Merge branch 'master' into master

pull/339/head
evelynmitchell 1 year ago committed by GitHub
commit 8e08408baa
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23

@ -7,20 +7,25 @@ load_dotenv()
# Load environment variables
llm = OpenAIChat(openai_api_key=os.getenv("OPENAI_API_KEY"))
agent = Agent(llm=llm, max_loops=1)
agent = Agent(
llm=llm,
max_loops=1,
)
# Create a workflow
workflow = ConcurrentWorkflow(max_workers=5)
# Create tasks
task1 = Task(agent, "What's the weather in miami")
task2 = Task(agent, "What's the weather in new york")
task3 = Task(agent, "What's the weather in london")
task1 = Task(agent=agent, description="What's the weather in miami")
task2 = Task(
agent=agent, description="What's the weather in new york"
)
task3 = Task(agent=agent, description="What's the weather in london")
# Add tasks to the workflow
workflow.add(task1)
workflow.add(task2)
workflow.add(task3)
# Run the workflow
# Run the workflow and print each task result
workflow.run()

@ -4,7 +4,7 @@ build-backend = "poetry.core.masonry.api"
[tool.poetry]
name = "swarms"
version = "3.1.9"
version = "3.2.5"
description = "Swarms - Pytorch"
license = "MIT"
authors = ["Kye Gomez <kye@apac.ai>"]

@ -2,27 +2,39 @@
import inspect
import os
import threading
from zeta import OpenAIChat
from swarms import OpenAIChat
from scripts.auto_tests_docs.docs import DOCUMENTATION_WRITER_SOP
from zeta.nn.modules._activations import (
AccurateGELUActivation,
ClippedGELUActivation,
FastGELUActivation,
GELUActivation,
LaplaceActivation,
LinearActivation,
MishActivation,
NewGELUActivation,
PytorchGELUTanh,
QuickGELUActivation,
ReLUSquaredActivation,
from swarms.structs.agent import Agent
from swarms.structs.autoscaler import AutoScaler
from swarms.structs.base import BaseStructure
from swarms.structs.base_swarm import AbstractSwarm
from swarms.structs.base_workflow import BaseWorkflow
from swarms.structs.concurrent_workflow import ConcurrentWorkflow
from swarms.structs.conversation import Conversation
from swarms.structs.groupchat import GroupChat, GroupChatManager
from swarms.structs.model_parallizer import ModelParallelizer
from swarms.structs.multi_agent_collab import MultiAgentCollaboration
from swarms.structs.nonlinear_workflow import NonlinearWorkflow
from swarms.structs.recursive_workflow import RecursiveWorkflow
from swarms.structs.schemas import (
Artifact,
ArtifactUpload,
StepInput,
TaskInput,
)
from swarms.structs.sequential_workflow import SequentialWorkflow
from swarms.structs.swarm_net import SwarmNetwork
from swarms.structs.utils import (
distribute_tasks,
extract_key_from_json,
extract_tokens_from_text,
find_agent_by_id,
find_token_in_text,
parse_tasks,
)
from zeta.nn.modules.dense_connect import DenseBlock
from zeta.nn.modules.dual_path_block import DualPathBlock
from zeta.nn.modules.feedback_block import FeedbackBlock
from zeta.nn.modules.highway_layer import HighwayLayer
from zeta.nn.modules.multi_scale_block import MultiScaleBlock
from zeta.nn.modules.recursive_block import RecursiveBlock
from dotenv import load_dotenv
load_dotenv()
@ -36,59 +48,76 @@ model = OpenAIChat(
)
def process_documentation(cls):
def process_documentation(
item,
module: str = "swarms.structs",
docs_folder_path: str = "docs/swarms/structs",
):
"""
Process the documentation for a given class using OpenAI model and save it in a Markdown file.
Process the documentation for a given class or function using OpenAI model and save it in a Python file.
"""
doc = inspect.getdoc(cls)
source = inspect.getsource(cls)
doc = inspect.getdoc(item)
source = inspect.getsource(item)
is_class = inspect.isclass(item)
item_type = "Class Name" if is_class else "Name"
input_content = (
"Class Name:"
f" {cls.__name__}\n\nDocumentation:\n{doc}\n\nSource"
f"{item_type}:"
f" {item.__name__}\n\nDocumentation:\n{doc}\n\nSource"
f" Code:\n{source}"
)
print(input_content)
# Process with OpenAI model (assuming the model's __call__ method takes this input and returns processed content)
# Process with OpenAI model
processed_content = model(
DOCUMENTATION_WRITER_SOP(input_content, "zeta")
DOCUMENTATION_WRITER_SOP(input_content, module)
)
doc_content = f"# {cls.__name__}\n\n{processed_content}\n"
doc_content = f"# {item.__name__}\n\n{processed_content}\n"
# Create the directory if it doesn't exist
dir_path = "docs/zeta/nn/modules"
dir_path = docs_folder_path
os.makedirs(dir_path, exist_ok=True)
# Write the processed documentation to a Markdown file
file_path = os.path.join(dir_path, f"{cls.__name__.lower()}.md")
# Write the processed documentation to a Python file
file_path = os.path.join(dir_path, f"{item.__name__.lower()}.md")
with open(file_path, "w") as file:
file.write(doc_content)
print(
f"Processed documentation for {item.__name__}. at {file_path}"
)
def main():
classes = [
DenseBlock,
HighwayLayer,
MultiScaleBlock,
FeedbackBlock,
DualPathBlock,
RecursiveBlock,
PytorchGELUTanh,
NewGELUActivation,
GELUActivation,
FastGELUActivation,
QuickGELUActivation,
ClippedGELUActivation,
AccurateGELUActivation,
MishActivation,
LinearActivation,
LaplaceActivation,
ReLUSquaredActivation,
items = [
Agent,
SequentialWorkflow,
AutoScaler,
Conversation,
TaskInput,
Artifact,
ArtifactUpload,
StepInput,
SwarmNetwork,
ModelParallelizer,
MultiAgentCollaboration,
AbstractSwarm,
GroupChat,
GroupChatManager,
parse_tasks,
find_agent_by_id,
distribute_tasks,
find_token_in_text,
extract_key_from_json,
extract_tokens_from_text,
ConcurrentWorkflow,
RecursiveWorkflow,
NonlinearWorkflow,
BaseWorkflow,
BaseStructure,
]
threads = []
for cls in classes:
for cls in items:
thread = threading.Thread(
target=process_documentation, args=(cls,)
)
@ -100,7 +129,7 @@ def main():
thread.join()
print(
"Documentation generated in 'docs/zeta/nn/modules' directory."
"Documentation generated in 'docs/swarms/structs' directory."
)

@ -0,0 +1,133 @@
import inspect
import os
import threading
from dotenv import load_dotenv
from scripts.auto_tests_docs.docs import DOCUMENTATION_WRITER_SOP
from swarms import OpenAIChat
from swarms.structs.agent import Agent
from swarms.structs.autoscaler import AutoScaler
from swarms.structs.base import BaseStructure
from swarms.structs.base_swarm import AbstractSwarm
from swarms.structs.base_workflow import BaseWorkflow
from swarms.structs.concurrent_workflow import ConcurrentWorkflow
from swarms.structs.conversation import Conversation
from swarms.structs.groupchat import GroupChat, GroupChatManager
from swarms.structs.model_parallizer import ModelParallelizer
from swarms.structs.multi_agent_collab import MultiAgentCollaboration
from swarms.structs.nonlinear_workflow import NonlinearWorkflow
from swarms.structs.recursive_workflow import RecursiveWorkflow
from swarms.structs.schemas import (
Artifact,
ArtifactUpload,
StepInput,
TaskInput,
)
from swarms.structs.sequential_workflow import SequentialWorkflow
from swarms.structs.swarm_net import SwarmNetwork
from swarms.structs.utils import (
distribute_tasks,
extract_key_from_json,
extract_tokens_from_text,
find_agent_by_id,
find_token_in_text,
parse_tasks,
)
load_dotenv()
api_key = os.getenv("OPENAI_API_KEY")
model = OpenAIChat(
model_name="gpt-4-1106-preview",
openai_api_key=api_key,
max_tokens=4000,
)
def process_documentation(
item,
module: str = "swarms.structs",
docs_folder_path: str = "docs/swarms/structs",
):
"""
Process the documentation for a given class or function using OpenAI model and save it in a Python file.
"""
doc = inspect.getdoc(item)
source = inspect.getsource(item)
is_class = inspect.isclass(item)
item_type = "Class Name" if is_class else "Name"
input_content = (
f"{item_type}:"
f" {item.__name__}\n\nDocumentation:\n{doc}\n\nSource"
f" Code:\n{source}"
)
# Process with OpenAI model
processed_content = model(
DOCUMENTATION_WRITER_SOP(input_content, module)
)
doc_content = f"# {item.__name__}\n\n{processed_content}\n"
# Create the directory if it doesn't exist
dir_path = docs_folder_path
os.makedirs(dir_path, exist_ok=True)
# Write the processed documentation to a Python file
file_path = os.path.join(dir_path, f"{item.__name__.lower()}.md")
with open(file_path, "w") as file:
file.write(doc_content)
print(
f"Processed documentation for {item.__name__}. at {file_path}"
)
def main(module: str = "docs/swarms/structs"):
items = [
Agent,
SequentialWorkflow,
AutoScaler,
Conversation,
TaskInput,
Artifact,
ArtifactUpload,
StepInput,
SwarmNetwork,
ModelParallelizer,
MultiAgentCollaboration,
AbstractSwarm,
GroupChat,
GroupChatManager,
parse_tasks,
find_agent_by_id,
distribute_tasks,
find_token_in_text,
extract_key_from_json,
extract_tokens_from_text,
ConcurrentWorkflow,
RecursiveWorkflow,
NonlinearWorkflow,
BaseWorkflow,
BaseStructure,
]
threads = []
for item in items:
thread = threading.Thread(
target=process_documentation, args=(item,)
)
threads.append(thread)
thread.start()
# Wait for all threads to complete
for thread in threads:
thread.join()
print(f"Documentation generated in {module} directory.")
if __name__ == "__main__":
main()

@ -5,9 +5,14 @@ from dotenv import load_dotenv
from swarms import (
OpenAIChat,
Conversation,
detect_markdown,
extract_code_from_markdown,
)
from swarms.tools.code_executor import CodeExecutor
conv = Conversation(
autosave=False,
time_enabled=True,
)
@ -18,7 +23,7 @@ load_dotenv()
api_key = os.environ.get("OPENAI_API_KEY")
# Initialize the language model
llm = OpenAIChat(openai_api_key=api_key, model_name="gpt-4")
llm = OpenAIChat(openai_api_key=api_key)
# Run the language model in a loop
@ -27,18 +32,33 @@ def interactive_conversation(llm, iters: int = 10):
for i in range(iters):
user_input = input("User: ")
conv.add("user", user_input)
if user_input.lower() == "quit":
break
task = (
conv.return_history_as_string()
) # Get the conversation history
# Run the language model
out = llm(task)
conv.add("assistant", out)
print(
f"Assistant: {out}",
)
# Code Interpreter
if detect_markdown(out):
code = extract_code_from_markdown(out)
if code:
print(f"Code: {code}")
executor = CodeExecutor()
out = executor.run(code)
conv.add("assistant", out)
# print(f"Assistant: {out}")
conv.display_conversation()
conv.export_conversation("conversation.txt")
# conv.export_conversation("conversation.txt")
# Replace with your LLM instance

@ -189,7 +189,9 @@ class BaseOpenAI(BaseLLM):
return True
client: Any = None #: :meta private:
model_name: str = Field(default="text-davinci-003", alias="model")
model_name: str = Field(
default="gpt-4-1106-preview", alias="model"
)
"""Model name to use."""
temperature: float = 0.7
"""What sampling temperature to use."""

@ -26,7 +26,10 @@ from swarms.structs.utils import (
find_agent_by_id,
find_token_in_text,
parse_tasks,
detect_markdown,
)
from swarms.structs.task import Task
from swarms.structs.block_wrapper import block
__all__ = [
"Agent",
@ -53,5 +56,8 @@ __all__ = [
"RecursiveWorkflow",
"NonlinearWorkflow",
"BaseWorkflow",
"BaseStructure"
"BaseStructure",
"detect_markdown",
"Task",
"block"
]

@ -0,0 +1,35 @@
from typing import Callable, Any
import logging
logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)
def block(
function: Callable[..., Any],
device: str = None,
verbose: bool = False,
) -> Callable[..., Any]:
"""
A decorator that transforms a function into a block.
Args:
function (Callable[..., Any]): The function to transform.
Returns:
Callable[..., Any]: The transformed function.
"""
def wrapper(*args, **kwargs):
# Here you can add code to execute the function on various hardwares
# For now, we'll just call the function normally
try:
return function(*args, **kwargs)
except Exception as error:
logger.error(f"Error in {function.__name__}: {error}")
raise error
# Set the wrapper function's name and docstring to those of the original function
wrapper.__name__ = function.__name__
wrapper.__doc__ = function.__doc__
return wrapper

@ -0,0 +1,19 @@
from tracemalloc import start
from typing import Any, Callable, Dict, List, Union
from dataclasses import dataclass, field
@dataclass
class BlockDevice:
device: str
cluster: str
description: str
def __init__(self, device: str, cluster: str, description: str):
self.device = device
self.cluster = cluster
self.description = description
def __str__(self):
return f"BlockDevice(device={self.device}, cluster={self.cluster}, description={self.description})"

@ -0,0 +1,65 @@
from typing import (
Any,
Dict,
Optional,
)
from swarms.structs.base import BaseStructure
class BlocksDict(BaseStructure):
"""
A class representing a dictionary of blocks.
Args:
name (str): The name of the blocks dictionary.
description (str): The description of the blocks dictionary.
blocks (Dict[str, Any]): The dictionary of blocks.
parent (Optional[Any], optional): The parent of the blocks dictionary. Defaults to None.
**kwargs: Additional keyword arguments.
Attributes:
parent (Optional[Any]): The parent of the blocks dictionary.
blocks (Dict[str, Any]): The dictionary of blocks.
Methods:
add(key: str, block: Any): Add a block to the dictionary.
remove(key: str): Remove a block from the dictionary.
get(key: str): Get a block from the dictionary.
"""
def __init__(
self,
name: str,
description: str,
blocks: Dict[str, Any],
parent: Optional[Any] = None,
**kwargs,
):
super().__init__(name=name, description=description, **kwargs)
self.parent = parent
self.blocks = blocks
def add(self, key: str, block: Any):
self.blocks[key] = block
def remove(self, key: str):
del self.blocks[key]
def get(self, key: str):
return self.blocks.get(key)
def update(self, key: str, block: Any):
self.blocks[key] = block
def keys(self):
return list(self.blocks.keys())
def values(self):
return list(self.blocks.values())
def items(self):
return list(self.blocks.items())
def clear(self):
self.blocks.clear()

@ -0,0 +1,138 @@
from typing import (
Any,
List,
Optional,
)
from swarms.structs.base import BaseStructure
class BlocksList(BaseStructure):
"""
A class representing a list of blocks.
Args:
name (str): The name of the blocks list.
description (str): The description of the blocks list.
blocks (List[Any]): The list of blocks.
parent (Optional[Any], optional): The parent of the blocks list. Defaults to None.
**kwargs: Additional keyword arguments.
Attributes:
parent (Optional[Any]): The parent of the blocks list.
blocks (List[Any]): The list of blocks.
Methods:
add(block: Any): Add a block to the list.
remove(block: Any): Remove a block from the list.
update(block: Any): Update a block in the list.
get(index: int): Get a block at the specified index.
get_all(): Get all blocks in the list.
get_by_name(name: str): Get blocks by name.
get_by_type(type: str): Get blocks by type.
get_by_id(id: str): Get blocks by ID.
get_by_parent(parent: str): Get blocks by parent.
get_by_parent_id(parent_id: str): Get blocks by parent ID.
get_by_parent_name(parent_name: str): Get blocks by parent name.
get_by_parent_type(parent_type: str): Get blocks by parent type.
get_by_parent_description(parent_description: str): Get blocks by parent description.
"""
def __init__(
self,
name: str,
description: str,
blocks: List[Any],
parent: Optional[Any] = None,
**kwargs,
):
super().__init__(name=name, description=description, **kwargs)
self.parent = parent
self.blocks = blocks
def add(self, block: Any):
self.blocks.append(block)
def remove(self, block: Any):
self.blocks.remove(block)
def update(self, block: Any):
self.blocks[self.blocks.index(block)] = block
def get(self, index: int):
return self.blocks[index]
def get_all(self):
return self.blocks
def get_by_name(self, name: str):
return [block for block in self.blocks if block.name == name]
def get_by_type(self, type: str):
return [block for block in self.blocks if block.type == type]
def get_by_id(self, id: str):
return [block for block in self.blocks if block.id == id]
def get_by_parent(self, parent: str):
return [
block for block in self.blocks if block.parent == parent
]
def get_by_parent_id(self, parent_id: str):
return [
block
for block in self.blocks
if block.parent_id == parent_id
]
def get_by_parent_name(self, parent_name: str):
return [
block
for block in self.blocks
if block.parent_name == parent_name
]
def get_by_parent_type(self, parent_type: str):
return [
block
for block in self.blocks
if block.parent_type == parent_type
]
def get_by_parent_description(self, parent_description: str):
return [
block
for block in self.blocks
if block.parent_description == parent_description
]
def __len__(self):
return len(self.blocks)
def __getitem__(self, index):
return self.blocks[index]
def __setitem__(self, index, value):
self.blocks[index] = value
def __delitem__(self, index):
del self.blocks[index]
def __iter__(self):
return iter(self.blocks)
def __reversed__(self):
return reversed(self.blocks)
def __contains__(self, item):
return item in self.blocks
def __str__(self):
return f"{self.name}({self.blocks})"
def __repr__(self):
return f"{self.name}({self.blocks})"
def __eq__(self, other):
return self.blocks == other.blocks

@ -31,7 +31,7 @@ class Conversation(BaseStructure):
time_enabled: bool = False,
database: AbstractDatabase = None,
autosave: bool = True,
save_filepath: str = "/runs/conversation.json",
save_filepath: str = "runs/conversation.json",
*args,
**kwargs,
):

@ -1,3 +1,4 @@
import re
import json
from typing import Dict, Any, List, Optional
from swarms.structs.agent import Agent
@ -118,3 +119,21 @@ def extract_tokens_from_text(
List[str]: The tokens that were found in the text.
"""
return [token for token in tokens if token in text]
def detect_markdown(text: str) -> bool:
"""
Checks if a string contains Markdown code enclosed in six backticks.
Parameters
----------
text : str
The text to check.
Returns
-------
bool
True if the text contains Markdown code enclosed in six backticks, False otherwise.
"""
pattern = r"``````[\s\S]*?``````"
return bool(re.search(pattern, text))

@ -1,24 +1,31 @@
""" This module lists all the telemetry related functions. """
from swarms.telemetry.log_all import log_all_calls, log_calls
# from swarms.telemetry.posthog_utils import log_activity_posthog
from swarms.telemetry.sys_info import (
get_cpu_info,
get_oi_version,
get_os_version,
get_package_mismatches,
get_pip_version,
get_python_version,
get_ram_info,
interpreter_info,
system_info,
)
from swarms.telemetry.user_utils import (
generate_unique_identifier,
generate_user_id,
get_machine_id,
get_system_info,
generate_unique_identifier,
)
__all__ = [
"log_all_calls",
"log_calls",
# "log_activity_posthog",
"generate_user_id",
"get_machine_id",
"get_system_info",
"generate_unique_identifier",
"get_python_version", # from swarms/telemetry/sys_info.py
"get_python_version",
"get_pip_version",
"get_oi_version",
"get_os_version",

@ -1,3 +1,23 @@
from swarms.tools.tool_func_doc_scraper import scrape_tool_func_docs
from swarms.tools.code_executor import CodeExecutor
from swarms.tools.tool_utils import (
tool_find_by_name,
extract_tool_commands,
parse_and_execute_tools,
execute_tools,
)
from swarms.tools.tool import BaseTool, Tool, StructuredTool, tool
__all__ = ["scrape_tool_func_docs"]
__all__ = [
"scrape_tool_func_docs",
"CodeExecutor",
"tool_find_by_name",
"extract_tool_commands",
"parse_and_execute_tools",
"execute_tools",
"BaseTool",
"Tool",
"StructuredTool",
"tool",
]

@ -0,0 +1,111 @@
import os
import tempfile
import subprocess
class CodeExecutor:
"""
A class for executing code snippets.
Args:
code (str, optional): The code snippet to be executed. Defaults to None.
Methods:
is_python_code(code: str = None) -> bool:
Checks if the given code is Python code.
run_python(code: str = None) -> str:
Executes the given Python code and returns the output.
run(code: str = None) -> str:
Executes the given code and returns the output.
__call__() -> str:
Executes the code and returns the output.
"""
def __init__(self, code: str = None):
self.code = code
def is_python_code(self, code: str = None) -> bool:
"""
Checks if the given code is Python code.
Args:
code (str, optional): The code to be checked. Defaults to None.
Returns:
bool: True if the code is Python code, False otherwise.
"""
code = code or self.code
return code.strip().startswith("python")
def run_python(self, code: str = None) -> str:
"""
Executes the given Python code and returns the output.
Args:
code (str, optional): The Python code to be executed. Defaults to None.
Returns:
str: The output of the code execution.
"""
code = code or self.code
try:
# Create a temporary file
with tempfile.NamedTemporaryFile(
suffix=".py", delete=False
) as temp:
temp.write(code.encode())
temp_filename = temp.name
# Execute the temporary file
output = subprocess.check_output(
f"python {temp_filename}",
shell=True,
)
# Delete the temporary file
os.remove(temp_filename)
return output.decode("utf-8")
except subprocess.CalledProcessError as error:
return error.output.decode("utf-8")
except Exception as error:
return str(error)
def run(self, code: str = None) -> str:
"""
Executes the given code and returns the output.
Args:
code (str, optional): The code to be executed. Defaults to None.
Returns:
str: The output of the code execution.
"""
code = code or self.code
try:
output = subprocess.check_output(
code,
shell=True,
)
return output.decode("utf-8")
except subprocess.CalledProcessError as e:
return e.output.decode("utf-8")
except Exception as e:
return str(e)
def __call__(self) -> str:
"""
Executes the code and returns the output.
Returns:
str: The output of the code execution.
"""
return self.run()
# model = CodeExecutor()
# out = model.run("python3")
# print(out)

@ -1,8 +1,17 @@
import re
import json
from typing import List, Any
def extract_tool_commands(self, text: str):
def tool_find_by_name(tool_name: str, tools: List[Any]):
"""Find the tool by name"""
for tool in tools:
if tool.name == tool_name:
return tool
return None
def extract_tool_commands(text: str):
"""
Extract the tool commands from the text
@ -39,9 +48,9 @@ def parse_and_execute_tools(response: str):
execute_tools(tool_name, params)
def execute_tools(self, tool_name, params):
def execute_tools(tool_name, params):
"""Execute the tool with the provided params"""
tool = self.tool_find_by_name(tool_name)
tool = tool_find_by_name(tool_name)
if tool:
# Execute the tool with the provided parameters
tool_result = tool.run(**params)

Loading…
Cancel
Save