tayljordan

{
  "MarineWeatherAPI": {
    "description": "Hourly and daily marine weather forecasts with 5 km resolution",
    "inputs": {
      "location": {
        "type": "object",
        "properties": {
          "latitude": "float",
          "longitude": "float",
          "bbox": "array<float> (optional)",

tayljordan

from sentence_transformers import SentenceTransformer, util
import fitz  # PyMuPDF

def extract_text_from_pdf(path):
    doc = fitz.open(path)
    text = ""
    for page in doc:
        text += page.get_text()
    return text

tayljordan

import json
from serp_maritime import run_search

academies = [
    "Maine Maritime Academy mission statement",
    "Massachusetts Maritime Academy mission statement",
    "SUNY Maritime College mission statement",
    "Texas A&M Maritime Academy mission statement",
    "California State University Maritime Academy mission statement",
    "Great Lakes Maritime Academy mission statement"

tayljordan

# STEP 1

import os
import fitz  # PyMuPDF

pdf_dir = "[PDF DIR]" # the PDF directory
out_dir = os.path.join(pdf_dir, "chunks")
os.makedirs(out_dir, exist_ok=True)

for pdf_file in os.listdir(pdf_dir):

tayljordan

import json
import os
import tempfile
from langchain_community.document_loaders import TextLoader
from langchain.indexes import VectorstoreIndexCreator
from langchain_openai import OpenAIEmbeddings

# Function to load and index the document text in chunks
def load_and_index(file_path, chunk_size=1000, overlap=100):
    with open(file_path, 'r', encoding='utf-8') as file: