From 673dc2a2002a7b92e88b606831acb99b544665ee Mon Sep 17 00:00:00 2001
From: tvhahn <tvhahn@gmail.com>
Date: Sat, 10 Feb 2024 22:47:00 -0800
Subject: [PATCH] Update code and dependencies

---
 notebooks/scratch/get_hash.ipynb | 30 ++++++++++++++++++------------
 pyproject.toml                   | 28 +++++++++++++++++++++++++---
 src/pyphm/datasets/milling.py    |  5 +++--
 3 files changed, 46 insertions(+), 17 deletions(-)

diff --git a/notebooks/scratch/get_hash.ipynb b/notebooks/scratch/get_hash.ipynb
index 590545f..0056dbf 100644
--- a/notebooks/scratch/get_hash.ipynb
+++ b/notebooks/scratch/get_hash.ipynb
@@ -2,18 +2,24 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": 3,
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "The autoreload extension is already loaded. To reload it, use:\n",
+      "  %reload_ext autoreload\n"
+     ]
+    }
+   ],
    "source": [
     "import hashlib\n",
     "from pathlib import Path\n",
     "import pandas as pd\n",
-    "import os\n",
-    "import numpy as np\n",
-    "import time\n",
-    "import datetime\n",
-    "import h5py\n",
+    "from pyphm.datasets.utils import calculate_md5, check_md5\n",
+    "\n",
     "\n",
     "%load_ext autoreload\n",
     "%autoreload 2"
@@ -21,22 +27,22 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": 4,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "c:\\_Python\\PyPHM\n",
-      "c:\\_Python\\PyPHM\\data\\raw\\airbus\n"
+      "/home/tim/Documents/PyPHM/notebooks\n",
+      "/home/tim/Documents/PyPHM/notebooks/data/raw/airbus\n"
      ]
     }
    ],
    "source": [
     "root_dir = Path.cwd().parent\n",
     "print(root_dir)\n",
-    "path_data_raw_folder = Path(root_dir / 'data/raw/airbus/' )\n",
+    "path_data_raw_folder = Path(root_dir / 'data/raw/milling/' )\n",
     "print(path_data_raw_folder)"
    ]
   },
@@ -73,7 +79,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.8.12"
+   "version": "3.11.7"
   },
   "orig_nbformat": 4
  },
diff --git a/pyproject.toml b/pyproject.toml
index 11a8c9b..bb75a3e 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -3,10 +3,32 @@ name = "pyphm"
 version = "0.0.5"
 description = "Machinery data, made easy"
 requires-python = ">=3.6"
+readme = "README.md"
+authors = [
+  { name = "Tim von Hahn", email = "t.vonhahn@queensu.ca" },
+]
+classifiers = [
+  "Programming Language :: Python :: 3",
+  "License :: OSI Approved :: MIT License",
+  "Operating System :: OS Independent",
+]
+dependencies = [
+  "pandas",
+  "numpy",
+  "py7zr",
+  "rarfile",
+  "tqdm",
+  "scipy",
+  "requests",
+  "h5py",
+  "tables",
+  "gdown"
+]
 
-[build-system]
-requires = ["setuptools>=42", "wheel"]
-build-backend = "setuptools.build_meta"
+[project.urls]
+Homepage = "https://github.com/tvhahn/PyPHM"
+Repository = "https://github.com/tvhahn/PyPHM"
+Documentation = "https://github.com/tvhahn/PyPHM"
 
 [project.optional-dependencies]
 doc = ["sphinx~=4.4.0", "myst-parser"]
diff --git a/src/pyphm/datasets/milling.py b/src/pyphm/datasets/milling.py
index e0b7b47..e7f329f 100644
--- a/src/pyphm/datasets/milling.py
+++ b/src/pyphm/datasets/milling.py
@@ -38,13 +38,14 @@ class MillingDataLoad(PHMDataset):
     """
 
     mirrors = [
+        "https://phm-datasets.s3.amazonaws.com/NASA/",
         # "https://github.com/tvhahn/ml-tool-wear/raw/master/data/raw/",
         "https://drive.google.com/file/d/1_4Hm8RO_7Av1LzGtFnhx6cIN-zi-W40j/view?usp=sharing",
-        # "https://ti.arc.nasa.gov/m/project/prognostic-repository/"     
     ]
 
     resources = [
-        ("mill.zip", "81d821fdef812183a7d38b6f83f7cefa"),
+        ("3.+Milling.zip", "4da3afb0aa50cb3dcdd8e20ed1ed1c7c"),
+        ("mill.zip", "81d821fdef812183a7d38b6f83f7cefa"),        
     ]
 
     def __init__(