Upload sample_sankaku_dataset.ipynb

sample_sankaku_dataset.ipynb

@@ -0,0 +1,141 @@
+{
+  "nbformat": 4,
+  "nbformat_minor": 0,
+  "metadata": {
+    "colab": {
+      "provenance": []
+    },
+    "kernelspec": {
+      "name": "python3",
+      "display_name": "Python 3"
+    },
+    "language_info": {
+      "name": "python"
+    }
+  },
+  "cells": [
+    {
+      "cell_type": "code",
+      "source": [
+        "# ============================================================\n",
+        "# Cell 1 – Install & setup\n",
+        "# ============================================================\n",
+        "!pip install -q cheesechaser pillow matplotlib\n",
+        "\n",
+        "import os\n",
+        "from google.colab import userdata\n",
+        "from pathlib import Path\n",
+        "import random\n",
+        "import json\n",
+        "from PIL import Image\n",
+        "import matplotlib.pyplot as plt\n",
+        "from cheesechaser.datapool import SankakuWebpDataPool\n",
+        "\n",
+        "# Load HF token from Colab secrets\n",
+        "HF_TOKEN = userdata.get('HF_TOKEN')\n",
+        "if not HF_TOKEN:\n",
+        "    raise ValueError(\"Please add HF_TOKEN to Colab secrets (🔑 icon on the left)\")\n",
+        "\n",
+        "os.environ[\"HF_TOKEN\"] = HF_TOKEN\n",
+        "\n",
+        "# ============================================================\n",
+        "# Cell 2 – Download a small sample (≈ 30–40 images)\n",
+        "# ============================================================\n",
+        "# Choose any range you like. IDs around 7M–36M are common.\n",
+        "# Using a small random-ish range so we don't download too much.\n",
+        "sample_ids = list(range(7000000, 7000040))   # 40 images – change freely\n",
+        "\n",
+        "DST_DIR = Path(\"/content/sankaku_sample\")\n",
+        "DST_DIR.mkdir(exist_ok=True, parents=True)\n",
+        "\n",
+        "pool = SankakuWebpDataPool(hf_token=HF_TOKEN)\n",
+        "\n",
+        "print(f\"Downloading {len(sample_ids)} images …\")\n",
+        "pool.batch_download_to_directory(\n",
+        "    resource_ids=sample_ids,\n",
+        "    dst_dir=str(DST_DIR),\n",
+        "    max_workers=8,\n",
+        "    save_metainfo=True,          # important – this creates the *_metainfo.json files\n",
+        ")\n",
+        "print(\"Download finished.\")\n",
+        "\n",
+        "# ============================================================\n",
+        "# Cell 3 – Pick ONE item and inspect its caption / tags\n",
+        "# ============================================================\n",
+        "# Find all downloaded images and their metainfo\n",
+        "images = list(DST_DIR.glob(\"*.webp\")) + list(DST_DIR.glob(\"*.jpg\")) + list(DST_DIR.glob(\"*.png\"))\n",
+        "if not images:\n",
+        "    raise RuntimeError(\"No images were downloaded. Check the ID range or your HF access.\")\n",
+        "\n",
+        "# Pick one random image\n",
+        "one_img = random.choice(images)\n",
+        "meta_path = one_img.with_name(one_img.stem + \"_metainfo.json\")\n",
+        "\n",
+        "print(\"=\" * 60)\n",
+        "print(f\"Inspecting: {one_img.name}\")\n",
+        "print(\"=\" * 60)\n",
+        "\n",
+        "# Show the image\n",
+        "display(Image.open(one_img))\n",
+        "\n",
+        "# Load and pretty-print the caption / tags\n",
+        "if meta_path.exists():\n",
+        "    with open(meta_path, \"r\", encoding=\"utf-8\") as f:\n",
+        "        meta = json.load(f)\n",
+        "\n",
+        "    print(\"\\nFull metainfo keys:\", list(meta.keys()))\n",
+        "    print(\"\\n--- Caption / Tags ---\")\n",
+        "\n",
+        "    # Sankaku metainfo usually contains a 'tags' list or a tag_string\n",
+        "    if \"tags\" in meta:\n",
+        "        # tags can be a list of dicts or a list of strings\n",
+        "        tags = meta[\"tags\"]\n",
+        "        if tags and isinstance(tags[0], dict):\n",
+        "            tag_names = [t.get(\"name\") or t.get(\"tagName\") or str(t) for t in tags]\n",
+        "        else:\n",
+        "            tag_names = tags\n",
+        "        print(\"Tags:\", \", \".join(tag_names))\n",
+        "    elif \"tag_string\" in meta:\n",
+        "        print(\"tag_string:\", meta[\"tag_string\"])\n",
+        "    else:\n",
+        "        # fallback – just dump the whole thing\n",
+        "        print(json.dumps(meta, indent=2, ensure_ascii=False)[:2000])\n",
+        "else:\n",
+        "    print(\"No metainfo JSON found for this image.\")\n",
+        "\n",
+        "# ============================================================\n",
+        "# Cell 4 – Show a random grid of the downloaded images\n",
+        "# ============================================================\n",
+        "random.shuffle(images)\n",
+        "n_show = min(16, len(images))          # up to 4×4 grid\n",
+        "selected = images[:n_show]\n",
+        "\n",
+        "cols = 4\n",
+        "rows = (n_show + cols - 1) // cols\n",
+        "\n",
+        "fig, axes = plt.subplots(rows, cols, figsize=(cols * 3.5, rows * 3.5))\n",
+        "axes = axes.flatten() if n_show > 1 else [axes]\n",
+        "\n",
+        "for ax, img_path in zip(axes, selected):\n",
+        "    img = Image.open(img_path)\n",
+        "    ax.imshow(img)\n",
+        "    ax.set_title(img_path.stem, fontsize=8)\n",
+        "    ax.axis(\"off\")\n",
+        "\n",
+        "# Hide unused axes\n",
+        "for ax in axes[n_show:]:\n",
+        "    ax.axis(\"off\")\n",
+        "\n",
+        "plt.tight_layout()\n",
+        "plt.show()\n",
+        "\n",
+        "print(f\"\\nShowing {n_show} random images from the sample.\")"
+      ],
+      "metadata": {
+        "id": "fQeIjNHoap4Q"
+      },
+      "execution_count": null,
+      "outputs": []
+    }
+  ]
+}