Jelajahi Sumber

Add Cohere LLM support (#751)

Sidharth Mohanty 1 tahun lalu
induk
melakukan
03a84daf9d
4 mengubah file dengan 79 tambahan dan 0 penghapusan
  1. 1 0
      docs/advanced/app_types.mdx
  2. 43 0
      embedchain/llm/cohere.py
  3. 2 0
      pyproject.toml
  4. 33 0
      tests/llm/test_cohere.py

+ 1 - 0
docs/advanced/app_types.mdx

@@ -27,6 +27,7 @@ The following LLM providers are supported by Embedchain:
 - GPT4ALL
 - AZURE_OPENAI
 - LLAMA2
+- COHERE
 
 You can choose one by importing it from `embedchain.llm`. E.g.:
 

+ 43 - 0
embedchain/llm/cohere.py

@@ -0,0 +1,43 @@
+import importlib
+import os
+from typing import Optional
+
+from langchain.llms import Cohere
+
+from embedchain.config import BaseLlmConfig
+from embedchain.helper.json_serializable import register_deserializable
+from embedchain.llm.base import BaseLlm
+
+
+@register_deserializable
+class CohereLlm(BaseLlm):
+    def __init__(self, config: Optional[BaseLlmConfig] = None):
+        if "COHERE_API_KEY" not in os.environ:
+            raise ValueError("Please set the COHERE_API_KEY environment variable.")
+
+        try:
+            importlib.import_module("cohere")
+        except ModuleNotFoundError:
+            raise ModuleNotFoundError(
+                "The required dependencies for Cohere are not installed."
+                'Please install with `pip install --upgrade "embedchain[cohere]"`'
+            ) from None
+
+        super().__init__(config=config)
+
+    def get_llm_model_answer(self, prompt):
+        if self.config.system_prompt:
+            raise ValueError("CohereLlm does not support `system_prompt`")
+        return CohereLlm._get_answer(prompt=prompt, config=self.config)
+
+    @staticmethod
+    def _get_answer(prompt: str, config: BaseLlmConfig) -> str:
+        llm = Cohere(
+            cohere_api_key=os.environ["COHERE_API_KEY"],
+            model=config.model,
+            max_tokens=config.max_tokens,
+            temperature=config.temperature,
+            p=config.top_p,
+        )
+
+        return llm(prompt)

+ 2 - 0
pyproject.toml

@@ -105,6 +105,7 @@ twilio = { version = "^8.5.0", optional = true }
 fastapi-poe = { version = "0.0.16", optional = true }
 discord = { version = "^2.3.2", optional = true }
 slack-sdk = { version = "3.21.3", optional = true }
+cohere = { version = "^4.27", optional= true }
 docx2txt = "^0.8"
 unstructured = {extras = ["local-inference"], version = "^0.10.18"}
 pillow = { version = "10.0.1", optional = true }
@@ -134,6 +135,7 @@ discord = ["discord"]
 slack = ["slack-sdk", "flask"]
 whatsapp = ["twilio", "flask"]
 images = ["torch", "ftfy", "regex", "pillow", "torchvision"]
+cohere = ["cohere"]
 
 [tool.poetry.group.docs.dependencies]
 

+ 33 - 0
tests/llm/test_cohere.py

@@ -0,0 +1,33 @@
+import os
+import unittest
+from unittest.mock import patch
+
+from embedchain.config import BaseLlmConfig
+from embedchain.llm.cohere import CohereLlm
+
+
+class TestCohereLlm(unittest.TestCase):
+    def setUp(self):
+        os.environ["COHERE_API_KEY"] = "test_api_key"
+        self.config = BaseLlmConfig(model="gptd-instruct-tft", max_tokens=50, temperature=0.7, top_p=0.8)
+
+    def test_init_raises_value_error_without_api_key(self):
+        os.environ.pop("COHERE_API_KEY")
+        with self.assertRaises(ValueError):
+            CohereLlm()
+
+    def test_get_llm_model_answer_raises_value_error_for_system_prompt(self):
+        llm = CohereLlm(self.config)
+        llm.config.system_prompt = "system_prompt"
+        with self.assertRaises(ValueError):
+            llm.get_llm_model_answer("prompt")
+
+    @patch("embedchain.llm.cohere.CohereLlm._get_answer")
+    def test_get_llm_model_answer(self, mock_get_answer):
+        mock_get_answer.return_value = "Test answer"
+
+        llm = CohereLlm(self.config)
+        answer = llm.get_llm_model_answer("Test query")
+
+        self.assertEqual(answer, "Test answer")
+        mock_get_answer.assert_called_once()