Add multi agent #23
Merged
Add multi agent #23
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
skrawcz
force-pushed
the
add_telephone
branch
from
257fbb6
to
31df27a
Compare
skrawcz
force-pushed
the
add-multi-agent
branch
3 times, most recently
from
abc053b
to
1655b60
Compare
skrawcz
force-pushed
the
add_telephone
branch
from
a4f970c
to
0b7bc6f
Compare
elijahbenizzy
force-pushed
the
add_telephone
branch
from
140e8bb
to
5fa877d
Compare
|
@skrawcz you're going to want to cherry-pick the commits for just this. I'm going to add input/state serialization to a separate PR. |
skrawcz
force-pushed
the
add-multi-agent
branch
from
1655b60
to
0841cf2
Compare
elijahbenizzy
approved these changes
Feb 26, 2024
burr/core/action.py
Outdated
| @@ -241,7 +241,7 @@ def __init__(self, *fields: str): | |||
| super(Result, self).__init__() | |||
| self._fields = fields | |||
|
|
|||
| def run(self, state: State) -> dict: | |||
| def run(self, state: State, **kwargs) -> dict: | |||
There was a problem hiding this comment.
This doesn't take in kwargs -- they're not allowed to be passed in as inputs is empty. Type-checking can't really express it so I figured it's better to just not have them there.
There was a problem hiding this comment.
ok will drop this commit then.
skrawcz
force-pushed
the
add-multi-agent
branch
from
0841cf2
to
febd6a7
Compare
skrawcz
commented
Feb 26, 2024
skrawcz
force-pushed
the
add-multi-agent
branch
from
febd6a7
to
fbd640d
Compare
Adds serialization hook for lcel example (+12 squashed commits) Squashed commits: [966cffd] Fixes python repl response If this doesn't work I should revert to the lecl response structure of their python repl function. [ac0a0a1] Gets it almost working the LCEL one uses the deprecated function paths of open ai API. I'm trying to use the new tool one here. It's a bit inconsistent it seems -- I did manage to get it to draw a graph, but it failed to finish because I hadn't set up how it should say to finish things. Otherwise a UI to debug this would be great. [12feac6] Adds working lcel multi-agent example Will now take this and rewrite with Hamilton/more obvious constructs. [4ef64f2] Adds beginnings of multi-agent example Need to figure out the application state model and what to update, i.e. wire through messages history properly. Otherwise rather than having a specific tool node, like the langgraph example, I added that call to the hamilton DAG for it -- could always change it to mirror their stuff. [f3ffe95] Adds telephone example README [03b6d07] Addresses comments [11d0b0f] Fixes formatting [17ffa67] Adds telephone app example This uses the dataflows from the hub and then shows two ways to integrate Hamilton. One via the Hamilton integration, the other via the function API. [557b24b] Adds image telephone example This uses the dataflows defined on the hub to create the image telephone app. [0233a95] Fixes cowsay application __main__ the variables were round the wrong way. [dbeddb0] WIP [0ba0158] WIP (+1 squashed commit) Squashed commits: [bc53419] Initializes telemetry UI Design: 1. CRA + tailwind (they don't like CRA but I really don't care) 2. prettier + eslint 3. Add to pre-commit-config.yaml rahte rthan husky (which is very painful IMO)
So that things run and docs don't have misspellings.
This will be useful for debugging. E.g. in the UI have the code to replay from this point int state. Then boom you can go and try to debug what was going on more easily.
skrawcz
force-pushed
the
add-multi-agent
branch
from
fbd640d
to
9010e72
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Adds beginnings of multi-agent example
Need to figure out the application state model and
what to update, i.e. wire through messages history
properly.
Otherwise rather than having a specific tool node, like
the langgraph example, I added that call to the hamilton
DAG for it -- could always change it to mirror their stuff.