jfjensen’s gists

jfjensen / langchain_chatopenai.py

Created September 15, 2023 11:25

Using Langchain to access an LLM on a vLLM server

	from langchain import PromptTemplate
	from langchain.prompts.chat import ChatPromptTemplate, HumanMessagePromptTemplate
	from langchain.chains import LLMChain
	from langchain.chat_models import ChatOpenAI

	import requests

	base_url = "http://localhost:9999/v1"
	x = requests.get(base_url + "/models")
	model = str(x.json()["data"][0]["id"])

jfjensen / langchain_openai.py

Created September 15, 2023 11:16

Using Langchain to access an LLM on a vLLM server

	from langchain import PromptTemplate
	from langchain.chains import LLMChain
	from langchain.llms import OpenAI

	import requests

	base_url = "http://localhost:9999/v1"
	x = requests.get(base_url + "/models")
	model = str(x.json()["data"][0]["id"])
	print(f"model: {model}")

jfjensen / vllm_guidance.py

Last active May 15, 2025 16:45

Using the Guidance AI to access an LLM on a vLLM server

	import guidance
	import requests

	base_url = "http://localhost:9999/v1"
	x = requests.get(base_url + "/models")
	model = str(x.json()["data"][0]["id"])

	guidance.llm = guidance.llms.OpenAI(
	model=model,
	endpoint=base_url,

jfjensen / vllm_openai_chat_api.py

Created September 14, 2023 20:03

Using the OpenAI Chat API to access an LLM on a vLLM server

jfjensen / vllm_openai_api.py

Created September 14, 2023 20:01

Using the OpenAI API to access an LLM on a vLLM server

jfjensen / reviews.py

Last active July 25, 2023 18:33

A Python script for scraping reviews from Amazon given one or more product ASIN codes.

	from playwright.sync_api import sync_playwright
	from selectolax.parser import HTMLParser
	from dataclasses import dataclass
	import time
	import csv
	from loguru import logger

	@dataclass
	class Item:
	asin: str

jfjensen / gtav_yolopv2.py

Last active October 16, 2022 13:06

This code allows you to perform road and vehicle detection in GTA V using YOLOPv2.

	import mss
	from win32gui import FindWindow, GetWindowRect, SetForegroundWindow
	import torch
	import cv2 as cv
	import numpy as np
	from time import time, sleep

	from utils.utils import letterbox, driving_area_mask, lane_line_mask,\
	split_for_trace_model, non_max_suppression, plot_one_box, scale_coords, clip_coords

jfjensen / ready_data.yaml

Created October 1, 2022 17:08

YAML file describing the training data

	# YOLOv5 🚀 by Ultralytics, GPL-3.0 license

	# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
	path: ../PROJECT/ready_data/train_data # dataset root dir
	train: images/train # train images (relative to 'path')
	val: images/val # val images (relative to 'path')
	test: ../PROJECT/ready_data/test_data # test images (optional)

	# Classes
	names:

jfjensen / albion_test.py

Created October 1, 2022 17:05

This code is used for real-time inference on a stream of screenshots from Albion Online

	import mss
	from win32gui import FindWindow, GetWindowRect, SetForegroundWindow
	import torch
	import cv2 as cv
	import numpy as np
	from time import time, sleep

	window_handle = FindWindow(None, "Albion Online Client")
	window_rect = GetWindowRect(window_handle)
	SetForegroundWindow(window_handle)

jfjensen / labels.txt

Created October 1, 2022 16:03

This is a textfile with the labels that we'll use for annotating the objects in the images for the dataset

Jes Fink-Jensen jfjensen