number Diff line change
Expand Up @@ -2,7 +2,7 @@
"agent_extractions": {},
"world_extraction": {
"Focus group": {
"ad_copy": "Don't miss out on this unique opportunity! Contact Paulo today to schedule a viewing of this spacious, modern apartment featuring elegant renovations, a quiet location despite being on a main avenue, and all the furnishings you need to move in quickly. Enjoy the convenience of nearby supermarkets, restaurants, and public transport, all while being close to the best Medicine School in the country. This apartment is perfect for students and anyone looking for a comfortable home!"
"ad_copy": "Your peaceful oasis in the heart of the city! Experience modern living in a tranquil setting. This spacious, elegantly designed apartment features integrated spaces, a modern kitchen with high-quality appliances, and is just steps away from the best Medicine School. Enjoy the convenience of nearby supermarkets, bakeries, and restaurants, all while living in a quiet retreat. Don't miss out on this unique opportunity to live in style and comfort!"
}
}
}
21 changes: 10 additions & 11 deletions data/extractions/synthetic_data_generation.out.csv
Original file line number Diff line number Diff line change
@@ -1,13 +1,12 @@
author,content
USER,"Get some work done together, help each other."
Lucas Martinez,I think working together is a great idea! What tasks do you want to tackle first?
Lucas Martinez,"Hey Clara, I'm ready to get started on our projects. What do you think we should focus on first?"
Clara Thompson,I think we should start by brainstorming new content ideas for our upcoming campaign. What do you think?
Lucas Martinez,I think brainstorming new content ideas is a great starting point! I have a few ideas in mind that we can discuss. What do you think?
Clara Thompson,I'm excited to hear your ideas! What do you have in mind?
Lucas Martinez,I have a few ideas in mind! One is to create interactive social media posts that engage our audience. Another idea is to encourage user-generated content by running a contest. What do you think?
Clara Thompson,I really like your ideas! Creating interactive social media posts and running a contest for user-generated content could really engage our audience. Let's discuss how we can implement these.
Lucas Martinez,Great! Let's start by discussing the platforms we want to use for the interactive posts. I think Instagram and Facebook could be effective. What do you think?
Clara Thompson,I agree that Instagram and Facebook are great choices! We could also consider Twitter for quick updates or TikTok for more engaging video content. What do you think?
Lucas Martinez,I think incorporating Twitter for updates and TikTok for engaging videos is a fantastic idea! We can create content that fits each platform's style. Let's brainstorm some specific ideas for each one.
Clara Thompson,"Absolutely! Let's brainstorm some specific ideas for Twitter and TikTok. For Twitter, we could share quick tips related to our content, and for TikTok, maybe we can create fun behind-the-scenes videos of our process. What do you think?"
Emily Carter,That sounds like a great idea! I'm always up for collaborating and helping each other out. What specific tasks do you have in mind?
Lucas Thompson,I was thinking we could work on some social media content together or analyze the performance of our recent campaigns. What do you think?
Emily Carter,"I think both options are great! Creating social media content could help us engage our audience, while analyzing our recent campaigns will give us insights into what’s working. Maybe we can start with the content creation and then move on to the analysis?"
Lucas Thompson,"That sounds like a solid plan! For the content creation, do you have any specific themes or ideas in mind that we should focus on?"
Emily Carter,"I was thinking we could focus on themes like seasonal promotions, user-generated content, or even behind-the-scenes looks at our agency. What do you think about those ideas?"
Lucas Thompson,"I love those ideas! We could definitely create some engaging posts around seasonal promotions, and user-generated content would be a fantastic way to involve our audience. Maybe we can also share some behind-the-scenes moments to give a personal touch. How about we brainstorm some specific posts for each theme?"
Emily Carter,"Great! For the seasonal promotions, we could create posts highlighting special offers or discounts. For user-generated content, we can encourage our audience to share their experiences with our services and feature their posts. As for behind-the-scenes moments, we could share photos or videos of our team at work or during team-building activities. What do you think?"
Lucas Thompson,"I think those are fantastic ideas! Highlighting special offers for seasonal promotions will definitely catch attention. Encouraging user-generated content is a great way to build community, and sharing behind-the-scenes moments will give our audience a glimpse into our culture. Maybe we can also create a specific hashtag for the user-generated content to make it easier to track?"
Emily Carter,That's a great idea! We could use a hashtag like #CreativeSolutionsCommunity or #ShareYourExperience. It should be something that resonates with our audience and encourages them to participate. What do you think?
Lucas Thompson,I really like both hashtags! They resonate well with our audience and encourage participation. I think we should go with #CreativeSolutionsCommunity since it emphasizes our brand and community aspect. What do you think?
Binary file not shown.
Original file line number Diff line number Diff line change
@@ -0,0 +1,5 @@
{
"title": "Resume of Lisa Carter",
"content": "```\n# Lisa Carter\n\n## Contact Information\n- **Email:** [email protected]\n- **Phone:** (123) 456-7890\n- **LinkedIn:** [linkedin.com/in/lisacarter](https://linkedin.com/in/lisacarter)\n\n## Summary\nLisa Carter is a highly skilled Data Scientist with a Master's degree in Data Science from the prestigious University of Toronto. With a specialization in artificial intelligence (AI) and machine learning, Lisa has a proven track record of significantly improving search relevance and enhancing user experience through meticulous data analysis and innovative model development. Her analytical prowess and technical expertise have enabled her to contribute effectively to various projects, making her a valuable asset in any data-driven environment.\n\n## Education\n**University of Toronto** \n*Master's in Data Science* \n*Graduation Date: Month Year* \n**Thesis:** Improving Search Relevance Using Context-Aware Models \nLisa's thesis focused on developing advanced algorithms that leverage contextual information to enhance the accuracy and relevance of search results. This research not only contributed to her academic growth but also provided practical insights applicable in real-world scenarios.\n\n**Relevant Coursework:** \n- **Machine Learning:** Explored supervised and unsupervised learning techniques, including regression, classification, clustering, and neural networks.\n- **Natural Language Processing:** Studied the intricacies of language processing, including sentiment analysis, text classification, and language generation.\n- **Data Visualization:** Learned to effectively communicate data insights through visual representation using tools like Matplotlib and Seaborn.\n\n## Work Experience\n### Microsoft, M365 Search Team \n*Data Scientist* \n*Date: Month Year - Present* \nIn her current role at Microsoft, Lisa plays a pivotal role in enhancing the search functionality of the M365 suite. Her responsibilities include:\n\n- **User Behavior Analysis:** Analyzed user behavior and feedback data to identify patterns and trends that inform search result enhancements. This involved utilizing A/B testing and user surveys to gather actionable insights.\n- **Model Development:** Developed and tested machine learning models for natural language understanding and query expansion, resulting in a 20% increase in user satisfaction scores.\n- **Cross-Functional Collaboration:** Collaborated with cross-functional teams, including product managers, software engineers, and UX designers, to address data quality issues and improve model accuracy. This collaboration ensured that the models were not only technically sound but also aligned with user needs.\n\n### Tech Startup (Internship) \n*Data Science Intern* \n*Date: Month Year - Month Year* \nDuring her internship at a dynamic tech startup, Lisa gained hands-on experience in the field of data science. Her key contributions included:\n\n- **Conversational AI Models:** Assisted in building conversational AI models that improved user interaction and engagement. This involved training models on diverse datasets to ensure robustness and reliability.\n- **Data Cleaning and Preprocessing:** Conducted extensive data cleaning and preprocessing to ensure high-quality datasets, which are crucial for effective model training and evaluation. This included handling missing values, outlier detection, and normalization.\n\n## Skills\n- **Programming Languages:** Proficient in Python, with extensive experience in libraries such as pandas, scikit-learn, and TensorFlow. Familiar with R for statistical analysis.\n- **Data Management:** Strong knowledge of SQL for database management and data retrieval. Experienced in using Power BI for data visualization and reporting.\n- **Analytical Skills:** Exceptional analytical and problem-solving skills, with a keen ability to interpret complex data sets and derive meaningful insights.\n- **Machine Learning Frameworks:** Experienced in using Azure ML for deploying machine learning models in cloud environments.\n\n## Interests\n- **Artificial Intelligence and Machine Learning:** Passionate about exploring the latest advancements in AI and machine learning technologies, and their applications in various industries.\n- **Natural Language Processing:** Enthusiastic about developing conversational agents and improving human-computer interaction through NLP techniques.\n- **Culinary Arts:** Enjoys cooking and experimenting with new recipes, often drawing inspiration from different cultures and cuisines.\n- **Music:** Plays the piano and enjoys composing music, finding it a great way to express creativity and unwind.\n\n## References\nReferences are available upon request. Lisa is happy to provide contacts from her academic and professional network who can speak to her skills, work ethic, and contributions in the field of data science.\n```",
"author": "Lisa Carter"
}
Original file line number Diff line number Diff line change
@@ -0,0 +1,56 @@
```
# Lisa Carter

## Contact Information
- **Email:** [email protected]
- **Phone:** (123) 456-7890
- **LinkedIn:** [linkedin.com/in/lisacarter](https://linkedin.com/in/lisacarter)

## Summary
Lisa Carter is a highly skilled Data Scientist with a Master's degree in Data Science from the prestigious University of Toronto. With a specialization in artificial intelligence (AI) and machine learning, Lisa has a proven track record of significantly improving search relevance and enhancing user experience through meticulous data analysis and innovative model development. Her analytical prowess and technical expertise have enabled her to contribute effectively to various projects, making her a valuable asset in any data-driven environment.

## Education
**University of Toronto**
*Master's in Data Science*
*Graduation Date: Month Year*
**Thesis:** Improving Search Relevance Using Context-Aware Models
Lisa's thesis focused on developing advanced algorithms that leverage contextual information to enhance the accuracy and relevance of search results. This research not only contributed to her academic growth but also provided practical insights applicable in real-world scenarios.

**Relevant Coursework:**
- **Machine Learning:** Explored supervised and unsupervised learning techniques, including regression, classification, clustering, and neural networks.
- **Natural Language Processing:** Studied the intricacies of language processing, including sentiment analysis, text classification, and language generation.
- **Data Visualization:** Learned to effectively communicate data insights through visual representation using tools like Matplotlib and Seaborn.

## Work Experience
### Microsoft, M365 Search Team
*Data Scientist*
*Date: Month Year - Present*
In her current role at Microsoft, Lisa plays a pivotal role in enhancing the search functionality of the M365 suite. Her responsibilities include:

- **User Behavior Analysis:** Analyzed user behavior and feedback data to identify patterns and trends that inform search result enhancements. This involved utilizing A/B testing and user surveys to gather actionable insights.
- **Model Development:** Developed and tested machine learning models for natural language understanding and query expansion, resulting in a 20% increase in user satisfaction scores.
- **Cross-Functional Collaboration:** Collaborated with cross-functional teams, including product managers, software engineers, and UX designers, to address data quality issues and improve model accuracy. This collaboration ensured that the models were not only technically sound but also aligned with user needs.

### Tech Startup (Internship)
*Data Science Intern*
*Date: Month Year - Month Year*
During her internship at a dynamic tech startup, Lisa gained hands-on experience in the field of data science. Her key contributions included:

- **Conversational AI Models:** Assisted in building conversational AI models that improved user interaction and engagement. This involved training models on diverse datasets to ensure robustness and reliability.
- **Data Cleaning and Preprocessing:** Conducted extensive data cleaning and preprocessing to ensure high-quality datasets, which are crucial for effective model training and evaluation. This included handling missing values, outlier detection, and normalization.

## Skills
- **Programming Languages:** Proficient in Python, with extensive experience in libraries such as pandas, scikit-learn, and TensorFlow. Familiar with R for statistical analysis.
- **Data Management:** Strong knowledge of SQL for database management and data retrieval. Experienced in using Power BI for data visualization and reporting.
- **Analytical Skills:** Exceptional analytical and problem-solving skills, with a keen ability to interpret complex data sets and derive meaningful insights.
- **Machine Learning Frameworks:** Experienced in using Azure ML for deploying machine learning models in cloud environments.

## Interests
- **Artificial Intelligence and Machine Learning:** Passionate about exploring the latest advancements in AI and machine learning technologies, and their applications in various industries.
- **Natural Language Processing:** Enthusiastic about developing conversational agents and improving human-computer interaction through NLP techniques.
- **Culinary Arts:** Enjoys cooking and experimenting with new recipes, often drawing inspiration from different cultures and cuisines.
- **Music:** Plays the piano and enjoys composing music, finding it a great way to express creativity and unwind.

## References
References are available upon request. Lisa is happy to provide contacts from her academic and professional network who can speak to her skills, work ethic, and contributions in the field of data science.
```
Binary file not shown.
Binary file not shown.
Binary file not shown.
Loading
Loading