zachwill

Does it work in one-shot or I have to prompt it multiple times?

• For the qX_0 variants, they are actually quite straight-forward so deepseek can come up with a correct result in 1 shot. It is already crazy enough, given that ChatGPT and Claude have never produced a working result for me.
• For the qX_K it's more complicated, I would say most of the time I need to re-prompt it 4 to 8 more times.
• The most difficult was q6_K, the code never works until I ask it to only optimize one specific part, while leaving the rest intact (so it does not mess up everything)

How much time does it spent to think?

Around 3-5 minutes per response.

zachwill

import json
import logging
import re
from typing import Any, AsyncGenerator, Optional, Union

import aiohttp
import openai
from azure.search.documents.aio import SearchClient
from azure.search.documents.models import QueryType

zachwill

"""
Multiclass SVMs (Crammer-Singer formulation).

A pure Python re-implementation of:

Large-scale Multiclass Support Vector Machine Training via Euclidean Projection onto the Simplex.
Mathieu Blondel, Akinori Fujino, and Naonori Ueda.
ICPR 2014.
http://www.mblondel.org/publications/mblondel-icpr2014.pdf
"""