add compute node for llama

This commit is contained in:
streetycat
2023-09-17 13:16:21 +00:00
parent dd3187fc8a
commit f9269e27d5
11 changed files with 334 additions and 20 deletions
+16 -11
View File
@@ -1,7 +1,7 @@
import logging
import requests
from typing import Optional
from typing import Optional, List
from pydantic import BaseModel
from .compute_task import ComputeTask, ComputeTaskState, ComputeTaskType
@@ -26,14 +26,18 @@ class LocalLlama_ComputeNode(Queue_ComputeNode):
class GenerateResponse(BaseModel):
error: Optional[int]
msg: Optional[str]
results: Optional[str]
results: Optional[List[str]]
try:
try:
prompt_msgs = []
for prompt in task.params["prompts"]:
prompt_msgs.append(prompt["content"])
body = {
"prompts": task.params["prompts"]
"prompts": prompt_msgs
}
response = requests.post("http://aigc:7880/generate", data = body, verify=False)
response = requests.post("http://aigc:7880/generate", json = body, verify=False, headers={"Content-Type": "application/json"})
response.close()
logger.info(f"LocalLlama_ComputeNode task responsed, request: {body}, status-code: {response.status_code}, headers: {response.headers}, content: {response.content}")
@@ -47,19 +51,20 @@ class LocalLlama_ComputeNode(Queue_ComputeNode):
}
}
else:
resp = GenerateResponse.parse_raw(response.content.decode("utf-8"))
if resp.error:
resp = response.json()
if "error" in resp:
return {
"state": ComputeTaskState.ERROR,
"error": {
"code": resp.error,
"message": "local llama failed:" + resp.msg
"code": resp["error"],
"message": "local llama failed:" + resp["msg"]
}
}
else:
return {
"content": str(resp.results),
"message": {}
"state": ComputeTaskState.DONE,
"content": str(resp["results"]),
"message": str(resp["results"])
}
except Exception as err:
import traceback