better version for low light images

.gitattributes

@@ -70,36 +70,3 @@ sample_06_050244.png filter=lfs diff=lfs merge=lfs -text
 sample_07_040111.png filter=lfs diff=lfs merge=lfs -text
 sample_08_170404.png filter=lfs diff=lfs merge=lfs -text
 sample_09_130480.png filter=lfs diff=lfs merge=lfs -text
-checkpoints/ckpt-34.data-00000-of-00001 filter=lfs diff=lfs merge=lfs -text
-img_00000.png filter=lfs diff=lfs merge=lfs -text
-img_00005.png filter=lfs diff=lfs merge=lfs -text
-img_00010.png filter=lfs diff=lfs merge=lfs -text
-img_00585.png filter=lfs diff=lfs merge=lfs -text
-img_00600.png filter=lfs diff=lfs merge=lfs -text
-img_00605.png filter=lfs diff=lfs merge=lfs -text
-img_00610.png filter=lfs diff=lfs merge=lfs -text
-img_00775.png filter=lfs diff=lfs merge=lfs -text
-img_00785.png filter=lfs diff=lfs merge=lfs -text
-img_01020.png filter=lfs diff=lfs merge=lfs -text
-img_01135.png filter=lfs diff=lfs merge=lfs -text
-img_01775.png filter=lfs diff=lfs merge=lfs -text
-img_01780.png filter=lfs diff=lfs merge=lfs -text
-img_01815.png filter=lfs diff=lfs merge=lfs -text
-img_01820.png filter=lfs diff=lfs merge=lfs -text
-img_02130.png filter=lfs diff=lfs merge=lfs -text
-img_02135.png filter=lfs diff=lfs merge=lfs -text
-img_02140.png filter=lfs diff=lfs merge=lfs -text
-img_02860.png filter=lfs diff=lfs merge=lfs -text
-img_02865.png filter=lfs diff=lfs merge=lfs -text
-img_03180.png filter=lfs diff=lfs merge=lfs -text
-img_03185.png filter=lfs diff=lfs merge=lfs -text
-img_03980.png filter=lfs diff=lfs merge=lfs -text
-img_03985.png filter=lfs diff=lfs merge=lfs -text
-img_06220.png filter=lfs diff=lfs merge=lfs -text
-img_06225.png filter=lfs diff=lfs merge=lfs -text
-img_06780.png filter=lfs diff=lfs merge=lfs -text
-img_06785.png filter=lfs diff=lfs merge=lfs -text
-img_07545.png filter=lfs diff=lfs merge=lfs -text
-img_07550.png filter=lfs diff=lfs merge=lfs -text
-img_09935.png filter=lfs diff=lfs merge=lfs -text
-img_09975.png filter=lfs diff=lfs merge=lfs -text

checkpoints/checkpoint

@@ -1,4 +0,0 @@
-model_checkpoint_path: "ckpt-34"
-all_model_checkpoint_paths: "ckpt-34"
-all_model_checkpoint_timestamps: 1770885364.6021488
-last_preserved_timestamp: 1770885364.6021488

checkpoints/ckpt-34.data-00000-of-00001

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:ef70f3bd501ffb9f7c237d49042ca432ac074beecbd4f6853ba23b4e70fb06df
-size 207904461

visible-to-thermal-f7984f (1).ipynb

@@ -1,3574 +0,0 @@
-{
- "cells": [
-  {
-   "cell_type": "code",
-   "execution_count": 1,
-   "id": "3df26cb5",
-   "metadata": {
-    "execution": {
-     "iopub.execute_input": "2026-02-12T21:58:41.908567Z",
-     "iopub.status.busy": "2026-02-12T21:58:41.908310Z",
-     "iopub.status.idle": "2026-02-12T21:59:00.902318Z",
-     "shell.execute_reply": "2026-02-12T21:59:00.901653Z"
-    },
-    "papermill": {
-     "duration": 19.000924,
-     "end_time": "2026-02-12T21:59:00.903787",
-     "exception": false,
-     "start_time": "2026-02-12T21:58:41.902863",
-     "status": "completed"
-    },
-    "tags": []
-   },
-   "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "2026-02-12 21:58:43.423256: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:477] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered\n",
-      "WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
-      "E0000 00:00:1770933523.610196      20 cuda_dnn.cc:8310] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered\n",
-      "E0000 00:00:1770933523.667872      20 cuda_blas.cc:1418] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered\n",
-      "I0000 00:00:1770933537.686778      20 gpu_device.cc:2022] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 15511 MB memory:  -> device: 0, name: Tesla P100-PCIE-16GB, pci bus id: 0000:00:04.0, compute capability: 6.0\n"
-     ]
-    }
-   ],
-   "source": [
-    "import glob\n",
-    "import os\n",
-    "import time\n",
-    "import numpy as np\n",
-    "import tensorflow as tf\n",
-    "from tensorflow.keras import layers, Model, Input\n",
-    "from tensorflow.keras.layers import Conv2D, Conv2DTranspose, LeakyReLU, Dropout, Concatenate\n",
-    "from tensorflow.keras.optimizers import Adam\n",
-    "from tensorflow.keras.applications import VGG19\n",
-    "\n",
-    "# -------------------- Settings --------------------\n",
-    "IMG_SIZE = 256\n",
-    "BATCH_SIZE = 1\n",
-    "EPOCHS = 190\n",
-    "BASE_DIR = \"/kaggle/input/llvip-dataset/LLVIP\"  # change if needed\n",
-    "CHECKPOINT_DIR = \"checkpoints\"\n",
-    "OUTPUT_DIR = \"outputs\"\n",
-    "LOG_INTERVAL = 100\n",
-    "SAVE_INTERVAL_EPOCHS = 10\n",
-    "\n",
-    "# Discriminator update steps per generator step (1 or 2)\n",
-    "D_steps_per_G = 1\n",
-    "\n",
-    "# -------------------- Dataset Helpers --------------------\n",
-    "def load_image_pair(visible_path, ir_path, image_size=(IMG_SIZE, IMG_SIZE)):\n",
-    "    vis = tf.io.read_file(visible_path)\n",
-    "    vis = tf.image.decode_png(vis, channels=3)\n",
-    "    vis = tf.image.resize(vis, image_size)\n",
-    "    vis = tf.cast(vis, tf.float32) / 127.5 - 1.0  # [-1,1]\n",
-    "\n",
-    "    ir = tf.io.read_file(ir_path)\n",
-    "    ir = tf.image.decode_png(ir, channels=3)\n",
-    "    ir = tf.image.resize(ir, image_size)\n",
-    "    ir = tf.cast(ir, tf.float32) / 127.5 - 1.0\n",
-    "\n",
-    "    return vis, ir\n",
-    "def augment_image(vis, ir):\n",
-    "    vis = tf.image.random_contrast(vis, lower=0.8, upper=1.2)\n",
-    "    vis = tf.image.random_brightness(vis, max_delta=0.1)\n",
-    "    ir = tf.image.random_contrast(ir, lower=0.8, upper=1.2)\n",
-    "    return vis, ir\n",
-    "\n",
-    "def make_dataset(\n",
-    "    visible_dir,\n",
-    "    ir_dir,\n",
-    "    image_size=256,\n",
-    "    batch_size=1,\n",
-    "    shuffle=False,\n",
-    "    start=None,\n",
-    "    limit=None   # NEW\n",
-    "):\n",
-    "    visible_files = sorted(glob.glob(os.path.join(visible_dir, \"*\")))\n",
-    "    ir_files = sorted(glob.glob(os.path.join(ir_dir, \"*\")))\n",
-    "\n",
-    "    print(f\"Found {len(visible_files)} visible images and {len(ir_files)} IR images\")\n",
-    "\n",
-    "    if len(visible_files) == 0 or len(ir_files) == 0:\n",
-    "        raise ValueError(\"❌ No images found! Check paths.\")\n",
-    "\n",
-    "    # ---- LIMIT DATASET SIZE ----\n",
-    "    if limit is not None:\n",
-    "        visible_files = visible_files[start:limit]\n",
-    "        ir_files = ir_files[start:limit]\n",
-    "        print(f\"Using only {limit} image pairs\")\n",
-    "\n",
-    "    dataset = tf.data.Dataset.from_tensor_slices((visible_files, ir_files))\n",
-    "\n",
-    "    if shuffle and len(visible_files) > 1:\n",
-    "        dataset = dataset.shuffle(buffer_size=len(visible_files))\n",
-    "\n",
-    "    dataset = dataset.map(\n",
-    "        lambda v, i: load_image_pair(v, i, image_size),\n",
-    "        num_parallel_calls=tf.data.AUTOTUNE\n",
-    "    )\n",
-    "\n",
-    "    dataset = dataset.batch(batch_size).prefetch(tf.data.AUTOTUNE)\n",
-    "    dataset = dataset.map(augment_image, num_parallel_calls=tf.data.AUTOTUNE)\n",
-    "\n",
-    "    return dataset\n",
-    "\n",
-    "\n",
-    "def get_train_dataset(base_dir=BASE_DIR):\n",
-    "    train_visible = os.path.join(base_dir, \"visible/train\")\n",
-    "    train_ir = os.path.join(base_dir, \"infrared/train\")\n",
-    "\n",
-    "    return make_dataset(\n",
-    "        train_visible,\n",
-    "        train_ir,\n",
-    "        (IMG_SIZE, IMG_SIZE),\n",
-    "        batch_size=BATCH_SIZE,\n",
-    "        start=0,\n",
-    "        limit=12025   # 🔥 10K TRAIN\n",
-    "    )\n",
-    "\n",
-    "\n",
-    "def get_val_dataset(base_dir=BASE_DIR):\n",
-    "    val_visible = os.path.join(base_dir, \"visible/train\")\n",
-    "    val_ir = os.path.join(base_dir, \"infrared/train\")\n",
-    "\n",
-    "    return make_dataset(\n",
-    "        val_visible,\n",
-    "        val_ir,\n",
-    "        (IMG_SIZE, IMG_SIZE),\n",
-    "        batch_size=1,\n",
-    "        shuffle=True,\n",
-    "        start=10000,\n",
-    "        limit=12025   # 🔥 2.5K TEST\n",
-    "    )\n",
-    "\n",
-    "\n",
-    "# -------------------- Instance Normalization (per-channel) --------------------\n",
-    "class InstanceNormalization(layers.Layer):\n",
-    "    def __init__(self, epsilon=1e-5, **kwargs):\n",
-    "        super().__init__(**kwargs)\n",
-    "        self.epsilon = epsilon\n",
-    "\n",
-    "    def build(self, input_shape):\n",
-    "        channels = int(input_shape[-1])\n",
-    "        self.gamma = self.add_weight(name='gamma', shape=(1,1,1,channels), initializer='ones', trainable=True)\n",
-    "        self.beta = self.add_weight(name='beta', shape=(1,1,1,channels), initializer='zeros', trainable=True)\n",
-    "\n",
-    "    def call(self, inputs):\n",
-    "        mean, var = tf.nn.moments(inputs, axes=[1,2], keepdims=True)\n",
-    "        normalized = (inputs - mean) / tf.sqrt(var + self.epsilon)\n",
-    "        return self.gamma * normalized + self.beta\n",
-    "\n",
-    "from tensorflow.keras.layers import Layer\n",
-    "\n",
-    "class AddNoise(Layer):\n",
-    "    def __init__(self, stddev=0.02, **kwargs):\n",
-    "        super().__init__(**kwargs)\n",
-    "        self.stddev = stddev\n",
-    "\n",
-    "    def call(self, inputs, training=None):\n",
-    "        # you might choose to only add noise during training, or always\n",
-    "        noise = tf.random.normal(tf.shape(inputs), mean=0.0, stddev=self.stddev)\n",
-    "        return inputs + noise\n",
-    "\n",
-    "    def get_config(self):\n",
-    "        config = super().get_config()\n",
-    "        config.update({\"stddev\": self.stddev})\n",
-    "        return config\n",
-    "\n",
-    "\n",
-    "def build_generator_attention(img_size=256):\n",
-    "    inputs = Input(shape=(img_size, img_size, 3))\n",
-    "    x = AddNoise(stddev=0.02)(inputs)\n",
-    "\n",
-    "    e1 = Conv2D(64, 4, strides=2, padding='same')(x)\n",
-    "    e1 = LeakyReLU(0.2)(e1)\n",
-    "\n",
-    "    e2 = Conv2D(128, 4, strides=2, padding='same')(e1)\n",
-    "    e2 = LeakyReLU(0.2)(e2)\n",
-    "\n",
-    "    e3 = Conv2D(256, 4, strides=2, padding='same')(e2)\n",
-    "    e3 = LeakyReLU(0.2)(e3)\n",
-    "\n",
-    "    e4 = Conv2D(512, 4, strides=2, padding='same')(e3)\n",
-    "    e4 = LeakyReLU(0.2)(e4)\n",
-    "\n",
-    "    # Bottleneck + Attention block (optional)\n",
-    "    b = Conv2D(512, 4, strides=2, padding='same')(e4)\n",
-    "    b = LeakyReLU(0.2)(b)\n",
-    "\n",
-    "    # Decoder\n",
-    "    d1 = Conv2DTranspose(512, 4, strides=2, padding='same')(b)\n",
-    "    d1 = tf.keras.layers.ReLU()(d1)\n",
-    "    d1 = Concatenate()([d1, e4])\n",
-    "\n",
-    "    d2 = Conv2DTranspose(256, 4, strides=2, padding='same')(d1)\n",
-    "    d2 = tf.keras.layers.ReLU()(d2)\n",
-    "    d2 = Concatenate()([d2, e3])\n",
-    "\n",
-    "    d3 = Conv2DTranspose(128, 4, strides=2, padding='same')(d2)\n",
-    "    d3 = tf.keras.layers.ReLU()(d3)\n",
-    "    d3 = Concatenate()([d3, e2])\n",
-    "\n",
-    "    d4 = Conv2DTranspose(64, 4, strides=2, padding='same')(d3)\n",
-    "    d4 = tf.keras.layers.ReLU()(d4)\n",
-    "    d4 = Concatenate()([d4, e1])\n",
-    "\n",
-    "    output = Conv2DTranspose(3, 4, strides=2, padding='same', activation='tanh')(d4)\n",
-    "\n",
-    "    return Model(inputs, output, name='Generator')\n",
-    "\n",
-    "\n",
-    "# -------------------- Stronger PatchGAN Discriminator --------------------\n",
-    "def build_patch_discriminator(img_size=IMG_SIZE):\n",
-    "    inp = Input(shape=(img_size, img_size, 3))\n",
-    "    tar = Input(shape=(img_size, img_size, 3))\n",
-    "    x = Concatenate()([inp, tar])  # condition on input\n",
-    "\n",
-    "    x = Conv2D(64, 4, strides=2, padding='same')(x)\n",
-    "    x = LeakyReLU(0.2)(x)\n",
-    "\n",
-    "    x = Conv2D(128, 4, strides=2, padding='same', use_bias=False)(x)\n",
-    "    x = InstanceNormalization()(x)\n",
-    "    x = LeakyReLU(0.2)(x)\n",
-    "\n",
-    "    x = Conv2D(256, 4, strides=2, padding='same', use_bias=False)(x)\n",
-    "    x = InstanceNormalization()(x)\n",
-    "    x = LeakyReLU(0.2)(x)\n",
-    "\n",
-    "    out = Conv2D(1, 4, strides=1, padding='same')(x)\n",
-    "    return Model([inp, tar], out, name='PatchDiscriminator')\n",
-    "\n",
-    "\n",
-    "# -------------------- Losses & VGG perceptual --------------------\n",
-    "\n",
-    "\n",
-    "# -------------------- Perceptual model setup --------------------\n",
-    "from tensorflow.keras.applications.vgg19 import VGG19, preprocess_input\n",
-    "from tensorflow.keras import Model\n",
-    "\n",
-    "from tensorflow.keras.applications import VGG19\n",
-    "from tensorflow.keras.models import Model\n",
-    "\n",
-    "# Initialize VGG19 without top layers and no weights\n",
-    "vgg_model = VGG19(include_top=False, weights=None, input_shape=(256, 256, 3))\n",
-    "\n",
-    "# Load the manually downloaded weights\n",
-    "weights_path = '/kaggle/input/models/saisumathappala/vgg19-base-model/tensorflow2/default/1/vgg19_weights_tf_dim_ordering_tf_kernels_notop.h5'\n",
-    "vgg_model.load_weights(weights_path)\n",
-    "\n",
-    "# Freeze all layers\n",
-    "vgg_model.trainable = False\n",
-    "\n",
-    "# Use block4_conv2 output for perceptual loss\n",
-    "perceptual_model = Model(inputs=vgg_model.input,\n",
-    "                         outputs=vgg_model.get_layer('block4_conv2').output)\n",
-    "# from tensorflow.keras.applications import VGG19\n",
-    "# from tensorflow.keras.models import Model\n",
-    "# from tensorflow.keras.applications.vgg19 import preprocess_input\n",
-    "\n",
-    "# base_model = tf.keras.applications.VGG19(\n",
-    "#     weights='imagenet',\n",
-    "#     input_shape=(IMG_SIZE, IMG_SIZE, 3),\n",
-    "#     include_top=False)\n",
-    "\n",
-    "# base_model.trainable = False\n",
-    "\n",
-    "# Perceptual layer\n",
-    "\n",
-    "\n",
-    "bce = tf.keras.losses.BinaryCrossentropy(from_logits=True)\n",
-    "def discriminator_loss_fn(real_logits, fake_logits):\n",
-    "    real_loss = bce(tf.ones_like(real_logits), real_logits)\n",
-    "    fake_loss = bce(tf.zeros_like(fake_logits), fake_logits)\n",
-    "    total_loss = (real_loss + fake_loss) * 0.5\n",
-    "    return total_loss\n",
-    "\n",
-    "# ---------------- GAN loss ----------------\n",
-    "def generator_adv_loss(d_outs):\n",
-    "    loss = 0.0\n",
-    "    for out in d_outs:\n",
-    "        loss += bce(tf.ones_like(out), out)\n",
-    "    return loss / len(d_outs)\n",
-    "\n",
-    "\n",
-    "# ---------------- Weighted L1 ----------------\n",
-    "def weighted_l1_loss(target, gen):\n",
-    "    t = (target + 1.0) / 2.0\n",
-    "    g = (gen + 1.0) / 2.0\n",
-    "    weight = tf.clip_by_value(t ** 2 * 4.0, 1.0, 4.0)\n",
-    "    return tf.reduce_mean(weight * tf.abs(t - g))\n",
-    "\n",
-    "\n",
-    "# ---------------- Perceptual ----------------\n",
-    "from tensorflow.keras.applications.vgg19 import preprocess_input\n",
-    "\n",
-    "def perceptual_loss(target, gen):\n",
-    "    target_rgb = preprocess_input((target + 1.0) * 127.5)\n",
-    "    gen_rgb = preprocess_input((gen + 1.0) * 127.5)\n",
-    "\n",
-    "    f_t = perceptual_model(target_rgb)\n",
-    "    f_g = perceptual_model(gen_rgb)\n",
-    "\n",
-    "    return tf.reduce_mean(tf.abs(f_t - f_g))\n",
-    "\n",
-    "def generator_total_loss(d_out, gen_out, target):\n",
-    "\n",
-    "    adv  = generator_adv_loss([d_out])   # wrap in list\n",
-    "    wl1  = weighted_l1_loss(target, gen_out)\n",
-    "    perc = perceptual_loss(target, gen_out)\n",
-    "\n",
-    "    # Recommended weights for LLVIP\n",
-    "    total_loss = (\n",
-    "        1.0   * adv +     # realism\n",
-    "        100.0 * wl1 +     # alignment\n",
-    "        10.0  * perc      # structure\n",
-    "    )\n",
-    "\n",
-    "    return total_loss, adv, wl1, perc\n",
-    "\n",
-    "\n",
-    "\n",
-    "\n",
-    "\n",
-    "generator = build_generator_attention(IMG_SIZE)\n",
-    "D_full = build_patch_discriminator(IMG_SIZE)\n",
-    "\n",
-    "# TTUR: different LRs for G and D\n",
-    "generator_optimizer = Adam(2e-4, beta_1=0.5)\n",
-    "d_optimizer = Adam(5e-5, beta_1=0.5)\n",
-    "\n",
-    "# Checkpoints\n",
-    "import tensorflow as tf\n",
-    "import numpy as np\n",
-    "import cv2\n",
-    "import os\n",
-    "from tqdm import tqdm\n",
-    "import matplotlib.pyplot as plt\n",
-    "\n",
-    "# -------------------- Checkpoint Paths --------------------\n",
-    "\n",
-    "# 1️⃣ Pretrained checkpoint (read-only Kaggle input)\n",
-    "PRETRAINED_DIR = \"/kaggle/input/models/saisumathappala/image-to-ir-gan/tensorflow2/default/2\"\n",
-    "\n",
-    "# 2️⃣ Training checkpoint (writeable)\n",
-    "CHECKPOINT_DIR = \"checkpoints\"\n",
-    "\n",
-    "os.makedirs(CHECKPOINT_DIR, exist_ok=True)\n",
-    "\n",
-    "\n",
-    "# -------------------- Checkpoint Objects --------------------\n",
-    "ckpt = tf.train.Checkpoint(\n",
-    "    generator=generator,\n",
-    "    D_full=D_full,\n",
-    "    generator_optimizer=generator_optimizer,\n",
-    "    d_optimizer=d_optimizer\n",
-    ")\n",
-    "\n",
-    "# Manager for TRAINING checkpoints\n",
-    "ckpt_manager = tf.train.CheckpointManager(\n",
-    "    ckpt,\n",
-    "    CHECKPOINT_DIR,\n",
-    "    max_to_keep=5\n",
-    ")\n"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 2,
-   "id": "0c5c2773",
-   "metadata": {
-    "execution": {
-     "iopub.execute_input": "2026-02-12T21:59:00.912705Z",
-     "iopub.status.busy": "2026-02-12T21:59:00.911688Z",
-     "iopub.status.idle": "2026-02-12T21:59:01.885299Z",
-     "shell.execute_reply": "2026-02-12T21:59:01.884435Z"
-    },
-    "papermill": {
-     "duration": 0.979,
-     "end_time": "2026-02-12T21:59:01.886677",
-     "exception": false,
-     "start_time": "2026-02-12T21:59:00.907677",
-     "status": "completed"
-    },
-    "tags": []
-   },
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "No training checkpoint found. Loading pretrained weights...\n",
-      "✅ Restored pretrained weights from: /kaggle/input/models/saisumathappala/image-to-ir-gan/tensorflow2/default/2/ckpt-34\n"
-     ]
-    }
-   ],
-   "source": [
-    "# -------------------- Restore Logic --------------------\n",
-    "\n",
-    "if ckpt_manager.latest_checkpoint:\n",
-    "    # Case 1 → Resume training normally\n",
-    "    ckpt.restore(ckpt_manager.latest_checkpoint).expect_partial()\n",
-    "    print(f\"✅ Resumed from training checkpoint: {ckpt_manager.latest_checkpoint}\")\n",
-    "\n",
-    "else:\n",
-    "    # Case 2 → First run → Load pretrained weights ONCE\n",
-    "    print(\"No training checkpoint found. Loading pretrained weights...\")\n",
-    "\n",
-    "    pretrained_ckpt = tf.train.latest_checkpoint(PRETRAINED_DIR)\n",
-    "\n",
-    "    if pretrained_ckpt:\n",
-    "        ckpt.restore(pretrained_ckpt).expect_partial()\n",
-    "        print(f\"✅ Restored pretrained weights from: {pretrained_ckpt}\")\n",
-    "    else:\n",
-    "        print(\"⚠️ No pretrained checkpoint found. Training from scratch.\")\n"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 3,
-   "id": "358eb256",
-   "metadata": {
-    "execution": {
-     "iopub.execute_input": "2026-02-12T21:59:01.895018Z",
-     "iopub.status.busy": "2026-02-12T21:59:01.894351Z",
-     "iopub.status.idle": "2026-02-12T21:59:01.900980Z",
-     "shell.execute_reply": "2026-02-12T21:59:01.900207Z"
-    },
-    "papermill": {
-     "duration": 0.011908,
-     "end_time": "2026-02-12T21:59:01.902226",
-     "exception": false,
-     "start_time": "2026-02-12T21:59:01.890318",
-     "status": "completed"
-    },
-    "tags": []
-   },
-   "outputs": [],
-   "source": [
-    "@tf.function\n",
-    "def d_train_step(input_vis, target_ir, gen_out):\n",
-    "\n",
-    "    with tf.GradientTape() as tape:\n",
-    "\n",
-    "        d_real = D_full([input_vis, target_ir], training=True)\n",
-    "        d_fake = D_full([input_vis, gen_out], training=True)\n",
-    "\n",
-    "        d_loss = discriminator_loss_fn(d_real, d_fake)\n",
-    "\n",
-    "    d_vars  = D_full.trainable_variables\n",
-    "    d_grads = tape.gradient(d_loss, d_vars)\n",
-    "\n",
-    "    d_optimizer.apply_gradients(zip(d_grads, d_vars))\n",
-    "\n",
-    "    return d_loss\n",
-    "\n",
-    "# -------------------- Generator Train Step --------------------\n",
-    "@tf.function\n",
-    "def g_train_step(input_vis, target_ir):\n",
-    "\n",
-    "    with tf.GradientTape() as tape:\n",
-    "\n",
-    "        gen_out = generator(input_vis, training=True)\n",
-    "\n",
-    "        d_fake = D_full([input_vis, gen_out], training=True)\n",
-    "\n",
-    "        total_g_loss, adv_loss, wl1_loss, perc_loss = \\\n",
-    "            generator_total_loss(d_fake, gen_out, target_ir)\n",
-    "\n",
-    "    g_vars  = generator.trainable_variables\n",
-    "    g_grads = tape.gradient(total_g_loss, g_vars)\n",
-    "\n",
-    "    generator_optimizer.apply_gradients(zip(g_grads, g_vars))\n",
-    "\n",
-    "    return total_g_loss, adv_loss, wl1_loss, perc_loss, gen_out\n"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 4,
-   "id": "f2e569db",
-   "metadata": {
-    "execution": {
-     "iopub.execute_input": "2026-02-12T21:59:01.910320Z",
-     "iopub.status.busy": "2026-02-12T21:59:01.909856Z",
-     "iopub.status.idle": "2026-02-12T21:59:01.916085Z",
-     "shell.execute_reply": "2026-02-12T21:59:01.915351Z"
-    },
-    "papermill": {
-     "duration": 0.011612,
-     "end_time": "2026-02-12T21:59:01.917295",
-     "exception": false,
-     "start_time": "2026-02-12T21:59:01.905683",
-     "status": "completed"
-    },
-    "tags": []
-   },
-   "outputs": [],
-   "source": [
-    "def train(train_ds, val_ds=None, epochs=EPOCHS):\n",
-    "\n",
-    "    global_step = 0\n",
-    "\n",
-    "    for epoch in range(1, epochs + 1):\n",
-    "\n",
-    "        start = time.time()\n",
-    "        print(f\"\\nEpoch {epoch}/{epochs}\")\n",
-    "\n",
-    "        for batch, (inp, tar) in enumerate(train_ds):\n",
-    "\n",
-    "            # --------------------\n",
-    "            # Generator forward\n",
-    "            # --------------------\n",
-    "            gen_out = generator(inp, training=True)\n",
-    "\n",
-    "            # --------------------\n",
-    "            # Train Discriminator\n",
-    "            # --------------------\n",
-    "            for _ in range(D_steps_per_G):\n",
-    "                d_loss_val = d_train_step(inp, tar, gen_out)\n",
-    "\n",
-    "            # --------------------\n",
-    "            # Train Generator\n",
-    "            # --------------------\n",
-    "            g_total, g_adv, g_l1, g_perc, gen_out = \\\n",
-    "                g_train_step(inp, tar)\n",
-    "\n",
-    "            global_step += 1\n",
-    "\n",
-    "            # Step logging\n",
-    "            if batch % 50 == 0:\n",
-    "                print(\n",
-    "                    f\"Step {global_step}: \"\n",
-    "                    f\"D={d_loss_val:.4f}, \"\n",
-    "                    f\"G={g_total:.4f}, \"\n",
-    "                    f\"Adv={g_adv:.4f}, \"\n",
-    "                    f\"L1={g_l1:.4f}, \"\n",
-    "                    f\"Perc={g_perc:.4f}\"\n",
-    "                )\n",
-    "\n",
-    "        # --------------------\n",
-    "        # Visualization (TEST SET)\n",
-    "        # --------------------\n",
-    "        if val_ds is not None:\n",
-    "            print(\"🖼 Generating validation samples...\")\n",
-    "            generate_and_save_images(generator, val_ds, epoch)\n",
-    "\n",
-    "        # --------------------\n",
-    "        # Checkpoint\n",
-    "        # --------------------\n",
-    "        if epoch % SAVE_INTERVAL_EPOCHS == 0:\n",
-    "            path = ckpt_manager.save()\n",
-    "            print(f\" Saved checkpoint: {path}\")\n",
-    "\n",
-    "        print(f\" Epoch {epoch} finished in {time.time()-start:.1f}s\")\n"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 5,
-   "id": "bb729cec",
-   "metadata": {
-    "execution": {
-     "iopub.execute_input": "2026-02-12T21:59:01.924705Z",
-     "iopub.status.busy": "2026-02-12T21:59:01.924476Z",
-     "iopub.status.idle": "2026-02-12T21:59:01.931295Z",
-     "shell.execute_reply": "2026-02-12T21:59:01.930717Z"
-    },
-    "papermill": {
-     "duration": 0.011607,
-     "end_time": "2026-02-12T21:59:01.932316",
-     "exception": false,
-     "start_time": "2026-02-12T21:59:01.920709",
-     "status": "completed"
-    },
-    "tags": []
-   },
-   "outputs": [],
-   "source": [
-    "import os\n",
-    "import tensorflow as tf\n",
-    "\n",
-    "# -------------------- Visualization Utilities --------------------\n",
-    "def to_uint8(x):\n",
-    "    \"\"\"Convert tensor from [-1,1] → uint8 [0,255].\"\"\"\n",
-    "    x = (x + 1.0) * 127.5\n",
-    "    x = tf.clip_by_value(x, 0, 255)\n",
-    "    return tf.cast(x, tf.uint8)\n",
-    "\n",
-    "def generate_and_save_images(model, val_ds, epoch, out_dir=OUTPUT_DIR, num_rows=5):\n",
-    "    \"\"\"\n",
-    "    Creates a 5-row image where each row = [Visible | Real IR | Generated IR].\n",
-    "    \"\"\"\n",
-    "    os.makedirs(out_dir, exist_ok=True)\n",
-    "\n",
-    "    # Collect up to num_rows samples\n",
-    "    rows = []\n",
-    "    for i, (v_inp, v_tar) in enumerate(val_ds.take(num_rows)):\n",
-    "        pred = model(v_inp, training=False)\n",
-    "\n",
-    "        vis = to_uint8(v_inp[0])\n",
-    "        targ = to_uint8(v_tar[0])\n",
-    "        gen = to_uint8(pred[0])\n",
-    "\n",
-    "        # Ensure same height\n",
-    "        h = min(vis.shape[0], targ.shape[0], gen.shape[0])\n",
-    "        w = min(vis.shape[1], targ.shape[1], gen.shape[1])\n",
-    "        vis, targ, gen = vis[:h, :w], targ[:h, :w], gen[:h, :w]\n",
-    "\n",
-    "        # Concatenate horizontally\n",
-    "        row = tf.concat([vis, targ, gen], axis=1)\n",
-    "        rows.append(row)\n",
-    "\n",
-    "    # Stack vertically → 5-row image\n",
-    "    grid = tf.concat(rows, axis=0)\n",
-    "\n",
-    "    out_path = os.path.join(out_dir, f\"epoch_{epoch:03d}.png\")\n",
-    "    tf.keras.preprocessing.image.save_img(out_path, grid.numpy())\n"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 6,
-   "id": "302e6898",
-   "metadata": {
-    "execution": {
-     "iopub.execute_input": "2026-02-12T21:59:01.939607Z",
-     "iopub.status.busy": "2026-02-12T21:59:01.939374Z",
-     "iopub.status.idle": "2026-02-12T21:59:01.945225Z",
-     "shell.execute_reply": "2026-02-12T21:59:01.944630Z"
-    },
-    "papermill": {
-     "duration": 0.010711,
-     "end_time": "2026-02-12T21:59:01.946333",
-     "exception": false,
-     "start_time": "2026-02-12T21:59:01.935622",
-     "status": "completed"
-    },
-    "tags": []
-   },
-   "outputs": [],
-   "source": [
-    "def train(train_ds, val_ds=None, epochs=EPOCHS):\n",
-    "    global_step = 0\n",
-    "    for epoch in range(1, epochs + 1):\n",
-    "        start = time.time()\n",
-    "        print(f\"\\nEpoch {epoch}/{epochs}\")\n",
-    "\n",
-    "        for batch, (inp, tar) in enumerate(train_ds):\n",
-    "            gen_out = generator(inp, training=True)\n",
-    "\n",
-    "            # Train Discriminator\n",
-    "            for _ in range(D_steps_per_G):\n",
-    "                d_loss_val = d_train_step(inp, tar, gen_out)\n",
-    "\n",
-    "            # Train Generator\n",
-    "            g_total_loss, g_adv, g_l1,g_prec, gen_out2 = g_train_step(inp, tar)\n",
-    "            global_step += 1\n",
-    "        print(f\" step {global_step}: \"\n",
-    "                      f\"D={d_loss_val:.4f}, G={g_total_loss:.4f}, Adv={g_adv:.4f}, L1={g_l1:.4f} , prec={g_prec:.4f}\")\n",
-    "\n",
-    "        # Save visualization\n",
-    "        if val_ds is not None:\n",
-    "            print(f\"🖼 Generating sample output for epoch {epoch}...\")\n",
-    "            generate_and_save_images(generator, val_ds, epoch)\n",
-    "\n",
-    "        # Save checkpoint\n",
-    "        if epoch % SAVE_INTERVAL_EPOCHS == 0:\n",
-    "            path = ckpt_manager.save()\n",
-    "            print(f\" Saved checkpoint: {path}\")\n",
-    "\n",
-    "        print(f\" Epoch {epoch} finished in {time.time() - start:.1f}s\")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 7,
-   "id": "2581acd4",
-   "metadata": {
-    "execution": {
-     "iopub.execute_input": "2026-02-12T21:59:01.953697Z",
-     "iopub.status.busy": "2026-02-12T21:59:01.953474Z",
-     "iopub.status.idle": "2026-02-12T21:59:02.537240Z",
-     "shell.execute_reply": "2026-02-12T21:59:02.536482Z"
-    },
-    "papermill": {
-     "duration": 0.589004,
-     "end_time": "2026-02-12T21:59:02.538691",
-     "exception": false,
-     "start_time": "2026-02-12T21:59:01.949687",
-     "status": "completed"
-    },
-    "tags": []
-   },
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Found 12025 visible images and 12025 IR images\n",
-      "Using only 12025 image pairs\n"
-     ]
-    }
-   ],
-   "source": [
-    "train_ds = get_train_dataset()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 8,
-   "id": "9f18cfa2",
-   "metadata": {
-    "execution": {
-     "iopub.execute_input": "2026-02-12T21:59:02.548373Z",
-     "iopub.status.busy": "2026-02-12T21:59:02.547591Z",
-     "iopub.status.idle": "2026-02-12T21:59:02.551403Z",
-     "shell.execute_reply": "2026-02-12T21:59:02.550725Z"
-    },
-    "papermill": {
-     "duration": 0.009773,
-     "end_time": "2026-02-12T21:59:02.552655",
-     "exception": false,
-     "start_time": "2026-02-12T21:59:02.542882",
-     "status": "completed"
-    },
-    "tags": []
-   },
-   "outputs": [],
-   "source": [
-    "# val_ds = train_ds.take(75)\n",
-    "# train_ds = train_ds.skip(75)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 9,
-   "id": "7b2a39df",
-   "metadata": {
-    "execution": {
-     "iopub.execute_input": "2026-02-12T21:59:02.561724Z",
-     "iopub.status.busy": "2026-02-12T21:59:02.561429Z",
-     "iopub.status.idle": "2026-02-12T21:59:02.573514Z",
-     "shell.execute_reply": "2026-02-12T21:59:02.572847Z"
-    },
-    "papermill": {
-     "duration": 0.018255,
-     "end_time": "2026-02-12T21:59:02.575056",
-     "exception": false,
-     "start_time": "2026-02-12T21:59:02.556801",
-     "status": "completed"
-    },
-    "tags": []
-   },
-   "outputs": [],
-   "source": [
-    "import matplotlib.pyplot as plt\n",
-    "import tensorflow as tf\n",
-    "import numpy as np\n",
-    "\n",
-    "def to_uint8(x):\n",
-    "    \"\"\"Convert tensor from [-1,1] → uint8 [0,255].\"\"\"\n",
-    "    x = (x + 1.0) * 127.5\n",
-    "    x = tf.clip_by_value(x, 0, 255)\n",
-    "    return tf.cast(x, tf.uint8)\n",
-    "\n",
-    "# Shuffle the dataset to take random samples\n",
-    "train_ds_shuffled = train_ds.shuffle(buffer_size=1000, reshuffle_each_iteration=True)\n",
-    "\n",
-    "# Take 10 images\n",
-    "sample_ds = train_ds_shuffled.take(10)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 10,
-   "id": "4f876063",
-   "metadata": {
-    "execution": {
-     "iopub.execute_input": "2026-02-12T21:59:02.584167Z",
-     "iopub.status.busy": "2026-02-12T21:59:02.583876Z",
-     "iopub.status.idle": "2026-02-12T21:59:02.587381Z",
-     "shell.execute_reply": "2026-02-12T21:59:02.586759Z"
-    },
-    "papermill": {
-     "duration": 0.009294,
-     "end_time": "2026-02-12T21:59:02.588465",
-     "exception": false,
-     "start_time": "2026-02-12T21:59:02.579171",
-     "status": "completed"
-    },
-    "tags": []
-   },
-   "outputs": [],
-   "source": [
-    "# train(train_ds, val_ds, epochs=EPOCHS)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "id": "a4a82990",
-   "metadata": {
-    "papermill": {
-     "duration": 0.003627,
-     "end_time": "2026-02-12T21:59:02.596030",
-     "exception": false,
-     "start_time": "2026-02-12T21:59:02.592403",
-     "status": "completed"
-    },
-    "tags": []
-   },
-   "outputs": [],
-   "source": []
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 11,
-   "id": "bb6cd8e0",
-   "metadata": {
-    "execution": {
-     "iopub.execute_input": "2026-02-12T21:59:02.604728Z",
-     "iopub.status.busy": "2026-02-12T21:59:02.604438Z",
-     "iopub.status.idle": "2026-02-12T22:08:25.100908Z",
-     "shell.execute_reply": "2026-02-12T22:08:25.099998Z"
-    },
-    "papermill": {
-     "duration": 562.502476,
-     "end_time": "2026-02-12T22:08:25.102199",
-     "exception": false,
-     "start_time": "2026-02-12T21:59:02.599723",
-     "status": "completed"
-    },
-    "tags": []
-   },
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "No training checkpoint found. Loading pretrained weights...\n",
-      "✅ Restored pretrained weights from: /kaggle/input/models/saisumathappala/image-to-ir-gan/tensorflow2/default/2/ckpt-34\n",
-      "No checkpoint found, using uninitialized model\n",
-      "Found 12025 visible images and 12025 IR images\n",
-      "Using only 12025 image pairs\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "  0%|          | 0/12025 [00:00<?, ?it/s]I0000 00:00:1770933543.075318      20 cuda_dnn.cc:529] Loaded cuDNN version 90300\n",
-      "100%|██████████| 12025/12025 [09:22<00:00, 21.39it/s]"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "==== Test Dataset Metrics ====\n",
-      "L1 Loss   : 0.1371\n",
-      "PSNR      : 20.8206\n",
-      "SSIM      : 0.5706\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "\n"
-     ]
-    }
-   ],
-   "source": [
-    "import tensorflow as tf\n",
-    "import numpy as np\n",
-    "import cv2\n",
-    "import os\n",
-    "from tqdm import tqdm\n",
-    "import matplotlib.pyplot as plt\n",
-    "# -------------------- Load Checkpoint --------------------\n",
-    "ckpt = tf.train.Checkpoint(generator=generator,\n",
-    "                           D_full=D_full,\n",
-    "                           generator_optimizer=generator_optimizer,\n",
-    "                           d_optimizer=d_optimizer)\n",
-    "ckpt_manager = tf.train.CheckpointManager(ckpt, CHECKPOINT_DIR, max_to_keep=1)\n",
-    "# -------------------- Restore Logic --------------------\n",
-    "\n",
-    "if ckpt_manager.latest_checkpoint:\n",
-    "    # Case 1 → Resume training normally\n",
-    "    ckpt.restore(ckpt_manager.latest_checkpoint).expect_partial()\n",
-    "    print(f\"✅ Resumed from training checkpoint: {ckpt_manager.latest_checkpoint}\")\n",
-    "\n",
-    "else:\n",
-    "    # Case 2 → First run → Load pretrained weights ONCE\n",
-    "    print(\"No training checkpoint found. Loading pretrained weights...\")\n",
-    "\n",
-    "    pretrained_ckpt = tf.train.latest_checkpoint(PRETRAINED_DIR)\n",
-    "\n",
-    "    if pretrained_ckpt:\n",
-    "        ckpt.restore(pretrained_ckpt).expect_partial()\n",
-    "        print(f\"✅ Restored pretrained weights from: {pretrained_ckpt}\")\n",
-    "    else:\n",
-    "        print(\"⚠️ No pretrained checkpoint found. Training from scratch.\")\n",
-    "\n",
-    "\n",
-    "# Restore latest checkpoint\n",
-    "if ckpt_manager.latest_checkpoint:\n",
-    "    ckpt.restore(ckpt_manager.latest_checkpoint).expect_partial()\n",
-    "    print(f\"Restored from {ckpt_manager.latest_checkpoint}\")\n",
-    "else:\n",
-    "    print(\"No checkpoint found, using uninitialized model\")\n",
-    "\n",
-    "# -------------------- Dataset Evaluation --------------------\n",
-    "def l1_loss(y_true, y_pred):\n",
-    "    return tf.reduce_mean(tf.abs(y_true - y_pred))\n",
-    "\n",
-    "def evaluate(test_ds):\n",
-    "    l1_list, psnr_list, ssim_list = [], [], []\n",
-    "    for vis, ir in tqdm(test_ds):\n",
-    "        pred = generator(vis, training=False)\n",
-    "        l1 = l1_loss(ir, pred).numpy()\n",
-    "        psnr = tf.image.psnr(ir, pred, max_val=2.0).numpy()\n",
-    "        ssim = tf.image.ssim(ir, pred, max_val=2.0).numpy()\n",
-    "\n",
-    "        l1_list.append(l1)\n",
-    "        psnr_list.append(np.mean(psnr))\n",
-    "        ssim_list.append(np.mean(ssim))\n",
-    "\n",
-    "    print(\"==== Test Dataset Metrics ====\")\n",
-    "    print(f\"L1 Loss   : {np.mean(l1_list):.4f}\")\n",
-    "    print(f\"PSNR      : {np.mean(psnr_list):.4f}\")\n",
-    "    print(f\"SSIM      : {np.mean(ssim_list):.4f}\")\n",
-    "\n",
-    "# Example usage\n",
-    "test_ds = get_val_dataset()  # Define this function\n",
-    "evaluate(train_ds)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 12,
-   "id": "268eb52f",
-   "metadata": {
-    "execution": {
-     "iopub.execute_input": "2026-02-12T22:08:25.423242Z",
-     "iopub.status.busy": "2026-02-12T22:08:25.422959Z",
-     "iopub.status.idle": "2026-02-12T22:08:25.430078Z",
-     "shell.execute_reply": "2026-02-12T22:08:25.429489Z"
-    },
-    "papermill": {
-     "duration": 0.166727,
-     "end_time": "2026-02-12T22:08:25.431112",
-     "exception": false,
-     "start_time": "2026-02-12T22:08:25.264385",
-     "status": "completed"
-    },
-    "tags": []
-   },
-   "outputs": [],
-   "source": [
-    "import os\n",
-    "import tensorflow as tf\n",
-    "\n",
-    "OUTPUT_DIR = \"outputs\"\n",
-    "\n",
-    "\n",
-    "# -------------------- Utils --------------------\n",
-    "def to_uint8(x):\n",
-    "    \"\"\"\n",
-    "    Convert tensor from [-1,1] → uint8 [0,255]\n",
-    "    \"\"\"\n",
-    "    x = (x + 1.0) * 127.5\n",
-    "    x = tf.clip_by_value(x, 0, 255)\n",
-    "    return tf.cast(x, tf.uint8)\n",
-    "\n",
-    "\n",
-    "# -------------------- Save 3 Images in 1 --------------------\n",
-    "def generate(\n",
-    "        model,\n",
-    "        dataset,\n",
-    "        epoch,\n",
-    "        out_dir=OUTPUT_DIR):\n",
-    "\n",
-    "    os.makedirs(out_dir, exist_ok=True)\n",
-    "\n",
-    "    for idx, (visible, real_ir) in enumerate(dataset):\n",
-    "        if idx%5==0:\n",
-    "\n",
-    "            # Generate fake IR\n",
-    "            fake_ir = model(visible, training=False)\n",
-    "    \n",
-    "            # Convert → uint8\n",
-    "            vis  = to_uint8(visible[0])\n",
-    "            real = to_uint8(real_ir[0])\n",
-    "            fake = to_uint8(fake_ir[0])\n",
-    "    \n",
-    "            # Safety crop (if any mismatch)\n",
-    "            h = min(vis.shape[0], real.shape[0], fake.shape[0])\n",
-    "            w = min(vis.shape[1], real.shape[1], fake.shape[1])\n",
-    "    \n",
-    "            vis  = vis[:h, :w]\n",
-    "            real = real[:h, :w]\n",
-    "            fake = fake[:h, :w]\n",
-    "\n",
-    "            # -------- Concatenate 3 images horizontally --------\n",
-    "            trio = tf.concat([vis, real, fake], axis=1)\n",
-    "    \n",
-    "            # Save path\n",
-    "            save_path = os.path.join(\n",
-    "                out_dir,\n",
-    "                f\"img_{idx:05d}.png\"\n",
-    "            )\n",
-    "    \n",
-    "            # Save image\n",
-    "            tf.keras.preprocessing.image.save_img(\n",
-    "                save_path,\n",
-    "                trio.numpy()\n",
-    "            )\n",
-    "            print(idx)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 13,
-   "id": "161508e1",
-   "metadata": {
-    "execution": {
-     "iopub.execute_input": "2026-02-12T22:08:25.750226Z",
-     "iopub.status.busy": "2026-02-12T22:08:25.749948Z",
-     "iopub.status.idle": "2026-02-12T22:12:22.229019Z",
-     "shell.execute_reply": "2026-02-12T22:12:22.228225Z"
-    },
-    "papermill": {
-     "duration": 236.642335,
-     "end_time": "2026-02-12T22:12:22.230235",
-     "exception": false,
-     "start_time": "2026-02-12T22:08:25.587900",
-     "status": "completed"
-    },
-    "tags": []
-   },
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "0\n",
-      "5\n",
-      "10\n",
-      "15\n",
-      "20\n",
-      "25\n",
-      "30\n",
-      "35\n",
-      "40\n",
-      "45\n",
-      "50\n",
-      "55\n",
-      "60\n",
-      "65\n",
-      "70\n",
-      "75\n",
-      "80\n",
-      "85\n",
-      "90\n",
-      "95\n",
-      "100\n",
-      "105\n",
-      "110\n",
-      "115\n",
-      "120\n",
-      "125\n",
-      "130\n",
-      "135\n",
-      "140\n",
-      "145\n",
-      "150\n",
-      "155\n",
-      "160\n",
-      "165\n",
-      "170\n",
-      "175\n",
-      "180\n",
-      "185\n",
-      "190\n",
-      "195\n",
-      "200\n",
-      "205\n",
-      "210\n",
-      "215\n",
-      "220\n",
-      "225\n",
-      "230\n",
-      "235\n",
-      "240\n",
-      "245\n",
-      "250\n",
-      "255\n",
-      "260\n",
-      "265\n",
-      "270\n",
-      "275\n",
-      "280\n",
-      "285\n",
-      "290\n",
-      "295\n",
-      "300\n",
-      "305\n",
-      "310\n",
-      "315\n",
-      "320\n",
-      "325\n",
-      "330\n",
-      "335\n",
-      "340\n",
-      "345\n",
-      "350\n",
-      "355\n",
-      "360\n",
-      "365\n",
-      "370\n",
-      "375\n",
-      "380\n",
-      "385\n",
-      "390\n",
-      "395\n",
-      "400\n",
-      "405\n",
-      "410\n",
-      "415\n",
-      "420\n",
-      "425\n",
-      "430\n",
-      "435\n",
-      "440\n",
-      "445\n",
-      "450\n",
-      "455\n",
-      "460\n",
-      "465\n",
-      "470\n",
-      "475\n",
-      "480\n",
-      "485\n",
-      "490\n",
-      "495\n",
-      "500\n",
-      "505\n",
-      "510\n",
-      "515\n",
-      "520\n",
-      "525\n",
-      "530\n",
-      "535\n",
-      "540\n",
-      "545\n",
-      "550\n",
-      "555\n",
-      "560\n",
-      "565\n",
-      "570\n",
-      "575\n",
-      "580\n",
-      "585\n",
-      "590\n",
-      "595\n",
-      "600\n",
-      "605\n",
-      "610\n",
-      "615\n",
-      "620\n",
-      "625\n",
-      "630\n",
-      "635\n",
-      "640\n",
-      "645\n",
-      "650\n",
-      "655\n",
-      "660\n",
-      "665\n",
-      "670\n",
-      "675\n",
-      "680\n",
-      "685\n",
-      "690\n",
-      "695\n",
-      "700\n",
-      "705\n",
-      "710\n",
-      "715\n",
-      "720\n",
-      "725\n",
-      "730\n",
-      "735\n",
-      "740\n",
-      "745\n",
-      "750\n",
-      "755\n",
-      "760\n",
-      "765\n",
-      "770\n",
-      "775\n",
-      "780\n",
-      "785\n",
-      "790\n",
-      "795\n",
-      "800\n",
-      "805\n",
-      "810\n",
-      "815\n",
-      "820\n",
-      "825\n",
-      "830\n",
-      "835\n",
-      "840\n",
-      "845\n",
-      "850\n",
-      "855\n",
-      "860\n",
-      "865\n",
-      "870\n",
-      "875\n",
-      "880\n",
-      "885\n",
-      "890\n",
-      "895\n",
-      "900\n",
-      "905\n",
-      "910\n",
-      "915\n",
-      "920\n",
-      "925\n",
-      "930\n",
-      "935\n",
-      "940\n",
-      "945\n",
-      "950\n",
-      "955\n",
-      "960\n",
-      "965\n",
-      "970\n",
-      "975\n",
-      "980\n",
-      "985\n",
-      "990\n",
-      "995\n",
-      "1000\n",
-      "1005\n",
-      "1010\n",
-      "1015\n",
-      "1020\n",
-      "1025\n",
-      "1030\n",
-      "1035\n",
-      "1040\n",
-      "1045\n",
-      "1050\n",
-      "1055\n",
-      "1060\n",
-      "1065\n",
-      "1070\n",
-      "1075\n",
-      "1080\n",
-      "1085\n",
-      "1090\n",
-      "1095\n",
-      "1100\n",
-      "1105\n",
-      "1110\n",
-      "1115\n",
-      "1120\n",
-      "1125\n",
-      "1130\n",
-      "1135\n",
-      "1140\n",
-      "1145\n",
-      "1150\n",
-      "1155\n",
-      "1160\n",
-      "1165\n",
-      "1170\n",
-      "1175\n",
-      "1180\n",
-      "1185\n",
-      "1190\n",
-      "1195\n",
-      "1200\n",
-      "1205\n",
-      "1210\n",
-      "1215\n",
-      "1220\n",
-      "1225\n",
-      "1230\n",
-      "1235\n",
-      "1240\n",
-      "1245\n",
-      "1250\n",
-      "1255\n",
-      "1260\n",
-      "1265\n",
-      "1270\n",
-      "1275\n",
-      "1280\n",
-      "1285\n",
-      "1290\n",
-      "1295\n",
-      "1300\n",
-      "1305\n",
-      "1310\n",
-      "1315\n",
-      "1320\n",
-      "1325\n",
-      "1330\n",
-      "1335\n",
-      "1340\n",
-      "1345\n",
-      "1350\n",
-      "1355\n",
-      "1360\n",
-      "1365\n",
-      "1370\n",
-      "1375\n",
-      "1380\n",
-      "1385\n",
-      "1390\n",
-      "1395\n",
-      "1400\n",
-      "1405\n",
-      "1410\n",
-      "1415\n",
-      "1420\n",
-      "1425\n",
-      "1430\n",
-      "1435\n",
-      "1440\n",
-      "1445\n",
-      "1450\n",
-      "1455\n",
-      "1460\n",
-      "1465\n",
-      "1470\n",
-      "1475\n",
-      "1480\n",
-      "1485\n",
-      "1490\n",
-      "1495\n",
-      "1500\n",
-      "1505\n",
-      "1510\n",
-      "1515\n",
-      "1520\n",
-      "1525\n",
-      "1530\n",
-      "1535\n",
-      "1540\n",
-      "1545\n",
-      "1550\n",
-      "1555\n",
-      "1560\n",
-      "1565\n",
-      "1570\n",
-      "1575\n",
-      "1580\n",
-      "1585\n",
-      "1590\n",
-      "1595\n",
-      "1600\n",
-      "1605\n",
-      "1610\n",
-      "1615\n",
-      "1620\n",
-      "1625\n",
-      "1630\n",
-      "1635\n",
-      "1640\n",
-      "1645\n",
-      "1650\n",
-      "1655\n",
-      "1660\n",
-      "1665\n",
-      "1670\n",
-      "1675\n",
-      "1680\n",
-      "1685\n",
-      "1690\n",
-      "1695\n",
-      "1700\n",
-      "1705\n",
-      "1710\n",
-      "1715\n",
-      "1720\n",
-      "1725\n",
-      "1730\n",
-      "1735\n",
-      "1740\n",
-      "1745\n",
-      "1750\n",
-      "1755\n",
-      "1760\n",
-      "1765\n",
-      "1770\n",
-      "1775\n",
-      "1780\n",
-      "1785\n",
-      "1790\n",
-      "1795\n",
-      "1800\n",
-      "1805\n",
-      "1810\n",
-      "1815\n",
-      "1820\n",
-      "1825\n",
-      "1830\n",
-      "1835\n",
-      "1840\n",
-      "1845\n",
-      "1850\n",
-      "1855\n",
-      "1860\n",
-      "1865\n",
-      "1870\n",
-      "1875\n",
-      "1880\n",
-      "1885\n",
-      "1890\n",
-      "1895\n",
-      "1900\n",
-      "1905\n",
-      "1910\n",
-      "1915\n",
-      "1920\n",
-      "1925\n",
-      "1930\n",
-      "1935\n",
-      "1940\n",
-      "1945\n",
-      "1950\n",
-      "1955\n",
-      "1960\n",
-      "1965\n",
-      "1970\n",
-      "1975\n",
-      "1980\n",
-      "1985\n",
-      "1990\n",
-      "1995\n",
-      "2000\n",
-      "2005\n",
-      "2010\n",
-      "2015\n",
-      "2020\n",
-      "2025\n",
-      "2030\n",
-      "2035\n",
-      "2040\n",
-      "2045\n",
-      "2050\n",
-      "2055\n",
-      "2060\n",
-      "2065\n",
-      "2070\n",
-      "2075\n",
-      "2080\n",
-      "2085\n",
-      "2090\n",
-      "2095\n",
-      "2100\n",
-      "2105\n",
-      "2110\n",
-      "2115\n",
-      "2120\n",
-      "2125\n",
-      "2130\n",
-      "2135\n",
-      "2140\n",
-      "2145\n",
-      "2150\n",
-      "2155\n",
-      "2160\n",
-      "2165\n",
-      "2170\n",
-      "2175\n",
-      "2180\n",
-      "2185\n",
-      "2190\n",
-      "2195\n",
-      "2200\n",
-      "2205\n",
-      "2210\n",
-      "2215\n",
-      "2220\n",
-      "2225\n",
-      "2230\n",
-      "2235\n",
-      "2240\n",
-      "2245\n",
-      "2250\n",
-      "2255\n",
-      "2260\n",
-      "2265\n",
-      "2270\n",
-      "2275\n",
-      "2280\n",
-      "2285\n",
-      "2290\n",
-      "2295\n",
-      "2300\n",
-      "2305\n",
-      "2310\n",
-      "2315\n",
-      "2320\n",
-      "2325\n",
-      "2330\n",
-      "2335\n",
-      "2340\n",
-      "2345\n",
-      "2350\n",
-      "2355\n",
-      "2360\n",
-      "2365\n",
-      "2370\n",
-      "2375\n",
-      "2380\n",
-      "2385\n",
-      "2390\n",
-      "2395\n",
-      "2400\n",
-      "2405\n",
-      "2410\n",
-      "2415\n",
-      "2420\n",
-      "2425\n",
-      "2430\n",
-      "2435\n",
-      "2440\n",
-      "2445\n",
-      "2450\n",
-      "2455\n",
-      "2460\n",
-      "2465\n",
-      "2470\n",
-      "2475\n",
-      "2480\n",
-      "2485\n",
-      "2490\n",
-      "2495\n",
-      "2500\n",
-      "2505\n",
-      "2510\n",
-      "2515\n",
-      "2520\n",
-      "2525\n",
-      "2530\n",
-      "2535\n",
-      "2540\n",
-      "2545\n",
-      "2550\n",
-      "2555\n",
-      "2560\n",
-      "2565\n",
-      "2570\n",
-      "2575\n",
-      "2580\n",
-      "2585\n",
-      "2590\n",
-      "2595\n",
-      "2600\n",
-      "2605\n",
-      "2610\n",
-      "2615\n",
-      "2620\n",
-      "2625\n",
-      "2630\n",
-      "2635\n",
-      "2640\n",
-      "2645\n",
-      "2650\n",
-      "2655\n",
-      "2660\n",
-      "2665\n",
-      "2670\n",
-      "2675\n",
-      "2680\n",
-      "2685\n",
-      "2690\n",
-      "2695\n",
-      "2700\n",
-      "2705\n",
-      "2710\n",
-      "2715\n",
-      "2720\n",
-      "2725\n",
-      "2730\n",
-      "2735\n",
-      "2740\n",
-      "2745\n",
-      "2750\n",
-      "2755\n",
-      "2760\n",
-      "2765\n",
-      "2770\n",
-      "2775\n",
-      "2780\n",
-      "2785\n",
-      "2790\n",
-      "2795\n",
-      "2800\n",
-      "2805\n",
-      "2810\n",
-      "2815\n",
-      "2820\n",
-      "2825\n",
-      "2830\n",
-      "2835\n",
-      "2840\n",
-      "2845\n",
-      "2850\n",
-      "2855\n",
-      "2860\n",
-      "2865\n",
-      "2870\n",
-      "2875\n",
-      "2880\n",
-      "2885\n",
-      "2890\n",
-      "2895\n",
-      "2900\n",
-      "2905\n",
-      "2910\n",
-      "2915\n",
-      "2920\n",
-      "2925\n",
-      "2930\n",
-      "2935\n",
-      "2940\n",
-      "2945\n",
-      "2950\n",
-      "2955\n",
-      "2960\n",
-      "2965\n",
-      "2970\n",
-      "2975\n",
-      "2980\n",
-      "2985\n",
-      "2990\n",
-      "2995\n",
-      "3000\n",
-      "3005\n",
-      "3010\n",
-      "3015\n",
-      "3020\n",
-      "3025\n",
-      "3030\n",
-      "3035\n",
-      "3040\n",
-      "3045\n",
-      "3050\n",
-      "3055\n",
-      "3060\n",
-      "3065\n",
-      "3070\n",
-      "3075\n",
-      "3080\n",
-      "3085\n",
-      "3090\n",
-      "3095\n",
-      "3100\n",
-      "3105\n",
-      "3110\n",
-      "3115\n",
-      "3120\n",
-      "3125\n",
-      "3130\n",
-      "3135\n",
-      "3140\n",
-      "3145\n",
-      "3150\n",
-      "3155\n",
-      "3160\n",
-      "3165\n",
-      "3170\n",
-      "3175\n",
-      "3180\n",
-      "3185\n",
-      "3190\n",
-      "3195\n",
-      "3200\n",
-      "3205\n",
-      "3210\n",
-      "3215\n",
-      "3220\n",
-      "3225\n",
-      "3230\n",
-      "3235\n",
-      "3240\n",
-      "3245\n",
-      "3250\n",
-      "3255\n",
-      "3260\n",
-      "3265\n",
-      "3270\n",
-      "3275\n",
-      "3280\n",
-      "3285\n",
-      "3290\n",
-      "3295\n",
-      "3300\n",
-      "3305\n",
-      "3310\n",
-      "3315\n",
-      "3320\n",
-      "3325\n",
-      "3330\n",
-      "3335\n",
-      "3340\n",
-      "3345\n",
-      "3350\n",
-      "3355\n",
-      "3360\n",
-      "3365\n",
-      "3370\n",
-      "3375\n",
-      "3380\n",
-      "3385\n",
-      "3390\n",
-      "3395\n",
-      "3400\n",
-      "3405\n",
-      "3410\n",
-      "3415\n",
-      "3420\n",
-      "3425\n",
-      "3430\n",
-      "3435\n",
-      "3440\n",
-      "3445\n",
-      "3450\n",
-      "3455\n",
-      "3460\n",
-      "3465\n",
-      "3470\n",
-      "3475\n",
-      "3480\n",
-      "3485\n",
-      "3490\n",
-      "3495\n",
-      "3500\n",
-      "3505\n",
-      "3510\n",
-      "3515\n",
-      "3520\n",
-      "3525\n",
-      "3530\n",
-      "3535\n",
-      "3540\n",
-      "3545\n",
-      "3550\n",
-      "3555\n",
-      "3560\n",
-      "3565\n",
-      "3570\n",
-      "3575\n",
-      "3580\n",
-      "3585\n",
-      "3590\n",
-      "3595\n",
-      "3600\n",
-      "3605\n",
-      "3610\n",
-      "3615\n",
-      "3620\n",
-      "3625\n",
-      "3630\n",
-      "3635\n",
-      "3640\n",
-      "3645\n",
-      "3650\n",
-      "3655\n",
-      "3660\n",
-      "3665\n",
-      "3670\n",
-      "3675\n",
-      "3680\n",
-      "3685\n",
-      "3690\n",
-      "3695\n",
-      "3700\n",
-      "3705\n",
-      "3710\n",
-      "3715\n",
-      "3720\n",
-      "3725\n",
-      "3730\n",
-      "3735\n",
-      "3740\n",
-      "3745\n",
-      "3750\n",
-      "3755\n",
-      "3760\n",
-      "3765\n",
-      "3770\n",
-      "3775\n",
-      "3780\n",
-      "3785\n",
-      "3790\n",
-      "3795\n",
-      "3800\n",
-      "3805\n",
-      "3810\n",
-      "3815\n",
-      "3820\n",
-      "3825\n",
-      "3830\n",
-      "3835\n",
-      "3840\n",
-      "3845\n",
-      "3850\n",
-      "3855\n",
-      "3860\n",
-      "3865\n",
-      "3870\n",
-      "3875\n",
-      "3880\n",
-      "3885\n",
-      "3890\n",
-      "3895\n",
-      "3900\n",
-      "3905\n",
-      "3910\n",
-      "3915\n",
-      "3920\n",
-      "3925\n",
-      "3930\n",
-      "3935\n",
-      "3940\n",
-      "3945\n",
-      "3950\n",
-      "3955\n",
-      "3960\n",
-      "3965\n",
-      "3970\n",
-      "3975\n",
-      "3980\n",
-      "3985\n",
-      "3990\n",
-      "3995\n",
-      "4000\n",
-      "4005\n",
-      "4010\n",
-      "4015\n",
-      "4020\n",
-      "4025\n",
-      "4030\n",
-      "4035\n",
-      "4040\n",
-      "4045\n",
-      "4050\n",
-      "4055\n",
-      "4060\n",
-      "4065\n",
-      "4070\n",
-      "4075\n",
-      "4080\n",
-      "4085\n",
-      "4090\n",
-      "4095\n",
-      "4100\n",
-      "4105\n",
-      "4110\n",
-      "4115\n",
-      "4120\n",
-      "4125\n",
-      "4130\n",
-      "4135\n",
-      "4140\n",
-      "4145\n",
-      "4150\n",
-      "4155\n",
-      "4160\n",
-      "4165\n",
-      "4170\n",
-      "4175\n",
-      "4180\n",
-      "4185\n",
-      "4190\n",
-      "4195\n",
-      "4200\n",
-      "4205\n",
-      "4210\n",
-      "4215\n",
-      "4220\n",
-      "4225\n",
-      "4230\n",
-      "4235\n",
-      "4240\n",
-      "4245\n",
-      "4250\n",
-      "4255\n",
-      "4260\n",
-      "4265\n",
-      "4270\n",
-      "4275\n",
-      "4280\n",
-      "4285\n",
-      "4290\n",
-      "4295\n",
-      "4300\n",
-      "4305\n",
-      "4310\n",
-      "4315\n",
-      "4320\n",
-      "4325\n",
-      "4330\n",
-      "4335\n",
-      "4340\n",
-      "4345\n",
-      "4350\n",
-      "4355\n",
-      "4360\n",
-      "4365\n",
-      "4370\n",
-      "4375\n",
-      "4380\n",
-      "4385\n",
-      "4390\n",
-      "4395\n",
-      "4400\n",
-      "4405\n",
-      "4410\n",
-      "4415\n",
-      "4420\n",
-      "4425\n",
-      "4430\n",
-      "4435\n",
-      "4440\n",
-      "4445\n",
-      "4450\n",
-      "4455\n",
-      "4460\n",
-      "4465\n",
-      "4470\n",
-      "4475\n",
-      "4480\n",
-      "4485\n",
-      "4490\n",
-      "4495\n",
-      "4500\n",
-      "4505\n",
-      "4510\n",
-      "4515\n",
-      "4520\n",
-      "4525\n",
-      "4530\n",
-      "4535\n",
-      "4540\n",
-      "4545\n",
-      "4550\n",
-      "4555\n",
-      "4560\n",
-      "4565\n",
-      "4570\n",
-      "4575\n",
-      "4580\n",
-      "4585\n",
-      "4590\n",
-      "4595\n",
-      "4600\n",
-      "4605\n",
-      "4610\n",
-      "4615\n",
-      "4620\n",
-      "4625\n",
-      "4630\n",
-      "4635\n",
-      "4640\n",
-      "4645\n",
-      "4650\n",
-      "4655\n",
-      "4660\n",
-      "4665\n",
-      "4670\n",
-      "4675\n",
-      "4680\n",
-      "4685\n",
-      "4690\n",
-      "4695\n",
-      "4700\n",
-      "4705\n",
-      "4710\n",
-      "4715\n",
-      "4720\n",
-      "4725\n",
-      "4730\n",
-      "4735\n",
-      "4740\n",
-      "4745\n",
-      "4750\n",
-      "4755\n",
-      "4760\n",
-      "4765\n",
-      "4770\n",
-      "4775\n",
-      "4780\n",
-      "4785\n",
-      "4790\n",
-      "4795\n",
-      "4800\n",
-      "4805\n",
-      "4810\n",
-      "4815\n",
-      "4820\n",
-      "4825\n",
-      "4830\n",
-      "4835\n",
-      "4840\n",
-      "4845\n",
-      "4850\n",
-      "4855\n",
-      "4860\n",
-      "4865\n",
-      "4870\n",
-      "4875\n",
-      "4880\n",
-      "4885\n",
-      "4890\n",
-      "4895\n",
-      "4900\n",
-      "4905\n",
-      "4910\n",
-      "4915\n",
-      "4920\n",
-      "4925\n",
-      "4930\n",
-      "4935\n",
-      "4940\n",
-      "4945\n",
-      "4950\n",
-      "4955\n",
-      "4960\n",
-      "4965\n",
-      "4970\n",
-      "4975\n",
-      "4980\n",
-      "4985\n",
-      "4990\n",
-      "4995\n",
-      "5000\n",
-      "5005\n",
-      "5010\n",
-      "5015\n",
-      "5020\n",
-      "5025\n",
-      "5030\n",
-      "5035\n",
-      "5040\n",
-      "5045\n",
-      "5050\n",
-      "5055\n",
-      "5060\n",
-      "5065\n",
-      "5070\n",
-      "5075\n",
-      "5080\n",
-      "5085\n",
-      "5090\n",
-      "5095\n",
-      "5100\n",
-      "5105\n",
-      "5110\n",
-      "5115\n",
-      "5120\n",
-      "5125\n",
-      "5130\n",
-      "5135\n",
-      "5140\n",
-      "5145\n",
-      "5150\n",
-      "5155\n",
-      "5160\n",
-      "5165\n",
-      "5170\n",
-      "5175\n",
-      "5180\n",
-      "5185\n",
-      "5190\n",
-      "5195\n",
-      "5200\n",
-      "5205\n",
-      "5210\n",
-      "5215\n",
-      "5220\n",
-      "5225\n",
-      "5230\n",
-      "5235\n",
-      "5240\n",
-      "5245\n",
-      "5250\n",
-      "5255\n",
-      "5260\n",
-      "5265\n",
-      "5270\n",
-      "5275\n",
-      "5280\n",
-      "5285\n",
-      "5290\n",
-      "5295\n",
-      "5300\n",
-      "5305\n",
-      "5310\n",
-      "5315\n",
-      "5320\n",
-      "5325\n",
-      "5330\n",
-      "5335\n",
-      "5340\n",
-      "5345\n",
-      "5350\n",
-      "5355\n",
-      "5360\n",
-      "5365\n",
-      "5370\n",
-      "5375\n",
-      "5380\n",
-      "5385\n",
-      "5390\n",
-      "5395\n",
-      "5400\n",
-      "5405\n",
-      "5410\n",
-      "5415\n",
-      "5420\n",
-      "5425\n",
-      "5430\n",
-      "5435\n",
-      "5440\n",
-      "5445\n",
-      "5450\n",
-      "5455\n",
-      "5460\n",
-      "5465\n",
-      "5470\n",
-      "5475\n",
-      "5480\n",
-      "5485\n",
-      "5490\n",
-      "5495\n",
-      "5500\n",
-      "5505\n",
-      "5510\n",
-      "5515\n",
-      "5520\n",
-      "5525\n",
-      "5530\n",
-      "5535\n",
-      "5540\n",
-      "5545\n",
-      "5550\n",
-      "5555\n",
-      "5560\n",
-      "5565\n",
-      "5570\n",
-      "5575\n",
-      "5580\n",
-      "5585\n",
-      "5590\n",
-      "5595\n",
-      "5600\n",
-      "5605\n",
-      "5610\n",
-      "5615\n",
-      "5620\n",
-      "5625\n",
-      "5630\n",
-      "5635\n",
-      "5640\n",
-      "5645\n",
-      "5650\n",
-      "5655\n",
-      "5660\n",
-      "5665\n",
-      "5670\n",
-      "5675\n",
-      "5680\n",
-      "5685\n",
-      "5690\n",
-      "5695\n",
-      "5700\n",
-      "5705\n",
-      "5710\n",
-      "5715\n",
-      "5720\n",
-      "5725\n",
-      "5730\n",
-      "5735\n",
-      "5740\n",
-      "5745\n",
-      "5750\n",
-      "5755\n",
-      "5760\n",
-      "5765\n",
-      "5770\n",
-      "5775\n",
-      "5780\n",
-      "5785\n",
-      "5790\n",
-      "5795\n",
-      "5800\n",
-      "5805\n",
-      "5810\n",
-      "5815\n",
-      "5820\n",
-      "5825\n",
-      "5830\n",
-      "5835\n",
-      "5840\n",
-      "5845\n",
-      "5850\n",
-      "5855\n",
-      "5860\n",
-      "5865\n",
-      "5870\n",
-      "5875\n",
-      "5880\n",
-      "5885\n",
-      "5890\n",
-      "5895\n",
-      "5900\n",
-      "5905\n",
-      "5910\n",
-      "5915\n",
-      "5920\n",
-      "5925\n",
-      "5930\n",
-      "5935\n",
-      "5940\n",
-      "5945\n",
-      "5950\n",
-      "5955\n",
-      "5960\n",
-      "5965\n",
-      "5970\n",
-      "5975\n",
-      "5980\n",
-      "5985\n",
-      "5990\n",
-      "5995\n",
-      "6000\n",
-      "6005\n",
-      "6010\n",
-      "6015\n",
-      "6020\n",
-      "6025\n",
-      "6030\n",
-      "6035\n",
-      "6040\n",
-      "6045\n",
-      "6050\n",
-      "6055\n",
-      "6060\n",
-      "6065\n",
-      "6070\n",
-      "6075\n",
-      "6080\n",
-      "6085\n",
-      "6090\n",
-      "6095\n",
-      "6100\n",
-      "6105\n",
-      "6110\n",
-      "6115\n",
-      "6120\n",
-      "6125\n",
-      "6130\n",
-      "6135\n",
-      "6140\n",
-      "6145\n",
-      "6150\n",
-      "6155\n",
-      "6160\n",
-      "6165\n",
-      "6170\n",
-      "6175\n",
-      "6180\n",
-      "6185\n",
-      "6190\n",
-      "6195\n",
-      "6200\n",
-      "6205\n",
-      "6210\n",
-      "6215\n",
-      "6220\n",
-      "6225\n",
-      "6230\n",
-      "6235\n",
-      "6240\n",
-      "6245\n",
-      "6250\n",
-      "6255\n",
-      "6260\n",
-      "6265\n",
-      "6270\n",
-      "6275\n",
-      "6280\n",
-      "6285\n",
-      "6290\n",
-      "6295\n",
-      "6300\n",
-      "6305\n",
-      "6310\n",
-      "6315\n",
-      "6320\n",
-      "6325\n",
-      "6330\n",
-      "6335\n",
-      "6340\n",
-      "6345\n",
-      "6350\n",
-      "6355\n",
-      "6360\n",
-      "6365\n",
-      "6370\n",
-      "6375\n",
-      "6380\n",
-      "6385\n",
-      "6390\n",
-      "6395\n",
-      "6400\n",
-      "6405\n",
-      "6410\n",
-      "6415\n",
-      "6420\n",
-      "6425\n",
-      "6430\n",
-      "6435\n",
-      "6440\n",
-      "6445\n",
-      "6450\n",
-      "6455\n",
-      "6460\n",
-      "6465\n",
-      "6470\n",
-      "6475\n",
-      "6480\n",
-      "6485\n",
-      "6490\n",
-      "6495\n",
-      "6500\n",
-      "6505\n",
-      "6510\n",
-      "6515\n",
-      "6520\n",
-      "6525\n",
-      "6530\n",
-      "6535\n",
-      "6540\n",
-      "6545\n",
-      "6550\n",
-      "6555\n",
-      "6560\n",
-      "6565\n",
-      "6570\n",
-      "6575\n",
-      "6580\n",
-      "6585\n",
-      "6590\n",
-      "6595\n",
-      "6600\n",
-      "6605\n",
-      "6610\n",
-      "6615\n",
-      "6620\n",
-      "6625\n",
-      "6630\n",
-      "6635\n",
-      "6640\n",
-      "6645\n",
-      "6650\n",
-      "6655\n",
-      "6660\n",
-      "6665\n",
-      "6670\n",
-      "6675\n",
-      "6680\n",
-      "6685\n",
-      "6690\n",
-      "6695\n",
-      "6700\n",
-      "6705\n",
-      "6710\n",
-      "6715\n",
-      "6720\n",
-      "6725\n",
-      "6730\n",
-      "6735\n",
-      "6740\n",
-      "6745\n",
-      "6750\n",
-      "6755\n",
-      "6760\n",
-      "6765\n",
-      "6770\n",
-      "6775\n",
-      "6780\n",
-      "6785\n",
-      "6790\n",
-      "6795\n",
-      "6800\n",
-      "6805\n",
-      "6810\n",
-      "6815\n",
-      "6820\n",
-      "6825\n",
-      "6830\n",
-      "6835\n",
-      "6840\n",
-      "6845\n",
-      "6850\n",
-      "6855\n",
-      "6860\n",
-      "6865\n",
-      "6870\n",
-      "6875\n",
-      "6880\n",
-      "6885\n",
-      "6890\n",
-      "6895\n",
-      "6900\n",
-      "6905\n",
-      "6910\n",
-      "6915\n",
-      "6920\n",
-      "6925\n",
-      "6930\n",
-      "6935\n",
-      "6940\n",
-      "6945\n",
-      "6950\n",
-      "6955\n",
-      "6960\n",
-      "6965\n",
-      "6970\n",
-      "6975\n",
-      "6980\n",
-      "6985\n",
-      "6990\n",
-      "6995\n",
-      "7000\n",
-      "7005\n",
-      "7010\n",
-      "7015\n",
-      "7020\n",
-      "7025\n",
-      "7030\n",
-      "7035\n",
-      "7040\n",
-      "7045\n",
-      "7050\n",
-      "7055\n",
-      "7060\n",
-      "7065\n",
-      "7070\n",
-      "7075\n",
-      "7080\n",
-      "7085\n",
-      "7090\n",
-      "7095\n",
-      "7100\n",
-      "7105\n",
-      "7110\n",
-      "7115\n",
-      "7120\n",
-      "7125\n",
-      "7130\n",
-      "7135\n",
-      "7140\n",
-      "7145\n",
-      "7150\n",
-      "7155\n",
-      "7160\n",
-      "7165\n",
-      "7170\n",
-      "7175\n",
-      "7180\n",
-      "7185\n",
-      "7190\n",
-      "7195\n",
-      "7200\n",
-      "7205\n",
-      "7210\n",
-      "7215\n",
-      "7220\n",
-      "7225\n",
-      "7230\n",
-      "7235\n",
-      "7240\n",
-      "7245\n",
-      "7250\n",
-      "7255\n",
-      "7260\n",
-      "7265\n",
-      "7270\n",
-      "7275\n",
-      "7280\n",
-      "7285\n",
-      "7290\n",
-      "7295\n",
-      "7300\n",
-      "7305\n",
-      "7310\n",
-      "7315\n",
-      "7320\n",
-      "7325\n",
-      "7330\n",
-      "7335\n",
-      "7340\n",
-      "7345\n",
-      "7350\n",
-      "7355\n",
-      "7360\n",
-      "7365\n",
-      "7370\n",
-      "7375\n",
-      "7380\n",
-      "7385\n",
-      "7390\n",
-      "7395\n",
-      "7400\n",
-      "7405\n",
-      "7410\n",
-      "7415\n",
-      "7420\n",
-      "7425\n",
-      "7430\n",
-      "7435\n",
-      "7440\n",
-      "7445\n",
-      "7450\n",
-      "7455\n",
-      "7460\n",
-      "7465\n",
-      "7470\n",
-      "7475\n",
-      "7480\n",
-      "7485\n",
-      "7490\n",
-      "7495\n",
-      "7500\n",
-      "7505\n",
-      "7510\n",
-      "7515\n",
-      "7520\n",
-      "7525\n",
-      "7530\n",
-      "7535\n",
-      "7540\n",
-      "7545\n",
-      "7550\n",
-      "7555\n",
-      "7560\n",
-      "7565\n",
-      "7570\n",
-      "7575\n",
-      "7580\n",
-      "7585\n",
-      "7590\n",
-      "7595\n",
-      "7600\n",
-      "7605\n",
-      "7610\n",
-      "7615\n",
-      "7620\n",
-      "7625\n",
-      "7630\n",
-      "7635\n",
-      "7640\n",
-      "7645\n",
-      "7650\n",
-      "7655\n",
-      "7660\n",
-      "7665\n",
-      "7670\n",
-      "7675\n",
-      "7680\n",
-      "7685\n",
-      "7690\n",
-      "7695\n",
-      "7700\n",
-      "7705\n",
-      "7710\n",
-      "7715\n",
-      "7720\n",
-      "7725\n",
-      "7730\n",
-      "7735\n",
-      "7740\n",
-      "7745\n",
-      "7750\n",
-      "7755\n",
-      "7760\n",
-      "7765\n",
-      "7770\n",
-      "7775\n",
-      "7780\n",
-      "7785\n",
-      "7790\n",
-      "7795\n",
-      "7800\n",
-      "7805\n",
-      "7810\n",
-      "7815\n",
-      "7820\n",
-      "7825\n",
-      "7830\n",
-      "7835\n",
-      "7840\n",
-      "7845\n",
-      "7850\n",
-      "7855\n",
-      "7860\n",
-      "7865\n",
-      "7870\n",
-      "7875\n",
-      "7880\n",
-      "7885\n",
-      "7890\n",
-      "7895\n",
-      "7900\n",
-      "7905\n",
-      "7910\n",
-      "7915\n",
-      "7920\n",
-      "7925\n",
-      "7930\n",
-      "7935\n",
-      "7940\n",
-      "7945\n",
-      "7950\n",
-      "7955\n",
-      "7960\n",
-      "7965\n",
-      "7970\n",
-      "7975\n",
-      "7980\n",
-      "7985\n",
-      "7990\n",
-      "7995\n",
-      "8000\n",
-      "8005\n",
-      "8010\n",
-      "8015\n",
-      "8020\n",
-      "8025\n",
-      "8030\n",
-      "8035\n",
-      "8040\n",
-      "8045\n",
-      "8050\n",
-      "8055\n",
-      "8060\n",
-      "8065\n",
-      "8070\n",
-      "8075\n",
-      "8080\n",
-      "8085\n",
-      "8090\n",
-      "8095\n",
-      "8100\n",
-      "8105\n",
-      "8110\n",
-      "8115\n",
-      "8120\n",
-      "8125\n",
-      "8130\n",
-      "8135\n",
-      "8140\n",
-      "8145\n",
-      "8150\n",
-      "8155\n",
-      "8160\n",
-      "8165\n",
-      "8170\n",
-      "8175\n",
-      "8180\n",
-      "8185\n",
-      "8190\n",
-      "8195\n",
-      "8200\n",
-      "8205\n",
-      "8210\n",
-      "8215\n",
-      "8220\n",
-      "8225\n",
-      "8230\n",
-      "8235\n",
-      "8240\n",
-      "8245\n",
-      "8250\n",
-      "8255\n",
-      "8260\n",
-      "8265\n",
-      "8270\n",
-      "8275\n",
-      "8280\n",
-      "8285\n",
-      "8290\n",
-      "8295\n",
-      "8300\n",
-      "8305\n",
-      "8310\n",
-      "8315\n",
-      "8320\n",
-      "8325\n",
-      "8330\n",
-      "8335\n",
-      "8340\n",
-      "8345\n",
-      "8350\n",
-      "8355\n",
-      "8360\n",
-      "8365\n",
-      "8370\n",
-      "8375\n",
-      "8380\n",
-      "8385\n",
-      "8390\n",
-      "8395\n",
-      "8400\n",
-      "8405\n",
-      "8410\n",
-      "8415\n",
-      "8420\n",
-      "8425\n",
-      "8430\n",
-      "8435\n",
-      "8440\n",
-      "8445\n",
-      "8450\n",
-      "8455\n",
-      "8460\n",
-      "8465\n",
-      "8470\n",
-      "8475\n",
-      "8480\n",
-      "8485\n",
-      "8490\n",
-      "8495\n",
-      "8500\n",
-      "8505\n",
-      "8510\n",
-      "8515\n",
-      "8520\n",
-      "8525\n",
-      "8530\n",
-      "8535\n",
-      "8540\n",
-      "8545\n",
-      "8550\n",
-      "8555\n",
-      "8560\n",
-      "8565\n",
-      "8570\n",
-      "8575\n",
-      "8580\n",
-      "8585\n",
-      "8590\n",
-      "8595\n",
-      "8600\n",
-      "8605\n",
-      "8610\n",
-      "8615\n",
-      "8620\n",
-      "8625\n",
-      "8630\n",
-      "8635\n",
-      "8640\n",
-      "8645\n",
-      "8650\n",
-      "8655\n",
-      "8660\n",
-      "8665\n",
-      "8670\n",
-      "8675\n",
-      "8680\n",
-      "8685\n",
-      "8690\n",
-      "8695\n",
-      "8700\n",
-      "8705\n",
-      "8710\n",
-      "8715\n",
-      "8720\n",
-      "8725\n",
-      "8730\n",
-      "8735\n",
-      "8740\n",
-      "8745\n",
-      "8750\n",
-      "8755\n",
-      "8760\n",
-      "8765\n",
-      "8770\n",
-      "8775\n",
-      "8780\n",
-      "8785\n",
-      "8790\n",
-      "8795\n",
-      "8800\n",
-      "8805\n",
-      "8810\n",
-      "8815\n",
-      "8820\n",
-      "8825\n",
-      "8830\n",
-      "8835\n",
-      "8840\n",
-      "8845\n",
-      "8850\n",
-      "8855\n",
-      "8860\n",
-      "8865\n",
-      "8870\n",
-      "8875\n",
-      "8880\n",
-      "8885\n",
-      "8890\n",
-      "8895\n",
-      "8900\n",
-      "8905\n",
-      "8910\n",
-      "8915\n",
-      "8920\n",
-      "8925\n",
-      "8930\n",
-      "8935\n",
-      "8940\n",
-      "8945\n",
-      "8950\n",
-      "8955\n",
-      "8960\n",
-      "8965\n",
-      "8970\n",
-      "8975\n",
-      "8980\n",
-      "8985\n",
-      "8990\n",
-      "8995\n",
-      "9000\n",
-      "9005\n",
-      "9010\n",
-      "9015\n",
-      "9020\n",
-      "9025\n",
-      "9030\n",
-      "9035\n",
-      "9040\n",
-      "9045\n",
-      "9050\n",
-      "9055\n",
-      "9060\n",
-      "9065\n",
-      "9070\n",
-      "9075\n",
-      "9080\n",
-      "9085\n",
-      "9090\n",
-      "9095\n",
-      "9100\n",
-      "9105\n",
-      "9110\n",
-      "9115\n",
-      "9120\n",
-      "9125\n",
-      "9130\n",
-      "9135\n",
-      "9140\n",
-      "9145\n",
-      "9150\n",
-      "9155\n",
-      "9160\n",
-      "9165\n",
-      "9170\n",
-      "9175\n",
-      "9180\n",
-      "9185\n",
-      "9190\n",
-      "9195\n",
-      "9200\n",
-      "9205\n",
-      "9210\n",
-      "9215\n",
-      "9220\n",
-      "9225\n",
-      "9230\n",
-      "9235\n",
-      "9240\n",
-      "9245\n",
-      "9250\n",
-      "9255\n",
-      "9260\n",
-      "9265\n",
-      "9270\n",
-      "9275\n",
-      "9280\n",
-      "9285\n",
-      "9290\n",
-      "9295\n",
-      "9300\n",
-      "9305\n",
-      "9310\n",
-      "9315\n",
-      "9320\n",
-      "9325\n",
-      "9330\n",
-      "9335\n",
-      "9340\n",
-      "9345\n",
-      "9350\n",
-      "9355\n",
-      "9360\n",
-      "9365\n",
-      "9370\n",
-      "9375\n",
-      "9380\n",
-      "9385\n",
-      "9390\n",
-      "9395\n",
-      "9400\n",
-      "9405\n",
-      "9410\n",
-      "9415\n",
-      "9420\n",
-      "9425\n",
-      "9430\n",
-      "9435\n",
-      "9440\n",
-      "9445\n",
-      "9450\n",
-      "9455\n",
-      "9460\n",
-      "9465\n",
-      "9470\n",
-      "9475\n",
-      "9480\n",
-      "9485\n",
-      "9490\n",
-      "9495\n",
-      "9500\n",
-      "9505\n",
-      "9510\n",
-      "9515\n",
-      "9520\n",
-      "9525\n",
-      "9530\n",
-      "9535\n",
-      "9540\n",
-      "9545\n",
-      "9550\n",
-      "9555\n",
-      "9560\n",
-      "9565\n",
-      "9570\n",
-      "9575\n",
-      "9580\n",
-      "9585\n",
-      "9590\n",
-      "9595\n",
-      "9600\n",
-      "9605\n",
-      "9610\n",
-      "9615\n",
-      "9620\n",
-      "9625\n",
-      "9630\n",
-      "9635\n",
-      "9640\n",
-      "9645\n",
-      "9650\n",
-      "9655\n",
-      "9660\n",
-      "9665\n",
-      "9670\n",
-      "9675\n",
-      "9680\n",
-      "9685\n",
-      "9690\n",
-      "9695\n",
-      "9700\n",
-      "9705\n",
-      "9710\n",
-      "9715\n",
-      "9720\n",
-      "9725\n",
-      "9730\n",
-      "9735\n",
-      "9740\n",
-      "9745\n",
-      "9750\n",
-      "9755\n",
-      "9760\n",
-      "9765\n",
-      "9770\n",
-      "9775\n",
-      "9780\n",
-      "9785\n",
-      "9790\n",
-      "9795\n",
-      "9800\n",
-      "9805\n",
-      "9810\n",
-      "9815\n",
-      "9820\n",
-      "9825\n",
-      "9830\n",
-      "9835\n",
-      "9840\n",
-      "9845\n",
-      "9850\n",
-      "9855\n",
-      "9860\n",
-      "9865\n",
-      "9870\n",
-      "9875\n",
-      "9880\n",
-      "9885\n",
-      "9890\n",
-      "9895\n",
-      "9900\n",
-      "9905\n",
-      "9910\n",
-      "9915\n",
-      "9920\n",
-      "9925\n",
-      "9930\n",
-      "9935\n",
-      "9940\n",
-      "9945\n",
-      "9950\n",
-      "9955\n",
-      "9960\n",
-      "9965\n",
-      "9970\n",
-      "9975\n",
-      "9980\n",
-      "9985\n",
-      "9990\n",
-      "9995\n",
-      "10000\n",
-      "10005\n",
-      "10010\n",
-      "10015\n",
-      "10020\n",
-      "10025\n",
-      "10030\n",
-      "10035\n",
-      "10040\n",
-      "10045\n",
-      "10050\n",
-      "10055\n",
-      "10060\n",
-      "10065\n",
-      "10070\n",
-      "10075\n",
-      "10080\n",
-      "10085\n",
-      "10090\n",
-      "10095\n",
-      "10100\n",
-      "10105\n",
-      "10110\n",
-      "10115\n",
-      "10120\n",
-      "10125\n",
-      "10130\n",
-      "10135\n",
-      "10140\n",
-      "10145\n",
-      "10150\n",
-      "10155\n",
-      "10160\n",
-      "10165\n",
-      "10170\n",
-      "10175\n",
-      "10180\n",
-      "10185\n",
-      "10190\n",
-      "10195\n",
-      "10200\n",
-      "10205\n",
-      "10210\n",
-      "10215\n",
-      "10220\n",
-      "10225\n",
-      "10230\n",
-      "10235\n",
-      "10240\n",
-      "10245\n",
-      "10250\n",
-      "10255\n",
-      "10260\n",
-      "10265\n",
-      "10270\n",
-      "10275\n",
-      "10280\n",
-      "10285\n",
-      "10290\n",
-      "10295\n",
-      "10300\n",
-      "10305\n",
-      "10310\n",
-      "10315\n",
-      "10320\n",
-      "10325\n",
-      "10330\n",
-      "10335\n",
-      "10340\n",
-      "10345\n",
-      "10350\n",
-      "10355\n",
-      "10360\n",
-      "10365\n",
-      "10370\n",
-      "10375\n",
-      "10380\n",
-      "10385\n",
-      "10390\n",
-      "10395\n",
-      "10400\n",
-      "10405\n",
-      "10410\n",
-      "10415\n",
-      "10420\n",
-      "10425\n",
-      "10430\n",
-      "10435\n",
-      "10440\n",
-      "10445\n",
-      "10450\n",
-      "10455\n",
-      "10460\n",
-      "10465\n",
-      "10470\n",
-      "10475\n",
-      "10480\n",
-      "10485\n",
-      "10490\n",
-      "10495\n",
-      "10500\n",
-      "10505\n",
-      "10510\n",
-      "10515\n",
-      "10520\n",
-      "10525\n",
-      "10530\n",
-      "10535\n",
-      "10540\n",
-      "10545\n",
-      "10550\n",
-      "10555\n",
-      "10560\n",
-      "10565\n",
-      "10570\n",
-      "10575\n",
-      "10580\n",
-      "10585\n",
-      "10590\n",
-      "10595\n",
-      "10600\n",
-      "10605\n",
-      "10610\n",
-      "10615\n",
-      "10620\n",
-      "10625\n",
-      "10630\n",
-      "10635\n",
-      "10640\n",
-      "10645\n",
-      "10650\n",
-      "10655\n",
-      "10660\n",
-      "10665\n",
-      "10670\n",
-      "10675\n",
-      "10680\n",
-      "10685\n",
-      "10690\n",
-      "10695\n",
-      "10700\n",
-      "10705\n",
-      "10710\n",
-      "10715\n",
-      "10720\n",
-      "10725\n",
-      "10730\n",
-      "10735\n",
-      "10740\n",
-      "10745\n",
-      "10750\n",
-      "10755\n",
-      "10760\n",
-      "10765\n",
-      "10770\n",
-      "10775\n",
-      "10780\n",
-      "10785\n",
-      "10790\n",
-      "10795\n",
-      "10800\n",
-      "10805\n",
-      "10810\n",
-      "10815\n",
-      "10820\n",
-      "10825\n",
-      "10830\n",
-      "10835\n",
-      "10840\n",
-      "10845\n",
-      "10850\n",
-      "10855\n",
-      "10860\n",
-      "10865\n",
-      "10870\n",
-      "10875\n",
-      "10880\n",
-      "10885\n",
-      "10890\n",
-      "10895\n",
-      "10900\n",
-      "10905\n",
-      "10910\n",
-      "10915\n",
-      "10920\n",
-      "10925\n",
-      "10930\n",
-      "10935\n",
-      "10940\n",
-      "10945\n",
-      "10950\n",
-      "10955\n",
-      "10960\n",
-      "10965\n",
-      "10970\n",
-      "10975\n",
-      "10980\n",
-      "10985\n",
-      "10990\n",
-      "10995\n",
-      "11000\n",
-      "11005\n",
-      "11010\n",
-      "11015\n",
-      "11020\n",
-      "11025\n",
-      "11030\n",
-      "11035\n",
-      "11040\n",
-      "11045\n",
-      "11050\n",
-      "11055\n",
-      "11060\n",
-      "11065\n",
-      "11070\n",
-      "11075\n",
-      "11080\n",
-      "11085\n",
-      "11090\n",
-      "11095\n",
-      "11100\n",
-      "11105\n",
-      "11110\n",
-      "11115\n",
-      "11120\n",
-      "11125\n",
-      "11130\n",
-      "11135\n",
-      "11140\n",
-      "11145\n",
-      "11150\n",
-      "11155\n",
-      "11160\n",
-      "11165\n",
-      "11170\n",
-      "11175\n",
-      "11180\n",
-      "11185\n",
-      "11190\n",
-      "11195\n",
-      "11200\n",
-      "11205\n",
-      "11210\n",
-      "11215\n",
-      "11220\n",
-      "11225\n",
-      "11230\n",
-      "11235\n",
-      "11240\n",
-      "11245\n",
-      "11250\n",
-      "11255\n",
-      "11260\n",
-      "11265\n",
-      "11270\n",
-      "11275\n",
-      "11280\n",
-      "11285\n",
-      "11290\n",
-      "11295\n",
-      "11300\n",
-      "11305\n",
-      "11310\n",
-      "11315\n",
-      "11320\n",
-      "11325\n",
-      "11330\n",
-      "11335\n",
-      "11340\n",
-      "11345\n",
-      "11350\n",
-      "11355\n",
-      "11360\n",
-      "11365\n",
-      "11370\n",
-      "11375\n",
-      "11380\n",
-      "11385\n",
-      "11390\n",
-      "11395\n",
-      "11400\n",
-      "11405\n",
-      "11410\n",
-      "11415\n",
-      "11420\n",
-      "11425\n",
-      "11430\n",
-      "11435\n",
-      "11440\n",
-      "11445\n",
-      "11450\n",
-      "11455\n",
-      "11460\n",
-      "11465\n",
-      "11470\n",
-      "11475\n",
-      "11480\n",
-      "11485\n",
-      "11490\n",
-      "11495\n",
-      "11500\n",
-      "11505\n",
-      "11510\n",
-      "11515\n",
-      "11520\n",
-      "11525\n",
-      "11530\n",
-      "11535\n",
-      "11540\n",
-      "11545\n",
-      "11550\n",
-      "11555\n",
-      "11560\n",
-      "11565\n",
-      "11570\n",
-      "11575\n",
-      "11580\n",
-      "11585\n",
-      "11590\n",
-      "11595\n",
-      "11600\n",
-      "11605\n",
-      "11610\n",
-      "11615\n",
-      "11620\n",
-      "11625\n",
-      "11630\n",
-      "11635\n",
-      "11640\n",
-      "11645\n",
-      "11650\n",
-      "11655\n",
-      "11660\n",
-      "11665\n",
-      "11670\n",
-      "11675\n",
-      "11680\n",
-      "11685\n",
-      "11690\n",
-      "11695\n",
-      "11700\n",
-      "11705\n",
-      "11710\n",
-      "11715\n",
-      "11720\n",
-      "11725\n",
-      "11730\n",
-      "11735\n",
-      "11740\n",
-      "11745\n",
-      "11750\n",
-      "11755\n",
-      "11760\n",
-      "11765\n",
-      "11770\n",
-      "11775\n",
-      "11780\n",
-      "11785\n",
-      "11790\n",
-      "11795\n",
-      "11800\n",
-      "11805\n",
-      "11810\n",
-      "11815\n",
-      "11820\n",
-      "11825\n",
-      "11830\n",
-      "11835\n",
-      "11840\n",
-      "11845\n",
-      "11850\n",
-      "11855\n",
-      "11860\n",
-      "11865\n",
-      "11870\n",
-      "11875\n",
-      "11880\n",
-      "11885\n",
-      "11890\n",
-      "11895\n",
-      "11900\n",
-      "11905\n",
-      "11910\n",
-      "11915\n",
-      "11920\n",
-      "11925\n",
-      "11930\n",
-      "11935\n",
-      "11940\n",
-      "11945\n",
-      "11950\n",
-      "11955\n",
-      "11960\n",
-      "11965\n",
-      "11970\n",
-      "11975\n",
-      "11980\n",
-      "11985\n",
-      "11990\n",
-      "11995\n",
-      "12000\n",
-      "12005\n",
-      "12010\n",
-      "12015\n",
-      "12020\n"
-     ]
-    }
-   ],
-   "source": [
-    "# test_ds = get_val_dataset()\n",
-    "# # Run evaluation\n",
-    "# print(test_ds)\n",
-    "generate(\n",
-    "        model=generator,\n",
-    "        dataset=train_ds,\n",
-    "        epoch=150  # change if you want more rows\n",
-    "    )"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "id": "a1bbabe5",
-   "metadata": {
-    "papermill": {
-     "duration": 0.191322,
-     "end_time": "2026-02-12T22:12:22.615266",
-     "exception": false,
-     "start_time": "2026-02-12T22:12:22.423944",
-     "status": "completed"
-    },
-    "tags": []
-   },
-   "outputs": [],
-   "source": []
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "id": "619ef3ab",
-   "metadata": {
-    "papermill": {
-     "duration": 0.201882,
-     "end_time": "2026-02-12T22:12:23.012230",
-     "exception": false,
-     "start_time": "2026-02-12T22:12:22.810348",
-     "status": "completed"
-    },
-    "tags": []
-   },
-   "outputs": [],
-   "source": []
-  }
- ],
- "metadata": {
-  "kaggle": {
-   "accelerator": "gpu",
-   "dataSources": [
-    {
-     "datasetId": 8436032,
-     "sourceId": 13308561,
-     "sourceType": "datasetVersion"
-    },
-    {
-     "isSourceIdPinned": false,
-     "modelId": 583820,
-     "modelInstanceId": 571510,
-     "sourceId": 748364,
-     "sourceType": "modelInstanceVersion"
-    },
-    {
-     "isSourceIdPinned": true,
-     "modelId": 584139,
-     "modelInstanceId": 571806,
-     "sourceId": 749667,
-     "sourceType": "modelInstanceVersion"
-    }
-   ],
-   "dockerImageVersionId": 31154,
-   "isGpuEnabled": true,
-   "isInternetEnabled": false,
-   "language": "python",
-   "sourceType": "notebook"
-  },
-  "kernelspec": {
-   "display_name": "Python 3",
-   "language": "python",
-   "name": "python3"
-  },
-  "language_info": {
-   "codemirror_mode": {
-    "name": "ipython",
-    "version": 3
-   },
-   "file_extension": ".py",
-   "mimetype": "text/x-python",
-   "name": "python",
-   "nbconvert_exporter": "python",
-   "pygments_lexer": "ipython3",
-   "version": "3.11.13"
-  },
-  "papermill": {
-   "default_parameters": {},
-   "duration": 828.406063,
-   "end_time": "2026-02-12T22:12:26.638920",
-   "environment_variables": {},
-   "exception": null,
-   "input_path": "__notebook__.ipynb",
-   "output_path": "__notebook__.ipynb",
-   "parameters": {},
-   "start_time": "2026-02-12T21:58:38.232857",
-   "version": "2.6.0"
-  }
- },
- "nbformat": 4,
- "nbformat_minor": 5
-}