91 lines
3.3 KiB
Python
91 lines
3.3 KiB
Python
import requests
|
|
import logging
|
|
from typing import Dict, Any
|
|
|
|
logger = logging.getLogger(__name__)
|
|
|
|
class OllamaClient:
|
|
def __init__(self, model_name: str, base_url: str = "http://localhost:11434"):
|
|
"""Initialize Ollama client.
|
|
|
|
Args:
|
|
model_name (str): Name of the Ollama model to use
|
|
host (str): Ollama server host address
|
|
port (int): Ollama server port
|
|
"""
|
|
self.model_name = model_name
|
|
self.base_url = base_url
|
|
self.headers = {"Content-Type": "application/json"}
|
|
|
|
def generate(self, prompt: str, strip_think: bool = True) -> str:
|
|
"""Process a document using the Ollama API.
|
|
|
|
Args:
|
|
document_text (str): The text content to process
|
|
|
|
Returns:
|
|
str: Processed text response from the model
|
|
|
|
Raises:
|
|
RequestException: If the API call fails
|
|
"""
|
|
try:
|
|
url = f"{self.base_url}/api/generate"
|
|
payload = {
|
|
"model": self.model_name,
|
|
"prompt": prompt,
|
|
"stream": False
|
|
}
|
|
|
|
logger.debug(f"Sending request to Ollama API: {url}")
|
|
response = requests.post(url, json=payload, headers=self.headers)
|
|
response.raise_for_status()
|
|
|
|
result = response.json()
|
|
logger.debug(f"Received response from Ollama API: {result}")
|
|
if strip_think:
|
|
# Remove the "thinking" part from the response
|
|
# the response is expected to be <think>...</think>response_text
|
|
# Check if the response contains <think> tag
|
|
if "<think>" in result.get("response", ""):
|
|
# Split the response and take the part after </think>
|
|
response_parts = result["response"].split("</think>")
|
|
if len(response_parts) > 1:
|
|
# Return the part after </think>
|
|
return response_parts[1].strip()
|
|
else:
|
|
# If no closing tag, return the full response
|
|
return result.get("response", "").strip()
|
|
else:
|
|
# If no <think> tag, return the full response
|
|
return result.get("response", "").strip()
|
|
else:
|
|
# If strip_think is False, return the full response
|
|
return result.get("response", "")
|
|
|
|
|
|
except requests.exceptions.RequestException as e:
|
|
logger.error(f"Error calling Ollama API: {str(e)}")
|
|
raise
|
|
|
|
def get_model_info(self) -> Dict[str, Any]:
|
|
"""Get information about the current model.
|
|
|
|
Returns:
|
|
Dict[str, Any]: Model information
|
|
|
|
Raises:
|
|
RequestException: If the API call fails
|
|
"""
|
|
try:
|
|
url = f"{self.base_url}/api/show"
|
|
payload = {"name": self.model_name}
|
|
|
|
response = requests.post(url, json=payload, headers=self.headers)
|
|
response.raise_for_status()
|
|
|
|
return response.json()
|
|
|
|
except requests.exceptions.RequestException as e:
|
|
logger.error(f"Error getting model info: {str(e)}")
|
|
raise |