add python package claude is supported in chatmol packge (e240a407) · Commits · github_fork / ChatMol

.gitignore

+2 −0

Original line number	Diff line number	Diff line
		@@ -128,3 +128,5 @@ dmypy.json

		# Pyre type checker
		.pyre/

		test*

README.md

+10 −1

Original line number	Diff line number	Diff line
		@@ -5,6 +5,7 @@ Started from a PyMol plugin.
		- [ChatMol](#chatmol)
		- [Table of contents](#table-of-contents)
		- [Overview](#overview)
		- [ChatMol python package](#chatmol-python-package)
		- [Copilot](#copilot)
		- [ChatMol Website](#chatmol-website)
		- [Requirements \& Installation](#requirements--installation)
		@@ -18,7 +19,15 @@ Started from a PyMol plugin.
		- [License](#license)

		## Overview
		The PyMOL ChatGPT Plugin seamlessly integrates OpenAI's large language models (GPT-3.5-turbo and text-davinci-003) into PyMOL, enabling users to interact with PyMOL using natural language instructions. This robust tool simplifies PyMOL tasks and offers suggestions, explanations, and guidance on a wide range of PyMOL-related topics. ChatMol provides various interaction modes with PyMOL, including the PyMOL command line, miniGUI chatbot, and web browsers.
		The PyMOL ChatGPT Plugin seamlessly integrates LLMs into PyMOL, enabling users to interact with PyMOL using natural language instructions. This tool simplifies PyMOL tasks and offers suggestions, explanations, and guidance on a wide range of PyMOL-related topics. ChatMol provides various interaction modes with PyMOL, including the PyMOL command line, miniGUI chatbot, python and web browsers.

		## ChatMol python package

		See this [README](./chatmol_pkg/README.md) for more details.

		```bash
		pip install chatmol
		```

		## Copilot
		This is ChatMol copilot, just like other copilot, it is designed to help your work.

chatmol_pkg/README.md

0 → 100644

+44 −0

Original line number	Diff line number	Diff line
		# ChatMol Python Package

		ChatMol is a Python package that provides a seamless integration of large language models into PyMOL, enabling users to interact with PyMOL using natural language instructions. This robust tool simplifies PyMOL tasks and offers suggestions, explanations, and guidance on a wide range of PyMOL-related topics. ChatMol provides various interaction modes with PyMOL, including the PyMOL command line, Python, miniGUI chatbot, and web browsers.

		## Installation

		```bash
		pip install chatmol
		```

		## Usage

		Here are some examples of how to use the package:

		```python
		import chatmol as cm
		output_chatmol_llm = cm.chatlite("download chain A of 3wzm and color it by secondary structure") # use the chatmol llm, free and no API key required
		print(output_chatmol_llm)
		```

		```python
		print(cm.defaul_client.gpt_model) # check the current ChatGPT model
		output_chatgpt = cm.chat_with_gpt("download 4eb0 and highlight residue number 208") # use the GPT-3.5-turbo llm, API key required
		print(output_chatgpt)
		```

		```python
		print(cm.defaul_client.claude_model) # check the current Claude model
		output_claude = cm.chat_with_claude("download 1pga from rcsb and show a transprant surface") # use the claude llm, API key required
		print(output_claude)
		```

		You can send results to PyMOL:

		```python
		import chatmol as cm
		ps = cm.start_pymol() # open a PyMOL session with XML-RPC server
		ps.chatlite("download 1pga")

		# send commands to PyMOL
		ps.server.do("esmfold MTYKLILNGKTLKGETTTEAVDAATAEKVFKQYANDNGVDGEWTYDDATKTFTVTE, 1pga_esmfold") # make sure you have pymolfold plugin installed
		```

		enjoy!
		No newline at end of file

chatmol_pkg/chatmol/init.py

0 → 100644

+24 −0

Original line number	Diff line number	Diff line
		from .utils import ChatMol
		from .pymol_server import PymolServer

		defaul_client = ChatMol()

		def chatlite(question):
		return defaul_client.chatlite(question)

		def chat_with_gpt(message):
		return defaul_client.chat_with_gpt(message)

		def chat_with_claude(message):
		return defaul_client.chat_with_claude(message)

		def clear_stashed_commands():
		return defaul_client.clear_stashed_commands()

		def clear_chat_history():
		return defaul_client.clear_chat_history()

		def start_pymol():
		pymolserver = PymolServer(defaul_client)
		return pymolserver

chatmol_pkg/chatmol/pymol_server.py

0 → 100644

+114 −0

Original line number	Diff line number	Diff line
		import subprocess
		from xmlrpc import client
		from .utils import ChatMol

		class PymolServer():
		def __init__(self, default_client:ChatMol):
		self.cm = default_client
		self.start_pymol()

		def start_pymol(self):
		# os.system("nohup pymol -R /dev/null 2>&1")
		subprocess.Popen(["pymol", "-R"], stdout=subprocess.DEVNULL, stderr=subprocess.DEVNULL)
		self.server = client.ServerProxy(uri="http://localhost:9123/RPC2")
		return 0

		def chatlite(self, question):
		answer = self.cm.chatlite(question)
		commands = answer.split('\n')
		print("Answers from ChatMol-Lite: ")
		for command in commands:
		if command == '':
		continue
		else:
		print(command)
		self.server.do(command)
		return commands

		def chatgpt(self, message, execute:bool=True, lite:bool=False):
		if lite:
		self.cm.chatlite(message)
		message = message.strip()
		if message == "e" or message == 'execute':
		if len(self.cm.stashed_commands) == 0:
		print("There is no stashed commands")
		else:
		for command in stashed_commands:
		self.server.do(command)
		self.cm.clear_stashed_commands()
		return 0

		if message == "new":
		self.cm.clear_chat_history()
		self.cm.clear_stashed_commands()
		return 0

		if message.endswith('?'):
		execute = False

		response = self.cm.chat_with_gpt(message) # Using the chat_with_gpt method
		print("ChatGPT:", response)

		try:
		command_blocks = []
		self.cm.clear_stashed_commands()
		for i, block in enumerate(response.split("```")):
		if i % 2 == 1:
		command_blocks.append(block)
		for command_block in command_blocks:
		for command in command_block.split("\n"):
		if command.strip() and not command.strip().startswith("#"):
		if command.strip() == "python" or command.strip() == "bash":
		continue # Skipping python commands
		if "#" in command:
		command, comment = command.split("#")
		if execute:
		print(command)
		self.server.do(command)
		else:
		self.cm.stashed_commands.append(command)
		except Exception as e:
		print(f"Error during command execution: {e}")

		def claude(self, message, execute:bool=True):
		message = message.strip()
		if message == "e" or message == 'execute':
		if len(self.cm.stashed_commands) == 0:
		print("There is no stashed commands")
		else:
		for command in stashed_commands:
		self.server.do(command)
		self.cm.clear_stashed_commands()
		return 0

		if message == "new":
		self.cm.clear_chat_history()
		self.cm.clear_stashed_commands()
		return 0

		if message.endswith('?'):
		execute = False

		response = self.cm.chat_with_claude(message) # Using the chat_with_gpt method
		print(self.cm.claude_model, response)

		try:
		command_blocks = []
		self.cm.clear_stashed_commands()
		for i, block in enumerate(response.split("```")):
		if i % 2 == 1:
		command_blocks.append(block)
		for command_block in command_blocks:
		for command in command_block.split("\n"):
		if command.strip() and not command.strip().startswith("#"):
		if command.strip() == "python" or command.strip() == "bash":
		continue # Skipping python commands
		if "#" in command:
		command, comment = command.split("#")
		if execute:
		print(command)
		self.server.do(command)
		else:
		self.cm.stashed_commands.append(command)
		except Exception as e:
		print(f"Error during command execution: {e}")
		No newline at end of file

Admin message