From 487be5cd274abcb8a5cf41f3f6d4b4d51e7517ca Mon Sep 17 00:00:00 2001
From: charlie-rasberry <charlie.rasberry@outlook.com>
Date: Fri, 19 Dec 2025 07:14:13 +0000
Subject: [PATCH] Everything is good to go for annotations.

---
 .../notebooks/datasets_reviews.ipynb          |  8 ++---
 .../{ => notebooks}/preprocessing_uber.ipynb  | 32 ++++++++++++++++---
 multitag/{ => notebooks}/uber_cleaned.ipynb   |  0
 multitag/shell.nix                            | 16 ----------
 multitag/src/infer.py                         |  0
 multitag/{ => src}/multitag.py                |  0
 multitag/{ => src}/preprocess.py              |  2 ++
 multitag/{ => src}/sampler.py                 |  0
 multitag/src/train.py                         |  0
 9 files changed, 33 insertions(+), 25 deletions(-)
 rename datasets_reviews.ipynb => multitag/notebooks/datasets_reviews.ipynb (99%)
 rename multitag/{ => notebooks}/preprocessing_uber.ipynb (95%)
 rename multitag/{ => notebooks}/uber_cleaned.ipynb (100%)
 delete mode 100644 multitag/shell.nix
 create mode 100644 multitag/src/infer.py
 rename multitag/{ => src}/multitag.py (100%)
 rename multitag/{ => src}/preprocess.py (99%)
 rename multitag/{ => src}/sampler.py (100%)
 create mode 100644 multitag/src/train.py

diff --git a/datasets_reviews.ipynb b/multitag/notebooks/datasets_reviews.ipynb
similarity index 99%
rename from datasets_reviews.ipynb
rename to multitag/notebooks/datasets_reviews.ipynb
index d71bf3a..3668e1a 100644
--- a/datasets_reviews.ipynb
+++ b/multitag/notebooks/datasets_reviews.ipynb
@@ -2,7 +2,7 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": 2,
    "id": "f3da59fb-eb6b-449f-b8d5-95ddacd456f2",
    "metadata": {},
    "outputs": [],
@@ -14,7 +14,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 3,
    "id": "0c897ead-dfb5-4d18-bcfc-949824a0868f",
    "metadata": {},
    "outputs": [],
@@ -348,7 +348,7 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3 (ipykernel)",
+   "display_name": "reclass",
    "language": "python",
    "name": "python3"
   },
@@ -362,7 +362,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.14.0"
+   "version": "3.14.2"
   }
  },
  "nbformat": 4,
diff --git a/multitag/preprocessing_uber.ipynb b/multitag/notebooks/preprocessing_uber.ipynb
similarity index 95%
rename from multitag/preprocessing_uber.ipynb
rename to multitag/notebooks/preprocessing_uber.ipynb
index 5f5e245..95e36d7 100644
--- a/multitag/preprocessing_uber.ipynb
+++ b/multitag/notebooks/preprocessing_uber.ipynb
@@ -2,7 +2,7 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 11,
+   "execution_count": 1,
    "id": "470fe7c6-1614-4daf-879f-e6c399117c7b",
    "metadata": {},
    "outputs": [],
@@ -13,12 +13,34 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 12,
+   "execution_count": 2,
+   "id": "afe1168c",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "cwd: /mnt/c/Users/ch/6013/multitag/notebooks\n",
+      "exists data: False\n"
+     ]
+    }
+   ],
+   "source": [
+    "import os\n",
+    "print(\"cwd:\", os.getcwd())\n",
+    "print(\"exists data:\", os.path.exists(\"mullitag\"))\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
    "id": "b855045e-2dd1-4fa1-ab5a-8ce8b50b02ee",
    "metadata": {},
    "outputs": [],
    "source": [
-    "df = pd.read_csv('data/uber_reviews.csv', low_memory=False)"
+    "\n",
+    "df = pd.read_csv('../data/uber_reviews.csv', low_memory=False)"
    ]
   },
   {
@@ -437,7 +459,7 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3 (ipykernel)",
+   "display_name": "reclass",
    "language": "python",
    "name": "python3"
   },
@@ -451,7 +473,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.14.0"
+   "version": "3.14.2"
   }
  },
  "nbformat": 4,
diff --git a/multitag/uber_cleaned.ipynb b/multitag/notebooks/uber_cleaned.ipynb
similarity index 100%
rename from multitag/uber_cleaned.ipynb
rename to multitag/notebooks/uber_cleaned.ipynb
diff --git a/multitag/shell.nix b/multitag/shell.nix
deleted file mode 100644
index 168277a..0000000
--- a/multitag/shell.nix
+++ /dev/null
@@ -1,16 +0,0 @@
-{ pkgs ? import <nixpkgs> {} }:
-
-pkgs.mkShell {
-  buildInputs = with pkgs; [
-    python313
-    python313Packages.tkinter
-    python313Packages.pandas
-    python313Packages.numpy
-  ];
-
-  
-  shellHook = ''
-    echo "Development environment loaded"
-    echo "Python: $(python --version)"
-  '';
-}
\ No newline at end of file
diff --git a/multitag/src/infer.py b/multitag/src/infer.py
new file mode 100644
index 0000000..e69de29
diff --git a/multitag/multitag.py b/multitag/src/multitag.py
similarity index 100%
rename from multitag/multitag.py
rename to multitag/src/multitag.py
diff --git a/multitag/preprocess.py b/multitag/src/preprocess.py
similarity index 99%
rename from multitag/preprocess.py
rename to multitag/src/preprocess.py
index dcdffe0..d80b034 100644
--- a/multitag/preprocess.py
+++ b/multitag/src/preprocess.py
@@ -1,3 +1,5 @@
+# preprocess.py
+
 import pandas as pd
 import re
 from langdetect import detect, LangDetectException
diff --git a/multitag/sampler.py b/multitag/src/sampler.py
similarity index 100%
rename from multitag/sampler.py
rename to multitag/src/sampler.py
diff --git a/multitag/src/train.py b/multitag/src/train.py
new file mode 100644
index 0000000..e69de29