Interactive UI for Swarms Framework #727

harshalmore31 · 2025-01-07T16:59:45Z

Interactive UI for Swarms Framework: Swarm-Specific Parsing, Agent Management, and LLM Integration with Hugging Face Deployment & Local Deployment

Description:

Developed a robust and feature-rich UI for the Swarms framework using Gradio, focused on enhancing usability, modularity, and interactive capabilities. Key additions include:
- Swarm-Specific Parsing:
  - Implemented parsers for all swarm types, including SequentialWorkflow, ConcurrentWorkflow, AgentRearrange, MixtureOfAgents, SpreadSheetSwarm, and auto.
  - Ensures consistent formatting of output data and detailed feedback for each workflow type.
- Dynamic Prompt Loading:
  - Integrated functionality to load agent prompts dynamically from a JSON file.
  - Default prompts are used if the file is absent or improperly formatted.
- Agent Management in UI:
  - Enabled creation of new agent prompts directly via the UI.
  - Supports editing and validation of custom prompts with seamless integration into the workflow configuration.
- UI Framework Enhancements:
  - Introduced a dedicated UI class for streamlined Gradio component management, improving scalability and readability of the codebase.
  - Added modular components like dropdowns, sliders, and conditional inputs for an intuitive user experience.
- LLM Integration:
  - Using LiteLLM, integrated the following models and providers:
    - OpenAI
    - Anthropic
    - Cohere
    - Gemini
    - Mistral
    - Groq
    - Perplexity

Key Features:

Interactive Swarm Configuration: Supports flexible configuration for workflows, including custom execution flows for AgentRearrange and advanced features like agent flow validation.
Agent-Centric Design: Facilitates agent selection, prompt management, and dynamic updates for added flexibility.
User Feedback and Debugging: Incorporates detailed logging, error handling, and status displays for enhanced user experience.

Issue:

Not linked to any specific issue. This PR focuses on establishing a functional and scalable UI layer for the Swarms framework.

Dependencies:

Gradio: Core framework for building the modular UI.
dotenv: Securely manages API keys and configurations.
litellm: Enables seamless integration with multiple language model providers and their respective models.

Additional Changes:

spreadsheet_swarm.py is modified accordingly for the working of UI

📚 Documentation preview 📚: https://swarms--727.org.readthedocs.build/en/727/

Added - Tasks - max_workers

- handles task assignment properly

…lows Fix AttributeError in ConcurrentWorkflow by Properly Initializing Tasks List

- other functionality remains same , code is just slightly rearranged if anything .

- when response is a choice , this determines how it is handled. - current implementation uses a placeholder for llm_output_parser , it needs to be updated ( next commit )

- implements the decision making , when 'choice' type output is received .

Add TaskQueueSwarm documentation

…-Access-Issue Ensure tool execution regardless of long-term memory usage

- added asdict to turn things into json compatible output

- adding updated log methods to run

- make the run update for tools

- edit "run" return type with condition on how to return if output type is JSON .

- add newer methods to the log_step_metadata for logging

- update tool usage fn to add and account for tool usage metadata.

- minor refactor

- update correct libraries & dependencies .

- create class to execute modular unittests - def setup for modular setup - objective to keep setup minimal , so that tests aren't bloated and fast to run - Since most param have a set default , init of necessary condition is a valid and supportive op wrt test speed .

- Implemented test_log_step_metadata_basic to verify the correct logging of step metadata including step_id, timestamp, tokens, and memory usage. - Confirmed that the token counts for total are accurately logged.

- Implemented test_log_step_metadata_no_long_term_memory to ensure that when long-term memory is None, the memory_usage for long_term is an empty dictionary in the log result.

- Implemented test_log_step_metadata_timestamp to verify that a timestamp is included in the log result when logging step metadata.

- Implemented test_token_counting_integration to verify the correct total token count when using a mocked tokenizer, ensuring that prompt and response token counts are accurately aggregated.

- Implemented test_agent_output_updating to verify that logging step metadata correctly updates the total token count and ensures that the agent's output steps are properly tracked confirming only one step is recorded.

--- updated-dependencies: - dependency-name: ruff dependency-type: direct:production ... Signed-off-by: dependabot[bot] <[email protected]>

Updates the requirements on [pygments](https://github.com/pygments/pygments) to permit the latest version. - [Release notes](https://github.com/pygments/pygments/releases) - [Changelog](https://github.com/pygments/pygments/blob/master/CHANGES) - [Commits](pygments/pygments@2.18.0...2.19.0) --- updated-dependencies: - dependency-name: pygments dependency-type: direct:production ... Signed-off-by: dependabot[bot] <[email protected]>

minor fix

[README]

…-0.5.1-and-lt-0.8.7 Update ruff requirement from >=0.5.1,<0.8.5 to >=0.5.1,<0.8.7

…-approx-eq-2.19 Update pygments requirement from ~=2.18 to ~=2.19

typo

evelynmitchell · 2025-01-07T19:08:05Z

Thanks Harshal!

In the HuggingFace space, the contrast in the UI is very low (grey on grey).

There is also no visual indicator that 'Select Provider' is a pulldow menu, and the text in it is even lighter than in the text box.

Can you increase the contrast a bit?

As a feature improvement, tags to filter the Agents on would be helpful (role, business department, frequently used with, last updated, ...) A long pulldown menu gets unusable quickly.

kyegomez · 2025-01-09T18:32:24Z

swarms/structs/ui/ui.py

+
+if __name__ == "__main__":
+    app = create_app()
+    app.launch()


comment that out

sambhavnoobcoder and others added 30 commits October 22, 2024 17:51

Merge branch 'kyegomez:master' into Longterm-Memory-Access-Issue

63e00b8

update __init__

8d7e07d

Added - Tasks - max_workers

Handles task assignment

5d17c50

- handles task assignment properly

Merge pull request kyegomez#613 from sambhavnoobcoder/ConcurrentWorkf…

1271980

…lows Fix AttributeError in ConcurrentWorkflow by Properly Initializing Tasks List

{EXAMPLE CLEANUP]

2807ab7

{EXAMPLE CLEANUP]

de75a5a

response_args should include image

c60789e

- other functionality remains same , code is just slightly rearranged if anything .

add section for 'choice' based responses .

e06e898

- when response is a choice , this determines how it is handled. - current implementation uses a placeholder for llm_output_parser , it needs to be updated ( next commit )

proper implemenation of llm_output_parser

a299a58

- implements the decision making , when 'choice' type output is received .

minor refactor

db821f0

style check

a0ab6af

changes

486a3cf

mkdocs and and swarms.models fix

e9818db

Merge pull request kyegomez#608 from Occupying-Mars/basic-fix

f31fa4f

Add TaskQueueSwarm documentation

Merge pull request kyegomez#611 from sambhavnoobcoder/Longterm-Memory…

46c206c

…-Access-Issue Ensure tool execution regardless of long-term memory usage

Update README.md

9fced20

adding asdict

38db41b

- added asdict to turn things into json compatible output

add log methods to run

0d2d7fd

- adding updated log methods to run

add section for tools update

e8b8d48

- make the run update for tools

edit run return

8b552ba

- edit "run" return type with condition on how to return if output type is JSON .

update the log_step_metadata

68806eb

- add newer methods to the log_step_metadata for logging

Add Update_tool_usage

2a8cf25

- update tool usage fn to add and account for tool usage metadata.

minor refactor

b40c76e

- minor refactor

Push libraries

92912b0

- update correct libraries & dependencies .

Add basic test for log_step_metadata method

4d300fc

- Implemented test_log_step_metadata_basic to verify the correct logging of step metadata including step_id, timestamp, tokens, and memory usage. - Confirmed that the token counts for total are accurately logged.

Add test for log_step_metadata with no long-term memory

672f3ee

- Implemented test_log_step_metadata_no_long_term_memory to ensure that when long-term memory is None, the memory_usage for long_term is an empty dictionary in the log result.

Add test for timestamp in log_step_metadata

e80d182

- Implemented test_log_step_metadata_timestamp to verify that a timestamp is included in the log result when logging step metadata.

Add integration test for token counting in log_step_metadata

8900d64

- Implemented test_token_counting_integration to verify the correct total token count when using a mocked tokenizer, ensuring that prompt and response token counts are accurately aggregated.

Add test for agent output updating in log_step_metadata

d72d2e7

- Implemented test_agent_output_updating to verify that logging step metadata correctly updates the total token count and ensures that the agent's output steps are properly tracked confirming only one step is recorded.

kyegomez and others added 22 commits January 3, 2025 13:09

[CLEANUP]

857099a

[CLEANUP]

6730883

[CLEANUP]

c439df4

[CLEANUP]

bb69f8e

[CLEANUP]

704f91f

[FEATS][CSVTOAgents] [Various Bug Fixes]

5ecdc87

[FEATS][CSVTOAgents] [Various Bug Fixes]

3df461b

[EXAMPLE][Multi-AgentRouter]

b136592

[CLEANUP]

f1f1133

[WRAP UP]

48140c5

[6.8.8]

abcb07a

Update ruff requirement from >=0.5.1,<0.8.5 to >=0.5.1,<0.8.7

9cf5510

--- updated-dependencies: - dependency-name: ruff dependency-type: direct:production ... Signed-off-by: dependabot[bot] <[email protected]>

[README]

d81038d

minor fix

Merge pull request kyegomez#720 from eltociear/patch-2

a1e791e

[README]

Merge pull request kyegomez#717 from kyegomez/dependabot/pip/ruff-gte…

27d4e2c

…-0.5.1-and-lt-0.8.7 Update ruff requirement from >=0.5.1,<0.8.5 to >=0.5.1,<0.8.7

Merge pull request kyegomez#718 from kyegomez/dependabot/pip/pygments…

3a0d2c4

…-approx-eq-2.19 Update pygments requirement from ~=2.18 to ~=2.19

typo

d1cc1be

Merge pull request kyegomez#723 from krankos/patch-1

b408faf

typo

ui.py and spreadsheet_swarm.py updated

28ef730

Update requirements.txt to add clusterops, litellm, and gradio

dfdd31a

Update requirements.txt to include new dependencies

6ae0b7c

github-actions bot added the structs label Jan 7, 2025

Enhance UI theme support with dark mode detection

3c074ae

evelynmitchell approved these changes Jan 8, 2025

View reviewed changes

kyegomez reviewed Jan 9, 2025

View reviewed changes

swarms/structs/ui/ui.py Outdated

if __name__ == "__main__":

app = create_app()

app.launch()

Copy link

Owner

kyegomez Jan 9, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

comment that out

Add auto-installation of required packages in ui.py

ae11915

kyegomez closed this Jan 9, 2025

kyegomez force-pushed the master branch from afd5419 to 7de9ee8 Compare January 9, 2025 18:51

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Interactive UI for Swarms Framework #727

Interactive UI for Swarms Framework #727

harshalmore31 commented Jan 7, 2025 •

edited by github-actions bot

Loading

evelynmitchell commented Jan 7, 2025

kyegomez Jan 9, 2025

Interactive UI for Swarms Framework #727

Interactive UI for Swarms Framework #727

Conversation

harshalmore31 commented Jan 7, 2025 • edited by github-actions bot Loading

Interactive UI for Swarms Framework: Swarm-Specific Parsing, Agent Management, and LLM Integration with Hugging Face Deployment & Local Deployment

Description:

Key Features:

Issue:

Dependencies:

Deployment:

Tag Maintainer:

Twitter Handle:

Additional Changes:

evelynmitchell commented Jan 7, 2025

kyegomez Jan 9, 2025

Choose a reason for hiding this comment

harshalmore31 commented Jan 7, 2025 •

edited by github-actions bot

Loading