renaming MDCrow

CONTRIBUTING.md

@@ -9,8 +9,8 @@ To get started with contributing to md-agent, follow these steps:
 ```
 git clone https://github.com/ur-whitelab/md-agent.git
 cd md-agent
-conda env create -n mdagent -f environment.yaml
-conda activate mdagent
+conda env create -n mdcrow -f environment.yaml
+conda activate mdcrow
 ```
 
 ### Installing the Package and Dependencies and Configuring Pre-commit Hooks

README.md

@@ -1,12 +1,12 @@
-MDAgent is a LLM-agent based toolset for Molecular Dynamics.
+MDCrow is a LLM-agent based toolset for Molecular Dynamics.
 It's built using Langchain and uses a collection of tools to set up and execute molecular dynamics simulations, particularly in OpenMM.
 
 
 ## Environment Setup
 To use the OpenMM features in the agent, please set up a conda environment, following these steps.
 ```
-conda env create -n mdagent -f environment.yaml
-conda activate mdagent
+conda env create -n mdcrow -f environment.yaml
+conda activate mdcrow
 ```
 
 If you already have a conda environment, you can install dependencies before you activate it with the following step.
@@ -25,14 +25,14 @@ Other tools require API keys, such as paper-qa for literature searches. We recom
 1. Copy the `.env.example` file and rename it to `.env`: `cp .env.example .env`
 2. Replace the placeholder values in `.env` with your actual keys
 
-You can ask MDAgent to conduct molecular dynamics tasks using OpenAI's GPT model
+You can ask MDCrow to conduct molecular dynamics tasks using OpenAI's GPT model
 ```
-from mdagent import MDAgent
+from mdcrow import MDCrow
 
-agent = MDAgent(model="gpt-3.5-turbo")
+agent = MDCrow(model="gpt-3.5-turbo")
 agent.run("Simulate protein 1ZNI at 300 K for 0.1 ps and calculate the RMSD over time.")
 ```
-Note: to distinguish Together models from the rest, you'll need to add "together\" prefix in model flag, such as `agent = MDAgent(model="together/meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo")`
+Note: to distinguish Together models from the rest, you'll need to add "together\" prefix in model flag, such as `agent = MDCrow(model="together/meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo")`
 
 ## LLM Providers
 By default, we support LLMs through OpenAI API. However, feel free to use other LLM providers. Make sure to install the necessary package for it. Here's list of packages required for alternative LLM providers we support:
@@ -43,6 +43,6 @@ By default, we support LLMs through OpenAI API. However, feel free to use other
 
 ## Contributing
 
-We welcome contributions to MDAgent! If you're interested in contributing to the project, please check out our [Contributor's Guide](CONTRIBUTING.md) for detailed instructions on getting started, feature development, and the pull request process.
+We welcome contributions to MDCrow! If you're interested in contributing to the project, please check out our [Contributor's Guide](CONTRIBUTING.md) for detailed instructions on getting started, feature development, and the pull request process.
 
-We value and appreciate all contributions to MDAgent.
+We value and appreciate all contributions to MDCrow.

mdcrow/__init__.py

@@ -0,0 +1,3 @@
+from .agent import Evaluator, MDCrow
+
+__all__ = ["MDCrow", "Evaluator"]

mdagent/agent/README_evaluate.md → mdcrow/agent/README_evaluate.md

@@ -1,17 +1,17 @@
 # README for Evaluate.py
 
 ## Overview
-`evaluate.py` is a Python script designed to facilitate the evaluation of MDAgent. It supports multiple use cases, including automated evaluation across different agent settings, loading previous evaluation results, and creating structured output tables. A list of generated or loaded evaluations can be accessed via `evaluator.evaluations`. Use `evaluator.reset()` to clear the evaluations list.
+`evaluate.py` is a Python script designed to facilitate the evaluation of MDCrow. It supports multiple use cases, including automated evaluation across different agent settings, loading previous evaluation results, and creating structured output tables. A list of generated or loaded evaluations can be accessed via `evaluator.evaluations`. Use `evaluator.reset()` to clear the evaluations list.
 
 ## Getting Started
-To use `evaluate.py`, ensure that `mdagent` package is installed in your Python environment (see main README for installation instructions).
+To use `evaluate.py`, ensure that `mdcrow` package is installed in your Python environment (see main README for installation instructions).
 
 ## Usage Examples
 
 ### Example 1: Evaluate Prompts with default MD-Agent Parameters
 Evaluate specific prompts with default settings:
 ```python
-from mdagent import Evaluator
+from mdcrow import Evaluator
 
 evaluator = Evaluator()
 prompts = [
@@ -21,13 +21,13 @@ prompts = [
 df = evaluator.automate(prompts)
 df # this displays DataFrame table in Jupyter notebook
 ```
-This will run MDAgent and evaluate the prompts using the default settings. The results will be
+This will run MDCrow and evaluate the prompts using the default settings. The results will be
 saved to json file in the "evaluation_results" directory and used to create pandas DataFrame.
 
 ### Example 2: Evaluate Prompts with specific MD-Agent Parameters
 Evaluate specific prompts using single agent instance with specified parameters:
 ```python
-from mdagent import Evaluator
+from mdcrow import Evaluator
 
 evaluator = Evaluator()
 agent_params = {
@@ -46,7 +46,7 @@ df_full = evaluator.create_table(simple=False) # to get a table with all details
 ### Example 3: Evaluate Prompts with Multiple Agent Parameters
 Evaluate specific prompts using multiple agent settings with `automate_all` method:
 ```python
-from mdagent import Evaluator
+from mdcrow import Evaluator
 
 evaluator = Evaluator()
 prompts = [
@@ -71,7 +71,7 @@ df = evaluator.automate_all(prompts, agent_params_list=agent_params_list)
 ### Example 4: Load Previous Evaluation Results and Create a Table
 Load previous evaluation results from a JSON file:
 ```python
-from mdagent import Evaluator
+from mdcrow import Evaluator
 
 evaluator = Evaluator()
 evaluator.load('evaluation_results/mega_eval_20240422-181241.json')
@@ -83,7 +83,7 @@ You can load multiple evaluation files by calling `evaluator.load()` multiple ti
 ### Example 5: Make Multi-Prompt Query with Agent Memory
 Use agent memory to link multiple prompts:
 ```python
-from mdagent import Evaluator
+from mdcrow import Evaluator
 
 evaluator = Evaluator()
 agent_params1 = {

mdcrow/agent/__init__.py

@@ -0,0 +1,4 @@
+from .agent import MDCrow
+from .evaluate import Evaluator
+
+__all__ = ["MDCrow", "Evaluator"]

mdagent/agent/agent.py → mdcrow/agent/agent.py

@@ -30,7 +30,7 @@ def get_agent(cls, model_name: str = "OpenAIFunctionsAgent"):
             )
 
 
-class MDAgent:
+class MDCrow:
     def __init__(
         self,
         tools=None,

mdagent/agent/evaluate.py → mdcrow/agent/evaluate.py

@@ -4,7 +4,7 @@
 
 import pandas as pd
 
-from .agent import MDAgent
+from .agent import MDCrow
 
 
 # TODO: turn off verbose for MD-Agent -- verbose option doesn't work
@@ -27,18 +27,18 @@ def __init__(self, eval_dir="evaluation_results"):
 
     def create_agent(self, agent_params={}):
         """
-        initializes MDAgent with given parameters
+        initializes MDCrow with given parameters
 
         Parameters:
-        - agent_params (dict): dictionary of parameters to initialize MDAgent
+        - agent_params (dict): dictionary of parameters to initialize MDCrow
 
         Returns:
-        - initialized MDAgent object
+        - initialized MDCrow object
         """
-        # initialize MDAgent with given parameters.
+        # initialize MDCrow with given parameters.
         if agent_params is None:  # this shouldn't happen though
             agent_params = {}
-        return MDAgent(**agent_params)
+        return MDCrow(**agent_params)
 
     def reset(self):
         """
@@ -121,7 +121,7 @@ def _flatten_dict(self, d, sep="_"):
     def _evaluate_all_steps(self, agent, user_prompt):
         """
         core function that evaluates while iterating every step of
-        MDAgent's response to a user prompt. Evaluation details are
+        MDCrow's response to a user prompt. Evaluation details are
         saved to json file.
         Used in run_and_evaluate method.
 
@@ -130,7 +130,7 @@ def _evaluate_all_steps(self, agent, user_prompt):
         which can be used to create a table.
 
         Parameters:
-        - agent (MDAgent): MDAgent object
+        - agent (MDCrow): MDCrow object
         - user_prompt (str): user prompt to evaluate
 
         Returns:
@@ -222,7 +222,7 @@ def run_and_evaluate(self, prompts, agent_params={}):
 
         Parameters:
         - prompts (list): list of prompts to evaluate
-        - agent_params (dict): dictionary of parameters to initialize MDAgent
+        - agent_params (dict): dictionary of parameters to initialize MDCrow
 
         Returns:
         - None
@@ -248,7 +248,7 @@ def run_and_evaluate(self, prompts, agent_params={}):
                         "error_msg": f"{type(e).__name__}: {e}",
                     }
                 )
-                print(f"Error occurred while running MDAgent. {type(e).__name__}: {e}")
+                print(f"Error occurred while running MDCrow. {type(e).__name__}: {e}")
 
     def create_table(self, simple=True):
         """
@@ -291,7 +291,7 @@ def automate(self, prompts, agent_params={}):
 
         Parameters:
         - prompts (list): list of prompts to evaluate
-        - agent_params (dict): dictionary of parameters to initialize MDAgent
+        - agent_params (dict): dictionary of parameters to initialize MDCrow
 
         Returns:
         - DataFrame table of evaluations
@@ -310,7 +310,7 @@ def automate_all(self, prompts, agent_params_list=None):
         Parameters:
         - prompts (list): list of prompts to evaluate
         - agent_params_list (list): list of dictionaries containing parameters
-            to initialize MDAgent. If None, it will evaluate with default agents.
+            to initialize MDCrow. If None, it will evaluate with default agents.
 
         Returns:
         - DataFrame table of evaluations

mdagent/agent/memory.py → mdcrow/agent/memory.py

@@ -6,7 +6,7 @@
 from langchain.prompts import PromptTemplate
 from langchain_core.output_parsers import StrOutputParser
 
-from mdagent.utils import PathRegistry
+from mdcrow.utils import PathRegistry
 
 agent_summary_template = PromptTemplate(
     input_variables=["agent_trace"],

mdagent/tools/base_tools/analysis_tools/distance_tools.py → mdcrow/tools/base_tools/analysis_tools/distance_tools.py

@@ -10,7 +10,7 @@
 from matplotlib.animation import FuncAnimation
 from pydantic import BaseModel, Field
 
-from mdagent.utils import FileType, PathRegistry, load_single_traj
+from mdcrow.utils import FileType, PathRegistry, load_single_traj
 
 
 class DistanceToolsUtils:

mdagent/tools/base_tools/analysis_tools/hydrogen_bonding_tools.py → mdcrow/tools/base_tools/analysis_tools/hydrogen_bonding_tools.py

@@ -2,7 +2,7 @@
 import mdtraj as md
 from langchain.tools import BaseTool
 
-from mdagent.utils import FileType, PathRegistry, load_single_traj
+from mdcrow.utils import FileType, PathRegistry, load_single_traj
 
 
 class HydrogenBondTool(BaseTool):

mdagent/tools/base_tools/analysis_tools/inertia.py → mdcrow/tools/base_tools/analysis_tools/inertia.py

@@ -5,7 +5,7 @@
 import numpy as np
 from langchain.tools import BaseTool
 
-from mdagent.utils import FileType, PathRegistry, load_single_traj, save_to_csv
+from mdcrow.utils import FileType, PathRegistry, load_single_traj, save_to_csv
 
 
 class MOIFunctions:

mdagent/tools/base_tools/analysis_tools/pca_tools.py → mdcrow/tools/base_tools/analysis_tools/pca_tools.py

@@ -10,7 +10,7 @@
 from pydantic import BaseModel, Field
 from sklearn.decomposition import PCA
 
-from mdagent.utils import FileType, PathRegistry, load_single_traj
+from mdcrow.utils import FileType, PathRegistry, load_single_traj
 
 
 class PCA_analysis:

mdagent/tools/base_tools/analysis_tools/plot_tools.py → mdcrow/tools/base_tools/analysis_tools/plot_tools.py

@@ -5,7 +5,7 @@
 import matplotlib.pyplot as plt
 from langchain.tools import BaseTool
 
-from mdagent.utils import FileType, PathRegistry
+from mdcrow.utils import FileType, PathRegistry
 
 
 class PlottingTools:

mdagent/tools/base_tools/analysis_tools/ppi_tools.py → mdcrow/tools/base_tools/analysis_tools/ppi_tools.py

@@ -6,7 +6,7 @@
 from langchain.tools import BaseTool
 from pydantic import BaseModel, Field
 
-from mdagent.utils import PathRegistry
+from mdcrow.utils import PathRegistry
 
 
 def ppi_distance(file_path, binding_site="protein"):

mdagent/tools/base_tools/analysis_tools/rdf_tool.py → mdcrow/tools/base_tools/analysis_tools/rdf_tool.py

@@ -5,7 +5,7 @@
 from langchain.tools import BaseTool
 from pydantic import BaseModel, Field
 
-from mdagent.utils import FileType, PathRegistry
+from mdcrow.utils import FileType, PathRegistry
 
 
 class RDFToolInput(BaseModel):

mdagent/tools/base_tools/analysis_tools/rgy.py → mdcrow/tools/base_tools/analysis_tools/rgy.py

@@ -5,7 +5,7 @@
 import numpy as np
 from langchain.tools import BaseTool
 
-from mdagent.utils import FileType, PathRegistry, load_single_traj
+from mdcrow.utils import FileType, PathRegistry, load_single_traj
 
 
 class RadiusofGyration:

mdagent/tools/base_tools/analysis_tools/rmsd_tools.py → mdcrow/tools/base_tools/analysis_tools/rmsd_tools.py

@@ -5,7 +5,7 @@
 from langchain.tools import BaseTool
 from pydantic import BaseModel, Field
 
-from mdagent.utils import PathRegistry, load_traj_with_ref, save_plot, save_to_csv
+from mdcrow.utils import PathRegistry, load_traj_with_ref, save_plot, save_to_csv
 
 
 def rmsd(path_registry, traj, ref_traj, mol_name, select="protein"):

mdagent/tools/base_tools/analysis_tools/sasa.py → mdcrow/tools/base_tools/analysis_tools/sasa.py

@@ -5,7 +5,7 @@
 import numpy as np
 from langchain.tools import BaseTool
 
-from mdagent.utils import FileType, PathRegistry, load_single_traj, save_to_csv
+from mdcrow.utils import FileType, PathRegistry, load_single_traj, save_to_csv
 
 
 class SASAFunctions:

mdagent/tools/base_tools/analysis_tools/secondary_structure.py → mdcrow/tools/base_tools/analysis_tools/secondary_structure.py

@@ -5,7 +5,7 @@
 import numpy as np
 from langchain.tools import BaseTool
 
-from mdagent.utils import FileType, PathRegistry, load_single_traj
+from mdcrow.utils import FileType, PathRegistry, load_single_traj
 
 
 def write_raw_x(

mdagent/tools/base_tools/analysis_tools/vis_tools.py → mdcrow/tools/base_tools/analysis_tools/vis_tools.py

@@ -6,7 +6,7 @@
 import nbformat as nbf
 from langchain.tools import BaseTool
 
-from mdagent.utils import PathRegistry
+from mdcrow.utils import PathRegistry
 
 
 class VisFunctions:

mdagent/tools/base_tools/preprocess_tools/clean_tools.py → mdcrow/tools/base_tools/preprocess_tools/clean_tools.py

@@ -5,7 +5,7 @@
 from pdbfixer import PDBFixer
 from pydantic import BaseModel, Field
 
-from mdagent.utils import FileType, PathRegistry
+from mdcrow.utils import FileType, PathRegistry
 
 
 class CleaningToolFunctionInput(BaseModel):

mdagent/tools/base_tools/preprocess_tools/packing.py → mdcrow/tools/base_tools/preprocess_tools/packing.py

@@ -7,7 +7,7 @@
 from langchain.tools import BaseTool
 from pydantic import BaseModel, Field, ValidationError
 
-from mdagent.utils import PathRegistry
+from mdcrow.utils import PathRegistry
 
 from .pdb_fix import Validate_Fix_PDB
 from .pdb_get import MolPDB

mdagent/tools/base_tools/preprocess_tools/pdb_fix.py → mdcrow/tools/base_tools/preprocess_tools/pdb_fix.py

@@ -8,7 +8,7 @@
 from pdbfixer import PDBFixer
 from pydantic import BaseModel, Field, ValidationError
 
-from mdagent.utils import PathRegistry
+from mdcrow.utils import PathRegistry
 
 from .elements import list_of_elements
 

mdagent/tools/base_tools/preprocess_tools/pdb_get.py → mdcrow/tools/base_tools/preprocess_tools/pdb_get.py

@@ -5,7 +5,7 @@
 from rdkit import Chem
 from rdkit.Chem import AllChem
 
-from mdagent.utils import FileType, PathRegistry
+from mdcrow.utils import FileType, PathRegistry
 
 
 def get_pdb(query_string: str, path_registry: PathRegistry):

mdagent/tools/base_tools/simulation_tools/create_simulation.py → mdcrow/tools/base_tools/simulation_tools/create_simulation.py

@@ -7,7 +7,7 @@
 from langchain_core.output_parsers import StrOutputParser
 from pydantic import BaseModel, Field
 
-from mdagent.utils import FileType, PathRegistry
+from mdcrow.utils import FileType, PathRegistry
 
 
 class ModifyScriptUtils: