0.532

changelog.md

@@ -1,9 +1,14 @@
 # Changelog
 
+#### 0.531
+🎉 AD `__contains__` can do a nested `in` 'x.y.z' in AD(x={'y': {'z': 10}}) == True
+
 #### 0.530
 ☠️ Stop using rich's print and revert back to builtin print
 🐞 Decouple AD and torch
 🎉 Add a new chart - spider / radar
+🎉 Add scp client with download upload functionality
+
 #### 0.529
 🧹 change code to remove future warnings in text_utils
 

conda/torch_snippets/meta.yaml

@@ -1,9 +1,9 @@
 package:
   name: torch_snippets
-  version: '0.528'
+  version: '0.530'
 source:
-  sha256: 06f5ca72a5cefe10bbf448c71886b15d0e5ec59a199528ff248c2a0918de32cd
-  url: https://files.pythonhosted.org/packages/fc/f6/0beee4f16392f42a8c5a23948e09549617479540e724e40b8e291b8c2b06/torch_snippets-0.528.tar.gz
+  sha256: b0c5de8108d4ec175ec40b73dcfd0e8cec29c0e83171cec8aaeb865f9200f696
+  url: https://files.pythonhosted.org/packages/33/29/ec172cc92aaf82303a25784f9470114ed76acfd52b734a1a9cfd93d20c64/torch_snippets-0.530.tar.gz
 about:
   description: "# Utilities for simple needs\n\n\n\n## torch snippets does a lot of\
     \ default importing for you\nWhether it is numpy, pandas, matplotlib or the useful\

nbs/adapters.ipynb

@@ -51,6 +51,14 @@
     "\n",
     "\n",
     "def b64_2_np(input: str) -> np.ndarray:\n",
+    "    \"\"\"Converts a base64 encoded image to a NumPy array.\n",
+    "\n",
+    "    Args:\n",
+    "        input (str): The base64 encoded image.\n",
+    "\n",
+    "    Returns:\n",
+    "        np.ndarray: The NumPy array representation of the image in RGB format.\n",
+    "    \"\"\"\n",
     "    input = bytes(input, \"utf-8\")\n",
     "    input = base64.b64decode(input)\n",
     "    img_nparr = np.frombuffer(input, np.uint8)\n",
@@ -99,6 +107,17 @@
     "def _process(\n",
     "    df: pd.DataFrame, label_column=\"readable_label\", default_label=\"Background\"\n",
     "):\n",
+    "    \"\"\"\n",
+    "    Process the given DataFrame and convert it into a list of records.\n",
+    "\n",
+    "    Args:\n",
+    "        df (pd.DataFrame): The input DataFrame to be processed.\n",
+    "        label_column (str, optional): The column name for the label. Defaults to \"readable_label\".\n",
+    "        default_label (str, optional): The default label value. Defaults to \"Background\".\n",
+    "\n",
+    "    Returns:\n",
+    "        list: A list of records, where each record is a dictionary representing a row in the DataFrame.\n",
+    "    \"\"\"\n",
     "    df[\"@xbr\"] = df[\"X\"]\n",
     "    df[\"@xtl\"] = df[\"x\"]\n",
     "    df[\"@ybr\"] = df[\"Y\"]\n",
@@ -148,6 +167,22 @@
     "    default_label=\"Background\",\n",
     "    extension=\"jpg\",\n",
     "):\n",
+    "    \"\"\"\n",
+    "    Convert CSV annotations to CVAT XML format.\n",
+    "\n",
+    "    Args:\n",
+    "        images_folder (str): Path to the folder containing the images.\n",
+    "        csvs_folder (str): Path to the folder containing the CSV annotations.\n",
+    "        xml_output_file (str): Path to the output XML file.\n",
+    "        items (list, optional): List of items to process. If None, all items will be processed. Defaults to None.\n",
+    "        parquet (bool, optional): Whether the annotations are stored in Parquet format. Defaults to False.\n",
+    "        relative_df (bool, optional): Whether the bounding box coordinates in the CSV are relative to the image size. Defaults to True.\n",
+    "        default_label (str, optional): Default label for the bounding boxes. Defaults to \"Background\".\n",
+    "        extension (str, optional): Image file extension. Defaults to \"jpg\".\n",
+    "\n",
+    "    Returns:\n",
+    "        None\n",
+    "    \"\"\"\n",
     "    images_folder, csvs_folder = [P(_) for _ in [images_folder, csvs_folder]]\n",
     "    data = AttrDict({\"annotations\": {\"image\": []}})\n",
     "    if items is None:\n",
@@ -174,6 +209,16 @@
     "\n",
     "\n",
     "def _get_attribute_columns(column):\n",
+    "    \"\"\"\n",
+    "    Get attribute columns from a given column.\n",
+    "\n",
+    "    Args:\n",
+    "        column (pandas.Series): The input column.\n",
+    "\n",
+    "    Returns:\n",
+    "        set: A set of attribute columns extracted from the input column.\n",
+    "    \"\"\"\n",
+    "\n",
     "    def _get_columns_from_row(item):\n",
     "        if item != item:\n",
     "            return []\n",
@@ -187,6 +232,16 @@
     "\n",
     "\n",
     "def _get_attribute_data(item, column_name):\n",
+    "    \"\"\"\n",
+    "    Retrieves the attribute data for a given item and column name.\n",
+    "\n",
+    "    Parameters:\n",
+    "    - item: The item to retrieve the attribute data from.\n",
+    "    - column_name: The name of the column to retrieve the attribute data for.\n",
+    "\n",
+    "    Returns:\n",
+    "    - The attribute data for the given item and column name, or np.nan if not found.\n",
+    "    \"\"\"\n",
     "    if item != item:\n",
     "        return item\n",
     "    if isinstance(item, dict):\n",
@@ -203,6 +258,15 @@
     "\n",
     "\n",
     "def _cvat_ann_2_csv(ann):\n",
+    "    \"\"\"\n",
+    "    Convert CVAT annotation to a pandas DataFrame in CSV format.\n",
+    "\n",
+    "    Args:\n",
+    "        ann (dict): CVAT annotation dictionary.\n",
+    "\n",
+    "    Returns:\n",
+    "        pandas.DataFrame: DataFrame containing the converted annotation data in CSV format.\n",
+    "    \"\"\"\n",
     "    if \"box\" not in ann:\n",
     "        return pd.DataFrame()\n",
     "    if isinstance(ann.box, AttrDict):\n",
@@ -236,8 +300,21 @@
     "\n",
     "\n",
     "def cvat_2_csvs(xmlfile, csvs_folder):\n",
+    "    \"\"\"\n",
+    "    Convert CVAT XML annotations to CSV files.\n",
+    "\n",
+    "    Args:\n",
+    "        xmlfile (str): Path to the CVAT XML file.\n",
+    "        csvs_folder (str): Path to the folder where the CSV files will be saved.\n",
+    "\n",
+    "    Returns:\n",
+    "        None\n",
+    "    \"\"\"\n",
     "    data = read_xml(xmlfile)\n",
-    "    for item in data.annotations.image:\n",
+    "    items = data.annotations.image\n",
+    "    if not isinstance(items, list):\n",
+    "        items = [items]\n",
+    "    for item in items:\n",
     "        try:\n",
     "            df = _cvat_ann_2_csv(item)\n",
     "            save_at = f'{csvs_folder}/{stem(item[\"@name\"])}.csv'\n",
@@ -271,7 +348,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.8.15"
+   "version": "3.8.undefined"
   },
   "orig_nbformat": 4,
   "vscode": {

nbs/bokeh_plotting.ipynb

@@ -65,6 +65,18 @@
     "# | export\n",
     "# | hide\n",
     "def parse_sz(size):\n",
+    "    \"\"\"\n",
+    "    Parses the size argument and returns a tuple of width and height.\n",
+    "\n",
+    "    Args:\n",
+    "        size (int or tuple): The size argument to be parsed.\n",
+    "\n",
+    "    Returns:\n",
+    "        tuple: A tuple of width and height.\n",
+    "\n",
+    "    Raises:\n",
+    "        NotImplementedError: If the size argument is not an int or a tuple of length 2.\n",
+    "    \"\"\"\n",
     "    if isinstance(size, int):\n",
     "        return size, size\n",
     "    elif isinstance(size, tuple):\n",
@@ -74,6 +86,17 @@
     "\n",
     "\n",
     "def get_bplot(sz=500, **kwargs):\n",
+    "    \"\"\"\n",
+    "    Create a Bokeh plot with specified size and tools.\n",
+    "\n",
+    "    Parameters:\n",
+    "    - sz (int): Size of the plot in pixels.\n",
+    "    - **kwargs: Additional keyword arguments for customizing the plot.\n",
+    "\n",
+    "    Returns:\n",
+    "    - plot (bokeh.plotting.Figure): Bokeh plot object.\n",
+    "\n",
+    "    \"\"\"\n",
     "    h, w = parse_sz(sz)\n",
     "    output_notebook()\n",
     "    plot = figure(\n",