{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "d8cbcb34-b7a9-4bd3-af85-6263ac3ee83d",
   "metadata": {},
   "source": [
    "# Training 3D U-Net model for multi-class semantic segmentation"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "936c827d-9270-4137-8049-3b2b8845f6f1",
   "metadata": {},
   "source": [
    "### Install pkgs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "922dd5d3-7999-47b2-affe-de4794faa2ed",
   "metadata": {},
   "outputs": [],
   "source": [
    "#!pip install git+https://github.com/copick/copick-utils.git matplotlib tqdm copick \n",
    "#!pip install -q \"monai-weekly[mlflow]\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6148c6d2",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Make a copick project\n",
    "import os\n",
    "import shutil\n",
    "\n",
    "config_blob = \"\"\"{\n",
    "    \"name\": \"czii_cryoet_mlchallenge_2024\",\n",
    "    \"description\": \"2024 CZII CryoET ML Challenge training data.\",\n",
    "    \"version\": \"1.0.0\",\n",
    "\n",
    "    \"pickable_objects\": [\n",
    "        {\n",
    "            \"name\": \"apo-ferritin\",\n",
    "            \"is_particle\": true,\n",
    "            \"pdb_id\": \"4V1W\",\n",
    "            \"label\": 1,\n",
    "            \"color\": [  0, 117, 220, 128],\n",
    "            \"radius\": 60,\n",
    "            \"map_threshold\": 0.0418\n",
    "        },\n",
    "        {\n",
    "            \"name\": \"beta-galactosidase\",\n",
    "            \"is_particle\": true,\n",
    "            \"pdb_id\": \"6X1Q\",\n",
    "            \"label\": 3,\n",
    "            \"color\": [ 76,   0,  92, 128],\n",
    "            \"radius\": 90,\n",
    "            \"map_threshold\": 0.0578\n",
    "        },\n",
    "        {\n",
    "            \"name\": \"ribosome\",\n",
    "            \"is_particle\": true,\n",
    "            \"pdb_id\": \"6EK0\",\n",
    "            \"label\": 4,\n",
    "            \"color\": [  0,  92,  49, 128],\n",
    "            \"radius\": 150,\n",
    "            \"map_threshold\": 0.0374\n",
    "        },\n",
    "        {\n",
    "            \"name\": \"thyroglobulin\",\n",
    "            \"is_particle\": true,\n",
    "            \"pdb_id\": \"6SCJ\",\n",
    "            \"label\": 5,\n",
    "            \"color\": [ 43, 206,  72, 128],\n",
    "            \"radius\": 130,\n",
    "            \"map_threshold\": 0.0278\n",
    "        },\n",
    "        {\n",
    "            \"name\": \"virus-like-particle\",\n",
    "            \"is_particle\": true,\n",
    "            \"label\": 6,\n",
    "            \"color\": [255, 204, 153, 128],\n",
    "            \"radius\": 135,\n",
    "            \"map_threshold\": 0.201\n",
    "        },\n",
    "        {\n",
    "            \"name\": \"membrane\",\n",
    "            \"is_particle\": false,\n",
    "            \"label\": 8,\n",
    "            \"color\": [100, 100, 100, 128]\n",
    "        },\n",
    "        {\n",
    "            \"name\": \"background\",\n",
    "            \"is_particle\": false,\n",
    "            \"label\": 9,\n",
    "            \"color\": [10, 150, 200, 128]\n",
    "        }\n",
    "    ],\n",
    "\n",
    "    \"overlay_root\": \"/kaggle/working/overlay\",\n",
    "\n",
    "    \"overlay_fs_args\": {\n",
    "        \"auto_mkdir\": true\n",
    "    },\n",
    "\n",
    "    \"static_root\": \"/kaggle/input/czii-cryo-et-object-identification/train/static\"\n",
    "}\"\"\"\n",
    "\n",
    "copick_config_path = \"/kaggle/working/copick.config\"\n",
    "output_overlay = \"/kaggle/working/overlay\"\n",
    "\n",
    "with open(copick_config_path, \"w\") as f:\n",
    "    f.write(config_blob)\n",
    "    \n",
    "# Update the overlay\n",
    "# Define source and destination directories\n",
    "source_dir = '/kaggle/input/czii-cryo-et-object-identification/train/overlay'\n",
    "destination_dir = '/kaggle/working/overlay'\n",
    "\n",
    "# Walk through the source directory\n",
    "for root, dirs, files in os.walk(source_dir):\n",
    "    # Create corresponding subdirectories in the destination\n",
    "    relative_path = os.path.relpath(root, source_dir)\n",
    "    target_dir = os.path.join(destination_dir, relative_path)\n",
    "    os.makedirs(target_dir, exist_ok=True)\n",
    "    \n",
    "    # Copy and rename each file\n",
    "    for file in files:\n",
    "        if file.startswith(\"curation_0_\"):\n",
    "            new_filename = file\n",
    "        else:\n",
    "            new_filename = f\"curation_0_{file}\"\n",
    "            \n",
    "        \n",
    "        # Define full paths for the source and destination files\n",
    "        source_file = os.path.join(root, file)\n",
    "        destination_file = os.path.join(target_dir, new_filename)\n",
    "        \n",
    "        # Copy the file with the new name\n",
    "        shutil.copy2(source_file, destination_file)\n",
    "        print(f\"Copied {source_file} to {destination_file}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "277b5d8e-9a75-4e24-a8d1-4b99ef32b39a",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import numpy as np\n",
    "from pathlib import Path\n",
    "import torch\n",
    "import torchinfo\n",
    "import zarr, copick\n",
    "from tqdm import tqdm\n",
    "from monai.data import DataLoader, Dataset, CacheDataset, decollate_batch\n",
    "from monai.transforms import (\n",
    "    Compose, \n",
    "    EnsureChannelFirstd, \n",
    "    Orientationd,  \n",
    "    AsDiscrete,  \n",
    "    RandFlipd, \n",
    "    RandRotate90d, \n",
    "    NormalizeIntensityd,\n",
    "    RandCropByLabelClassesd,\n",
    ")\n",
    "from monai.networks.nets import UNet\n",
    "from monai.losses import DiceLoss, FocalLoss, TverskyLoss\n",
    "from monai.metrics import DiceMetric, ConfusionMatrixMetric\n",
    "import mlflow\n",
    "import mlflow.pytorch"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4724011a-eb1e-4ff1-8451-fb9aac29550a",
   "metadata": {},
   "source": [
    "## Prepare the dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9126f6f4-25f6-4c02-868e-c2ffeb81ae9b",
   "metadata": {},
   "source": [
    "### 1. Get copick root"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "8cd3bff4-7135-4fa2-bb11-6f805603f71a",
   "metadata": {},
   "outputs": [],
   "source": [
    "root = copick.from_file(copick_config_path)\n",
    "\n",
    "copick_user_name = \"copickUtils\"\n",
    "copick_segmentation_name = \"paintedPicks\"\n",
    "voxel_size = 10\n",
    "tomo_type = \"denoised\""
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b2ee1fd6-a107-4044-a03e-2dc8a77e86e2",
   "metadata": {},
   "source": [
    "### 2. Generate multi-class segmentation masks from picks, and saved them to the copick overlay directory (one-time)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "ab607ddb-3c61-44f0-89c4-9215842ba996",
   "metadata": {},
   "outputs": [],
   "source": [
    "from copick_utils.segmentation import segmentation_from_picks\n",
    "import copick_utils.writers.write as write\n",
    "from collections import defaultdict\n",
    "\n",
    "# Just do this once\n",
    "generate_masks = True\n",
    "\n",
    "if generate_masks:\n",
    "    target_objects = defaultdict(dict)\n",
    "    for object in root.pickable_objects:\n",
    "        if object.is_particle:\n",
    "            target_objects[object.name]['label'] = object.label\n",
    "            target_objects[object.name]['radius'] = object.radius\n",
    "\n",
    "\n",
    "    for run in tqdm(root.runs):\n",
    "        tomo = run.get_voxel_spacing(10)\n",
    "        tomo = tomo.get_tomogram(tomo_type).numpy()\n",
    "        target = np.zeros(tomo.shape, dtype=np.uint8)\n",
    "        for pickable_object in root.pickable_objects:\n",
    "            pick = run.get_picks(object_name=pickable_object.name, user_id=\"curation\")\n",
    "            if len(pick):  \n",
    "                target = segmentation_from_picks.from_picks(pick[0], \n",
    "                                                            target, \n",
    "                                                            target_objects[pickable_object.name]['radius'] * 0.8,\n",
    "                                                            target_objects[pickable_object.name]['label']\n",
    "                                                            )\n",
    "        write.segmentation(run, target, copick_user_name, name=copick_segmentation_name)\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8f1f0486-d056-4f42-a7ed-842cd39d771f",
   "metadata": {},
   "source": [
    "### 3. Get tomograms and their segmentaion masks (from picks) arrays\n",
    "\n",
    "This 10439 dataset contains 27 tomograms. We will use the first 5 tomograms for training, and the following 2 tomograms for validation. Therefore, we will get the first 7 tomograms from the cryoET data portal in this notebook, and we will use the rest 20 tomograms for inference in a different notebook."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "3646b23b-93b6-4e8a-9915-5d6c8c570b0d",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 7/7 [01:10<00:00, 10.07s/it]\n"
     ]
    }
   ],
   "source": [
    "data_dicts = []\n",
    "for run in tqdm(root.runs):\n",
    "    tomogram = run.get_voxel_spacing(voxel_size).get_tomogram(tomo_type).numpy()\n",
    "    segmentation = run.get_segmentations(name=copick_segmentation_name, user_id=copick_user_name, voxel_size=voxel_size, is_multilabel=True)[0].numpy()\n",
    "    data_dicts.append({\"image\": tomogram, \"label\": segmentation})\n",
    "    \n",
    "print(np.unique(data_dicts[0]['label']))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bbbd22d1-0dba-41c2-b5d8-aac460a726e5",
   "metadata": {},
   "source": [
    "Since memebranes do not have picks, we should see 6 distinct labels for particles, with 0 being the default background label. "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3a98a9cb-8752-4c30-87e3-1ffd5c3452c9",
   "metadata": {},
   "source": [
    "### 4. Visualize the tomogram and painted segmentation from ground-truth picks"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "021533cd-5065-4bdc-b3e5-89124fde935f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1500x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Plot the images\n",
    "plt.figure(figsize=(15, 5))\n",
    "\n",
    "plt.subplot(1, 2, 1)\n",
    "plt.title('Tomogram')\n",
    "plt.imshow(data_dicts[0]['image'][100],cmap='gray')\n",
    "plt.axis('off')\n",
    "\n",
    "plt.subplot(1, 2, 2)\n",
    "plt.title('Painted Segmentation from Picks')\n",
    "plt.imshow(data_dicts[0]['label'][100], cmap='viridis')\n",
    "plt.axis('off')\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "89f14aa2-87d6-4237-b41c-d07e3e31d0f2",
   "metadata": {},
   "source": [
    "### 5. Prepare dataloaders"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "44234e02",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of training samples: 5\n",
      "Number of validation samples: 2\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Loading dataset: 100%|██████████| 5/5 [00:00<00:00, 15.14it/s]\n",
      "Loading dataset: 100%|██████████| 2/2 [00:00<00:00, 10.21it/s]\n"
     ]
    }
   ],
   "source": [
    "my_num_samples = 16\n",
    "train_batch_size = 1\n",
    "val_batch_size = 1\n",
    "\n",
    "train_files, val_files = data_dicts[:5], data_dicts[5:7]\n",
    "print(f\"Number of training samples: {len(train_files)}\")\n",
    "print(f\"Number of validation samples: {len(val_files)}\")\n",
    "\n",
    "# Non-random transforms to be cached\n",
    "non_random_transforms = Compose([\n",
    "    EnsureChannelFirstd(keys=[\"image\", \"label\"], channel_dim=\"no_channel\"),\n",
    "    NormalizeIntensityd(keys=\"image\"),\n",
    "    Orientationd(keys=[\"image\", \"label\"], axcodes=\"RAS\")\n",
    "])\n",
    "\n",
    "# Random transforms to be applied during training\n",
    "random_transforms = Compose([\n",
    "    RandCropByLabelClassesd(\n",
    "        keys=[\"image\", \"label\"],\n",
    "        label_key=\"label\",\n",
    "        spatial_size=[96, 96, 96],\n",
    "        num_classes=8,\n",
    "        num_samples=my_num_samples\n",
    "    ),\n",
    "    RandRotate90d(keys=[\"image\", \"label\"], prob=0.5, spatial_axes=[0, 2]),\n",
    "    RandFlipd(keys=[\"image\", \"label\"], prob=0.5, spatial_axis=0),    \n",
    "])\n",
    "\n",
    "# Create the cached dataset with non-random transforms\n",
    "train_ds = CacheDataset(data=train_files, transform=non_random_transforms, cache_rate=1.0)\n",
    "\n",
    "# Wrap the cached dataset to apply random transforms during iteration\n",
    "train_ds = Dataset(data=train_ds, transform=random_transforms)\n",
    "\n",
    "# DataLoader remains the same\n",
    "train_loader = DataLoader(\n",
    "    train_ds,\n",
    "    batch_size=train_batch_size,\n",
    "    shuffle=True,\n",
    "    num_workers=4,\n",
    "    pin_memory=torch.cuda.is_available()\n",
    ")\n",
    "\n",
    "# Validation transforms\n",
    "val_transforms = Compose([\n",
    "    EnsureChannelFirstd(keys=[\"image\", \"label\"], channel_dim=\"no_channel\"),\n",
    "    NormalizeIntensityd(keys=\"image\"),\n",
    "    RandCropByLabelClassesd(\n",
    "        keys=[\"image\", \"label\"],\n",
    "        label_key=\"label\",\n",
    "        spatial_size=[96, 96, 96],\n",
    "        num_classes=8,\n",
    "        num_samples=my_num_samples,  # Use 1 to get a single, consistent crop per image\n",
    "    ),\n",
    "])\n",
    "\n",
    "# Create validation dataset\n",
    "val_ds = CacheDataset(data=val_files, transform=non_random_transforms, cache_rate=1.0)\n",
    "\n",
    "# Wrap the cached dataset to apply random transforms during iteration\n",
    "val_ds = Dataset(data=val_ds, transform=random_transforms)\n",
    "\n",
    "# Create validation DataLoader\n",
    "val_loader = DataLoader(\n",
    "    val_ds,\n",
    "    batch_size=val_batch_size,\n",
    "    num_workers=4,\n",
    "    pin_memory=torch.cuda.is_available(),\n",
    "    shuffle=False,  # Ensure the data order remains consistent\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c7222751-50e7-443b-ab8f-4a380aecb021",
   "metadata": {},
   "source": [
    "## Model setup"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "6cd8eb5b-200c-49a5-939f-5221e0a023e6",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cuda\n"
     ]
    }
   ],
   "source": [
    "device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')\n",
    "print(device)\n",
    "# Create UNet, DiceLoss and Adam optimizer\n",
    "model = UNet(\n",
    "    spatial_dims=3,\n",
    "    in_channels=1,\n",
    "    out_channels=len(root.pickable_objects)+1,\n",
    "    channels=(48, 64, 80, 80),\n",
    "    strides=(2, 2, 1),\n",
    "    num_res_units=1,\n",
    ").to(device)\n",
    "\n",
    "lr = 1e-3\n",
    "optimizer = torch.optim.Adam(model.parameters(), lr)\n",
    "#loss_function = DiceLoss(include_background=True, to_onehot_y=True, softmax=True)  # softmax=True for multiclass\n",
    "loss_function = TverskyLoss(include_background=True, to_onehot_y=True, softmax=True)  # softmax=True for multiclass\n",
    "dice_metric = DiceMetric(include_background=False, reduction=\"mean\", ignore_empty=True)  # must use onehot for multiclass\n",
    "recall_metric = ConfusionMatrixMetric(include_background=False, metric_name=\"recall\", reduction=\"None\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "70284f0c-6e29-42d3-87d7-d7f49209dcba",
   "metadata": {
    "scrolled": true,
    "tags": []
   },
   "outputs": [],
   "source": [
    "post_pred = AsDiscrete(argmax=True, to_onehot=len(root.pickable_objects)+1)\n",
    "post_label = AsDiscrete(to_onehot=len(root.pickable_objects)+1)\n",
    "\n",
    "def train(train_loader, model, loss_function, metrics_function, optimizer, max_epochs=100):\n",
    "    val_interval = 2\n",
    "    best_metric = -1\n",
    "    best_metric_epoch = -1\n",
    "    epoch_loss_values = []\n",
    "    metric_values = []\n",
    "    for epoch in range(max_epochs):\n",
    "        print(\"-\" * 10)\n",
    "        print(f\"epoch {epoch + 1}/{max_epochs}\")\n",
    "        model.train()\n",
    "        epoch_loss = 0\n",
    "        step = 0\n",
    "        for batch_data in train_loader:\n",
    "            step += 1\n",
    "            inputs = batch_data[\"image\"].to(device)\n",
    "            labels = batch_data[\"label\"].to(device)\n",
    "            optimizer.zero_grad()\n",
    "            outputs = model(inputs)\n",
    "            loss = loss_function(outputs, labels)\n",
    "            loss.backward()\n",
    "            optimizer.step()\n",
    "            epoch_loss += loss.item()\n",
    "            print(f\"batch {step}/{len(train_ds) // train_loader.batch_size}, \" f\"train_loss: {loss.item():.4f}\")\n",
    "        epoch_loss /= step\n",
    "        epoch_loss_values.append(epoch_loss)\n",
    "        print(f\"epoch {epoch + 1} average loss: {epoch_loss:.4f}\")\n",
    "        mlflow.log_metric(\"train_loss\", epoch_loss, step=epoch+1)\n",
    "\n",
    "        if (epoch + 1) % val_interval == 0:\n",
    "            model.eval()\n",
    "            with torch.no_grad():\n",
    "                for val_data in val_loader:\n",
    "                    val_inputs = val_data[\"image\"].to(device)\n",
    "                    val_labels = val_data[\"label\"].to(device)\n",
    "                    val_outputs = model(val_inputs)\n",
    "                    metric_val_outputs = [post_pred(i) for i in decollate_batch(val_outputs)]\n",
    "                    metric_val_labels = [post_label(i) for i in decollate_batch(val_labels)]\n",
    "                    \n",
    "                    \n",
    "                    # compute metric for current iteration\n",
    "                    metrics_function(y_pred=metric_val_outputs, y=metric_val_labels)\n",
    "\n",
    "                metrics = metrics_function.aggregate(reduction=\"mean_batch\")\n",
    "                metric_per_class = [\"{:.4g}\".format(x) for x in metrics]\n",
    "                metric = torch.mean(metrics).numpy(force=True)\n",
    "                mlflow.log_metric(\"validation metric\", metric, step=epoch+1)\n",
    "                for i,m in enumerate(metrics):\n",
    "                    mlflow.log_metric(f\"validation metric class {i+1}\", m, step=epoch+1)\n",
    "                metrics_function.reset()\n",
    "\n",
    "                metric_values.append(metric)\n",
    "                if metric > best_metric:\n",
    "                    best_metric = metric\n",
    "                    best_metric_epoch = epoch + 1\n",
    "                    torch.save(model.state_dict(), os.path.join('./', \"best_metric_model.pth\"))\n",
    "                    \n",
    "                    print(\"saved new best metric model\")\n",
    "                print(\n",
    "                    f\"current epoch: {epoch + 1} current mean recall per class: {', '.join(metric_per_class)}\"\n",
    "                    f\"\\nbest mean recall: {best_metric:.4f} \"\n",
    "                    f\"at epoch: {best_metric_epoch}\"\n",
    "                )"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b8f55d1f-4601-47dd-a666-b3c519c4632c",
   "metadata": {},
   "source": [
    "## Training and tracking"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "0b8e2a2c-29dc-4615-9ee6-830f693d0c58",
   "metadata": {
    "scrolled": true,
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "----------\n",
      "epoch 1/200\n",
      "batch 1/5, train_loss: 0.9563\n",
      "batch 2/5, train_loss: 0.9532\n",
      "batch 3/5, train_loss: 0.9472\n",
      "batch 4/5, train_loss: 0.9418\n",
      "batch 5/5, train_loss: 0.9325\n",
      "epoch 1 average loss: 0.9462\n",
      "----------\n",
      "epoch 2/200\n",
      "batch 1/5, train_loss: 0.9285\n",
      "batch 2/5, train_loss: 0.9273\n",
      "batch 3/5, train_loss: 0.9196\n",
      "batch 4/5, train_loss: 0.9198\n",
      "batch 5/5, train_loss: 0.9161\n",
      "epoch 2 average loss: 0.9222\n",
      "saved new best metric model\n",
      "current epoch: 2 current mean recall per class: 0.03788, 0.01056, 0.02347, 0.003582, 0.3222, 0.01952, 0.008017\n",
      "best mean recall: 0.0607 at epoch: 2\n",
      "----------\n",
      "epoch 3/200\n",
      "batch 1/5, train_loss: 0.9038\n",
      "batch 2/5, train_loss: 0.8937\n",
      "batch 3/5, train_loss: 0.9015\n",
      "batch 4/5, train_loss: 0.8887\n",
      "batch 5/5, train_loss: 0.8954\n",
      "epoch 3 average loss: 0.8966\n",
      "----------\n",
      "epoch 4/200\n",
      "batch 1/5, train_loss: 0.8856\n",
      "batch 2/5, train_loss: 0.8819\n",
      "batch 3/5, train_loss: 0.8863\n",
      "batch 4/5, train_loss: 0.8716\n",
      "batch 5/5, train_loss: 0.8680\n",
      "epoch 4 average loss: 0.8787\n",
      "current epoch: 4 current mean recall per class: 0.01587, 0.002457, 0.0362, 0.000348, 0.2891, 0.01627, 0.004173\n",
      "best mean recall: 0.0607 at epoch: 2\n",
      "----------\n",
      "epoch 5/200\n",
      "batch 1/5, train_loss: 0.8765\n",
      "batch 2/5, train_loss: 0.8596\n",
      "batch 3/5, train_loss: 0.8657\n",
      "batch 4/5, train_loss: 0.8546\n",
      "batch 5/5, train_loss: 0.8524\n",
      "epoch 5 average loss: 0.8618\n",
      "----------\n",
      "epoch 6/200\n",
      "batch 1/5, train_loss: 0.8652\n",
      "batch 2/5, train_loss: 0.8553\n",
      "batch 3/5, train_loss: 0.8497\n",
      "batch 4/5, train_loss: 0.8292\n",
      "batch 5/5, train_loss: 0.8628\n",
      "epoch 6 average loss: 0.8524\n",
      "current epoch: 6 current mean recall per class: 0.009953, 0.009403, 0.0225, 1.372e-05, 0.337, 0.0004583, 0.0007762\n",
      "best mean recall: 0.0607 at epoch: 2\n",
      "----------\n",
      "epoch 7/200\n",
      "batch 1/5, train_loss: 0.8390\n",
      "batch 2/5, train_loss: 0.8286\n",
      "batch 3/5, train_loss: 0.8505\n",
      "batch 4/5, train_loss: 0.8376\n",
      "batch 5/5, train_loss: 0.8239\n",
      "epoch 7 average loss: 0.8359\n",
      "----------\n",
      "epoch 8/200\n",
      "batch 1/5, train_loss: 0.8359\n",
      "batch 2/5, train_loss: 0.8416\n",
      "batch 3/5, train_loss: 0.8356\n",
      "batch 4/5, train_loss: 0.8166\n",
      "batch 5/5, train_loss: 0.8465\n",
      "epoch 8 average loss: 0.8352\n",
      "saved new best metric model\n",
      "current epoch: 8 current mean recall per class: 0.001088, 0.04174, 0.04748, 0.004388, 0.5121, 0, 0\n",
      "best mean recall: 0.0867 at epoch: 8\n",
      "----------\n",
      "epoch 9/200\n",
      "batch 1/5, train_loss: 0.8348\n",
      "batch 2/5, train_loss: 0.8477\n",
      "batch 3/5, train_loss: 0.8322\n",
      "batch 4/5, train_loss: 0.8267\n",
      "batch 5/5, train_loss: 0.8141\n",
      "epoch 9 average loss: 0.8311\n",
      "----------\n",
      "epoch 10/200\n",
      "batch 1/5, train_loss: 0.8286\n",
      "batch 2/5, train_loss: 0.8204\n",
      "batch 3/5, train_loss: 0.8136\n",
      "batch 4/5, train_loss: 0.8086\n",
      "batch 5/5, train_loss: 0.8123\n",
      "epoch 10 average loss: 0.8167\n",
      "saved new best metric model\n",
      "current epoch: 10 current mean recall per class: 0.0008377, 0.1311, 0.07802, 0.01412, 0.4434, 0, 0\n",
      "best mean recall: 0.0953 at epoch: 10\n",
      "----------\n",
      "epoch 11/200\n",
      "batch 1/5, train_loss: 0.8184\n",
      "batch 2/5, train_loss: 0.8031\n",
      "batch 3/5, train_loss: 0.8086\n",
      "batch 4/5, train_loss: 0.8392\n",
      "batch 5/5, train_loss: 0.8043\n",
      "epoch 11 average loss: 0.8147\n",
      "----------\n",
      "epoch 12/200\n",
      "batch 1/5, train_loss: 0.8087\n",
      "batch 2/5, train_loss: 0.8324\n",
      "batch 3/5, train_loss: 0.8388\n",
      "batch 4/5, train_loss: 0.8308\n",
      "batch 5/5, train_loss: 0.8195\n",
      "epoch 12 average loss: 0.8260\n",
      "saved new best metric model\n",
      "current epoch: 12 current mean recall per class: 0.0007773, 0.1026, 0.07875, 0.05783, 0.5366, 0, 0\n",
      "best mean recall: 0.1109 at epoch: 12\n",
      "----------\n",
      "epoch 13/200\n",
      "batch 1/5, train_loss: 0.8169\n",
      "batch 2/5, train_loss: 0.7823\n",
      "batch 3/5, train_loss: 0.8170\n",
      "batch 4/5, train_loss: 0.8280\n",
      "batch 5/5, train_loss: 0.8023\n",
      "epoch 13 average loss: 0.8093\n",
      "----------\n",
      "epoch 14/200\n",
      "batch 1/5, train_loss: 0.8106\n",
      "batch 2/5, train_loss: 0.7956\n",
      "batch 3/5, train_loss: 0.8052\n",
      "batch 4/5, train_loss: 0.8113\n",
      "batch 5/5, train_loss: 0.7977\n",
      "epoch 14 average loss: 0.8041\n",
      "current epoch: 14 current mean recall per class: 0.0002548, 0.08705, 0.1202, 0.07057, 0.2911, 0, 0\n",
      "best mean recall: 0.1109 at epoch: 12\n",
      "----------\n",
      "epoch 15/200\n",
      "batch 1/5, train_loss: 0.8072\n",
      "batch 2/5, train_loss: 0.8317\n",
      "batch 3/5, train_loss: 0.7806\n",
      "batch 4/5, train_loss: 0.7874\n",
      "batch 5/5, train_loss: 0.7992\n",
      "epoch 15 average loss: 0.8012\n",
      "----------\n",
      "epoch 16/200\n",
      "batch 1/5, train_loss: 0.7988\n",
      "batch 2/5, train_loss: 0.8164\n",
      "batch 3/5, train_loss: 0.7992\n",
      "batch 4/5, train_loss: 0.7804\n",
      "batch 5/5, train_loss: 0.7886\n",
      "epoch 16 average loss: 0.7967\n",
      "current epoch: 16 current mean recall per class: 0.001154, 0.03475, 0.1546, 0.1119, 0.2166, 0, 0\n",
      "best mean recall: 0.1109 at epoch: 12\n",
      "----------\n",
      "epoch 17/200\n",
      "batch 1/5, train_loss: 0.7927\n",
      "batch 2/5, train_loss: 0.8342\n",
      "batch 3/5, train_loss: 0.8057\n",
      "batch 4/5, train_loss: 0.7740\n",
      "batch 5/5, train_loss: 0.7892\n",
      "epoch 17 average loss: 0.7992\n",
      "----------\n",
      "epoch 18/200\n",
      "batch 1/5, train_loss: 0.7948\n",
      "batch 2/5, train_loss: 0.7667\n",
      "batch 3/5, train_loss: 0.8141\n",
      "batch 4/5, train_loss: 0.7895\n",
      "batch 5/5, train_loss: 0.7862\n",
      "epoch 18 average loss: 0.7903\n",
      "current epoch: 18 current mean recall per class: 0.0004407, 0.03009, 0.1646, 0.1414, 0.3861, 0, 0\n",
      "best mean recall: 0.1109 at epoch: 12\n",
      "----------\n",
      "epoch 19/200\n",
      "batch 1/5, train_loss: 0.7797\n",
      "batch 2/5, train_loss: 0.7824\n",
      "batch 3/5, train_loss: 0.8100\n",
      "batch 4/5, train_loss: 0.7700\n",
      "batch 5/5, train_loss: 0.8089\n",
      "epoch 19 average loss: 0.7902\n",
      "----------\n",
      "epoch 20/200\n",
      "batch 1/5, train_loss: 0.8190\n",
      "batch 2/5, train_loss: 0.8038\n",
      "batch 3/5, train_loss: 0.7806\n",
      "batch 4/5, train_loss: 0.7982\n",
      "batch 5/5, train_loss: 0.7884\n",
      "epoch 20 average loss: 0.7980\n",
      "current epoch: 20 current mean recall per class: 0.002856, 0.02608, 0.1848, 0.05934, 0.464, 0, 0\n",
      "best mean recall: 0.1109 at epoch: 12\n",
      "----------\n",
      "epoch 21/200\n",
      "batch 1/5, train_loss: 0.7804\n",
      "batch 2/5, train_loss: 0.7839\n",
      "batch 3/5, train_loss: 0.7719\n",
      "batch 4/5, train_loss: 0.7915\n",
      "batch 5/5, train_loss: 0.7958\n",
      "epoch 21 average loss: 0.7847\n",
      "----------\n",
      "epoch 22/200\n",
      "batch 1/5, train_loss: 0.7575\n",
      "batch 2/5, train_loss: 0.7720\n",
      "batch 3/5, train_loss: 0.8154\n",
      "batch 4/5, train_loss: 0.7646\n",
      "batch 5/5, train_loss: 0.7777\n",
      "epoch 22 average loss: 0.7775\n",
      "saved new best metric model\n",
      "current epoch: 22 current mean recall per class: 0.000393, 0.05365, 0.2128, 0.1571, 0.5161, 0, 0\n",
      "best mean recall: 0.1343 at epoch: 22\n",
      "----------\n",
      "epoch 23/200\n",
      "batch 1/5, train_loss: 0.7537\n",
      "batch 2/5, train_loss: 0.7625\n",
      "batch 3/5, train_loss: 0.7651\n",
      "batch 4/5, train_loss: 0.8017\n",
      "batch 5/5, train_loss: 0.7809\n",
      "epoch 23 average loss: 0.7728\n",
      "----------\n",
      "epoch 24/200\n",
      "batch 1/5, train_loss: 0.7576\n",
      "batch 2/5, train_loss: 0.7812\n",
      "batch 3/5, train_loss: 0.7528\n",
      "batch 4/5, train_loss: 0.7943\n",
      "batch 5/5, train_loss: 0.7782\n",
      "epoch 24 average loss: 0.7728\n",
      "current epoch: 24 current mean recall per class: 0.006077, 0.04232, 0.2099, 0.1068, 0.442, 0, 0\n",
      "best mean recall: 0.1343 at epoch: 22\n",
      "----------\n",
      "epoch 25/200\n",
      "batch 1/5, train_loss: 0.7614\n",
      "batch 2/5, train_loss: 0.7570\n",
      "batch 3/5, train_loss: 0.7643\n",
      "batch 4/5, train_loss: 0.7963\n",
      "batch 5/5, train_loss: 0.7859\n",
      "epoch 25 average loss: 0.7730\n",
      "----------\n",
      "epoch 26/200\n",
      "batch 1/5, train_loss: 0.7670\n",
      "batch 2/5, train_loss: 0.7739\n",
      "batch 3/5, train_loss: 0.7364\n",
      "batch 4/5, train_loss: 0.7609\n",
      "batch 5/5, train_loss: 0.8102\n",
      "epoch 26 average loss: 0.7697\n",
      "current epoch: 26 current mean recall per class: 0.0004247, 0.07503, 0.1958, 0.0875, 0.4244, 0, 0\n",
      "best mean recall: 0.1343 at epoch: 22\n",
      "----------\n",
      "epoch 27/200\n",
      "batch 1/5, train_loss: 0.7634\n",
      "batch 2/5, train_loss: 0.7619\n",
      "batch 3/5, train_loss: 0.7441\n",
      "batch 4/5, train_loss: 0.7690\n",
      "batch 5/5, train_loss: 0.7768\n",
      "epoch 27 average loss: 0.7630\n",
      "----------\n",
      "epoch 28/200\n",
      "batch 1/5, train_loss: 0.7833\n",
      "batch 2/5, train_loss: 0.7645\n",
      "batch 3/5, train_loss: 0.7401\n",
      "batch 4/5, train_loss: 0.7555\n",
      "batch 5/5, train_loss: 0.7772\n",
      "epoch 28 average loss: 0.7641\n",
      "saved new best metric model\n",
      "current epoch: 28 current mean recall per class: 0.004878, 0.09144, 0.2084, 0.1895, 0.585, 0, 0\n",
      "best mean recall: 0.1542 at epoch: 28\n",
      "----------\n",
      "epoch 29/200\n",
      "batch 1/5, train_loss: 0.7622\n",
      "batch 2/5, train_loss: 0.7461\n",
      "batch 3/5, train_loss: 0.7427\n",
      "batch 4/5, train_loss: 0.7907\n",
      "batch 5/5, train_loss: 0.7655\n",
      "epoch 29 average loss: 0.7614\n",
      "----------\n",
      "epoch 30/200\n",
      "batch 1/5, train_loss: 0.7555\n",
      "batch 2/5, train_loss: 0.7517\n",
      "batch 3/5, train_loss: 0.7516\n",
      "batch 4/5, train_loss: 0.7829\n",
      "batch 5/5, train_loss: 0.7592\n",
      "epoch 30 average loss: 0.7602\n",
      "current epoch: 30 current mean recall per class: 0.003908, 0.08625, 0.2439, 0.1567, 0.4272, 0, 0\n",
      "best mean recall: 0.1542 at epoch: 28\n",
      "----------\n",
      "epoch 31/200\n",
      "batch 1/5, train_loss: 0.7691\n",
      "batch 2/5, train_loss: 0.7383\n",
      "batch 3/5, train_loss: 0.7476\n",
      "batch 4/5, train_loss: 0.7999\n",
      "batch 5/5, train_loss: 0.7402\n",
      "epoch 31 average loss: 0.7590\n",
      "----------\n",
      "epoch 32/200\n",
      "batch 1/5, train_loss: 0.7507\n",
      "batch 2/5, train_loss: 0.7443\n",
      "batch 3/5, train_loss: 0.7714\n",
      "batch 4/5, train_loss: 0.7725\n",
      "batch 5/5, train_loss: 0.7692\n",
      "epoch 32 average loss: 0.7616\n",
      "current epoch: 32 current mean recall per class: 0.01514, 0.1372, 0.2125, 0.1438, 0.511, 0, 0\n",
      "best mean recall: 0.1542 at epoch: 28\n",
      "----------\n",
      "epoch 33/200\n",
      "batch 1/5, train_loss: 0.7702\n",
      "batch 2/5, train_loss: 0.7552\n",
      "batch 3/5, train_loss: 0.7805\n",
      "batch 4/5, train_loss: 0.7249\n",
      "batch 5/5, train_loss: 0.7539\n",
      "epoch 33 average loss: 0.7569\n",
      "----------\n",
      "epoch 34/200\n",
      "batch 1/5, train_loss: 0.7347\n",
      "batch 2/5, train_loss: 0.7345\n",
      "batch 3/5, train_loss: 0.7933\n",
      "batch 4/5, train_loss: 0.7623\n",
      "batch 5/5, train_loss: 0.7649\n",
      "epoch 34 average loss: 0.7579\n",
      "current epoch: 34 current mean recall per class: 0.007701, 0.1506, 0.2504, 0.162, 0.4547, 0, 0\n",
      "best mean recall: 0.1542 at epoch: 28\n",
      "----------\n",
      "epoch 35/200\n",
      "batch 1/5, train_loss: 0.7450\n",
      "batch 2/5, train_loss: 0.7374\n",
      "batch 3/5, train_loss: 0.7404\n",
      "batch 4/5, train_loss: 0.7468\n",
      "batch 5/5, train_loss: 0.7819\n",
      "epoch 35 average loss: 0.7503\n",
      "----------\n",
      "epoch 36/200\n",
      "batch 1/5, train_loss: 0.7559\n",
      "batch 2/5, train_loss: 0.7536\n",
      "batch 3/5, train_loss: 0.7658\n",
      "batch 4/5, train_loss: 0.7775\n",
      "batch 5/5, train_loss: 0.7392\n",
      "epoch 36 average loss: 0.7584\n",
      "current epoch: 36 current mean recall per class: 0.008777, 0.1589, 0.2273, 0.1552, 0.4931, 1.363e-05, 3.054e-05\n",
      "best mean recall: 0.1542 at epoch: 28\n",
      "----------\n",
      "epoch 37/200\n",
      "batch 1/5, train_loss: 0.7550\n",
      "batch 2/5, train_loss: 0.7792\n",
      "batch 3/5, train_loss: 0.7035\n",
      "batch 4/5, train_loss: 0.7087\n",
      "batch 5/5, train_loss: 0.7162\n",
      "epoch 37 average loss: 0.7325\n",
      "----------\n",
      "epoch 38/200\n",
      "batch 1/5, train_loss: 0.7933\n",
      "batch 2/5, train_loss: 0.7117\n",
      "batch 3/5, train_loss: 0.7210\n",
      "batch 4/5, train_loss: 0.7566\n",
      "batch 5/5, train_loss: 0.7213\n",
      "epoch 38 average loss: 0.7408\n",
      "saved new best metric model\n",
      "current epoch: 38 current mean recall per class: 0.008157, 0.2048, 0.2539, 0.1441, 0.5733, 0.01294, 0.01277\n",
      "best mean recall: 0.1728 at epoch: 38\n",
      "----------\n",
      "epoch 39/200\n",
      "batch 1/5, train_loss: 0.7677\n",
      "batch 2/5, train_loss: 0.7429\n",
      "batch 3/5, train_loss: 0.7243\n",
      "batch 4/5, train_loss: 0.7003\n",
      "batch 5/5, train_loss: 0.7143\n",
      "epoch 39 average loss: 0.7299\n",
      "----------\n",
      "epoch 40/200\n",
      "batch 1/5, train_loss: 0.6936\n",
      "batch 2/5, train_loss: 0.7361\n",
      "batch 3/5, train_loss: 0.7196\n",
      "batch 4/5, train_loss: 0.7367\n",
      "batch 5/5, train_loss: 0.7745\n",
      "epoch 40 average loss: 0.7321\n",
      "current epoch: 40 current mean recall per class: 0.01636, 0.1996, 0.2386, 0.1818, 0.4502, 0.009808, 0.04731\n",
      "best mean recall: 0.1728 at epoch: 38\n",
      "----------\n",
      "epoch 41/200\n",
      "batch 1/5, train_loss: 0.7315\n",
      "batch 2/5, train_loss: 0.7325\n",
      "batch 3/5, train_loss: 0.7355\n",
      "batch 4/5, train_loss: 0.7141\n",
      "batch 5/5, train_loss: 0.7345\n",
      "epoch 41 average loss: 0.7296\n",
      "----------\n",
      "epoch 42/200\n",
      "batch 1/5, train_loss: 0.7374\n",
      "batch 2/5, train_loss: 0.6936\n",
      "batch 3/5, train_loss: 0.7011\n",
      "batch 4/5, train_loss: 0.7408\n",
      "batch 5/5, train_loss: 0.7053\n",
      "epoch 42 average loss: 0.7156\n",
      "saved new best metric model\n",
      "current epoch: 42 current mean recall per class: 0.0005693, 0.2954, 0.2787, 0.1356, 0.469, 0.1473, 0.1208\n",
      "best mean recall: 0.2068 at epoch: 42\n",
      "----------\n",
      "epoch 43/200\n",
      "batch 1/5, train_loss: 0.6923\n",
      "batch 2/5, train_loss: 0.7366\n",
      "batch 3/5, train_loss: 0.6750\n",
      "batch 4/5, train_loss: 0.7054\n",
      "batch 5/5, train_loss: 0.7546\n",
      "epoch 43 average loss: 0.7128\n",
      "----------\n",
      "epoch 44/200\n",
      "batch 1/5, train_loss: 0.6816\n",
      "batch 2/5, train_loss: 0.7143\n",
      "batch 3/5, train_loss: 0.7087\n",
      "batch 4/5, train_loss: 0.7283\n",
      "batch 5/5, train_loss: 0.6945\n",
      "epoch 44 average loss: 0.7055\n",
      "saved new best metric model\n",
      "current epoch: 44 current mean recall per class: 3.024e-05, 0.3381, 0.2306, 0.1592, 0.3851, 0.1907, 0.2848\n",
      "best mean recall: 0.2269 at epoch: 44\n",
      "----------\n",
      "epoch 45/200\n",
      "batch 1/5, train_loss: 0.7558\n",
      "batch 2/5, train_loss: 0.7098\n",
      "batch 3/5, train_loss: 0.6796\n",
      "batch 4/5, train_loss: 0.6718\n",
      "batch 5/5, train_loss: 0.7355\n",
      "epoch 45 average loss: 0.7105\n",
      "----------\n",
      "epoch 46/200\n",
      "batch 1/5, train_loss: 0.7047\n",
      "batch 2/5, train_loss: 0.7060\n",
      "batch 3/5, train_loss: 0.7084\n",
      "batch 4/5, train_loss: 0.6907\n",
      "batch 5/5, train_loss: 0.7363\n",
      "epoch 46 average loss: 0.7092\n",
      "current epoch: 46 current mean recall per class: 0, 0.1685, 0.3246, 0.1348, 0.4879, 0.1783, 0.1965\n",
      "best mean recall: 0.2269 at epoch: 44\n",
      "----------\n",
      "epoch 47/200\n",
      "batch 1/5, train_loss: 0.6642\n",
      "batch 2/5, train_loss: 0.7189\n",
      "batch 3/5, train_loss: 0.6991\n",
      "batch 4/5, train_loss: 0.6934\n",
      "batch 5/5, train_loss: 0.7083\n",
      "epoch 47 average loss: 0.6968\n",
      "----------\n",
      "epoch 48/200\n",
      "batch 1/5, train_loss: 0.7000\n",
      "batch 2/5, train_loss: 0.6997\n",
      "batch 3/5, train_loss: 0.7084\n",
      "batch 4/5, train_loss: 0.7230\n",
      "batch 5/5, train_loss: 0.7099\n",
      "epoch 48 average loss: 0.7082\n",
      "saved new best metric model\n",
      "current epoch: 48 current mean recall per class: 0.0002188, 0.3376, 0.2592, 0.1604, 0.4819, 0.2779, 0.1553\n",
      "best mean recall: 0.2389 at epoch: 48\n",
      "----------\n",
      "epoch 49/200\n",
      "batch 1/5, train_loss: 0.6701\n",
      "batch 2/5, train_loss: 0.7467\n",
      "batch 3/5, train_loss: 0.6860\n",
      "batch 4/5, train_loss: 0.7077\n",
      "batch 5/5, train_loss: 0.6912\n",
      "epoch 49 average loss: 0.7004\n",
      "----------\n",
      "epoch 50/200\n",
      "batch 1/5, train_loss: 0.6682\n",
      "batch 2/5, train_loss: 0.6846\n",
      "batch 3/5, train_loss: 0.7177\n",
      "batch 4/5, train_loss: 0.7120\n",
      "batch 5/5, train_loss: 0.6535\n",
      "epoch 50 average loss: 0.6872\n",
      "saved new best metric model\n",
      "current epoch: 50 current mean recall per class: 0.0004473, 0.2363, 0.2911, 0.2392, 0.381, 0.3509, 0.363\n",
      "best mean recall: 0.2660 at epoch: 50\n",
      "----------\n",
      "epoch 51/200\n",
      "batch 1/5, train_loss: 0.6992\n",
      "batch 2/5, train_loss: 0.6661\n",
      "batch 3/5, train_loss: 0.6962\n",
      "batch 4/5, train_loss: 0.6884\n",
      "batch 5/5, train_loss: 0.6915\n",
      "epoch 51 average loss: 0.6883\n",
      "----------\n",
      "epoch 52/200\n",
      "batch 1/5, train_loss: 0.7193\n",
      "batch 2/5, train_loss: 0.6511\n",
      "batch 3/5, train_loss: 0.6452\n",
      "batch 4/5, train_loss: 0.6829\n",
      "batch 5/5, train_loss: 0.7056\n",
      "epoch 52 average loss: 0.6808\n",
      "current epoch: 52 current mean recall per class: 0.000115, 0.2228, 0.2364, 0.262, 0.4065, 0.2696, 0.2173\n",
      "best mean recall: 0.2660 at epoch: 50\n",
      "----------\n",
      "epoch 53/200\n",
      "batch 1/5, train_loss: 0.6894\n",
      "batch 2/5, train_loss: 0.6541\n",
      "batch 3/5, train_loss: 0.6503\n",
      "batch 4/5, train_loss: 0.7217\n",
      "batch 5/5, train_loss: 0.6602\n",
      "epoch 53 average loss: 0.6751\n",
      "----------\n",
      "epoch 54/200\n",
      "batch 1/5, train_loss: 0.6713\n",
      "batch 2/5, train_loss: 0.6929\n",
      "batch 3/5, train_loss: 0.6724\n",
      "batch 4/5, train_loss: 0.6520\n",
      "batch 5/5, train_loss: 0.7057\n",
      "epoch 54 average loss: 0.6789\n",
      "current epoch: 54 current mean recall per class: 0, 0.3563, 0.2943, 0.09173, 0.4517, 0.2635, 0.2236\n",
      "best mean recall: 0.2660 at epoch: 50\n",
      "----------\n",
      "epoch 55/200\n",
      "batch 1/5, train_loss: 0.6473\n",
      "batch 2/5, train_loss: 0.6335\n",
      "batch 3/5, train_loss: 0.6470\n",
      "batch 4/5, train_loss: 0.7205\n",
      "batch 5/5, train_loss: 0.6928\n",
      "epoch 55 average loss: 0.6682\n",
      "----------\n",
      "epoch 56/200\n",
      "batch 1/5, train_loss: 0.6716\n",
      "batch 2/5, train_loss: 0.6545\n",
      "batch 3/5, train_loss: 0.6769\n",
      "batch 4/5, train_loss: 0.7253\n",
      "batch 5/5, train_loss: 0.6608\n",
      "epoch 56 average loss: 0.6778\n",
      "current epoch: 56 current mean recall per class: 8.574e-06, 0.3118, 0.2547, 0.2974, 0.3674, 0.2162, 0.1524\n",
      "best mean recall: 0.2660 at epoch: 50\n",
      "----------\n",
      "epoch 57/200\n",
      "batch 1/5, train_loss: 0.6521\n",
      "batch 2/5, train_loss: 0.6774\n",
      "batch 3/5, train_loss: 0.7280\n",
      "batch 4/5, train_loss: 0.6990\n",
      "batch 5/5, train_loss: 0.6600\n",
      "epoch 57 average loss: 0.6833\n",
      "----------\n",
      "epoch 58/200\n",
      "batch 1/5, train_loss: 0.6860\n",
      "batch 2/5, train_loss: 0.6641\n",
      "batch 3/5, train_loss: 0.6813\n",
      "batch 4/5, train_loss: 0.6785\n",
      "batch 5/5, train_loss: 0.7004\n",
      "epoch 58 average loss: 0.6821\n",
      "current epoch: 58 current mean recall per class: 4.424e-05, 0.2988, 0.2912, 0.2541, 0.4526, 0.3317, 0.09739\n",
      "best mean recall: 0.2660 at epoch: 50\n",
      "----------\n",
      "epoch 59/200\n",
      "batch 1/5, train_loss: 0.7178\n",
      "batch 2/5, train_loss: 0.6595\n",
      "batch 3/5, train_loss: 0.6515\n",
      "batch 4/5, train_loss: 0.6557\n",
      "batch 5/5, train_loss: 0.7006\n",
      "epoch 59 average loss: 0.6770\n",
      "----------\n",
      "epoch 60/200\n",
      "batch 1/5, train_loss: 0.6695\n",
      "batch 2/5, train_loss: 0.6849\n",
      "batch 3/5, train_loss: 0.6456\n",
      "batch 4/5, train_loss: 0.6601\n",
      "batch 5/5, train_loss: 0.6413\n",
      "epoch 60 average loss: 0.6603\n",
      "saved new best metric model\n",
      "current epoch: 60 current mean recall per class: 6.278e-06, 0.2826, 0.2694, 0.292, 0.4893, 0.3045, 0.2534\n",
      "best mean recall: 0.2702 at epoch: 60\n",
      "----------\n",
      "epoch 61/200\n",
      "batch 1/5, train_loss: 0.6708\n",
      "batch 2/5, train_loss: 0.6856\n",
      "batch 3/5, train_loss: 0.6658\n",
      "batch 4/5, train_loss: 0.6730\n",
      "batch 5/5, train_loss: 0.6501\n",
      "epoch 61 average loss: 0.6691\n",
      "----------\n",
      "epoch 62/200\n",
      "batch 1/5, train_loss: 0.6776\n",
      "batch 2/5, train_loss: 0.6583\n",
      "batch 3/5, train_loss: 0.6307\n",
      "batch 4/5, train_loss: 0.6451\n",
      "batch 5/5, train_loss: 0.7093\n",
      "epoch 62 average loss: 0.6642\n",
      "saved new best metric model\n",
      "current epoch: 62 current mean recall per class: 0, 0.3622, 0.2893, 0.2502, 0.4832, 0.3797, 0.3468\n",
      "best mean recall: 0.3016 at epoch: 62\n",
      "----------\n",
      "epoch 63/200\n",
      "batch 1/5, train_loss: 0.6724\n",
      "batch 2/5, train_loss: 0.6230\n",
      "batch 3/5, train_loss: 0.6345\n",
      "batch 4/5, train_loss: 0.6660\n",
      "batch 5/5, train_loss: 0.6961\n",
      "epoch 63 average loss: 0.6584\n",
      "----------\n",
      "epoch 64/200\n",
      "batch 1/5, train_loss: 0.6904\n",
      "batch 2/5, train_loss: 0.6519\n",
      "batch 3/5, train_loss: 0.6615\n",
      "batch 4/5, train_loss: 0.6749\n",
      "batch 5/5, train_loss: 0.6710\n",
      "epoch 64 average loss: 0.6700\n",
      "current epoch: 64 current mean recall per class: 1.077e-05, 0.2704, 0.2622, 0.1943, 0.4585, 0.4113, 0.1999\n",
      "best mean recall: 0.3016 at epoch: 62\n",
      "----------\n",
      "epoch 65/200\n",
      "batch 1/5, train_loss: 0.6654\n",
      "batch 2/5, train_loss: 0.7003\n",
      "batch 3/5, train_loss: 0.6304\n",
      "batch 4/5, train_loss: 0.6502\n",
      "batch 5/5, train_loss: 0.6910\n",
      "epoch 65 average loss: 0.6675\n",
      "----------\n",
      "epoch 66/200\n",
      "batch 1/5, train_loss: 0.7013\n",
      "batch 2/5, train_loss: 0.6398\n",
      "batch 3/5, train_loss: 0.6466\n",
      "batch 4/5, train_loss: 0.6182\n",
      "batch 5/5, train_loss: 0.6914\n",
      "epoch 66 average loss: 0.6595\n",
      "current epoch: 66 current mean recall per class: 8.846e-06, 0.3218, 0.3026, 0.1903, 0.3533, 0.4496, 0.09209\n",
      "best mean recall: 0.3016 at epoch: 62\n",
      "----------\n",
      "epoch 67/200\n",
      "batch 1/5, train_loss: 0.6408\n",
      "batch 2/5, train_loss: 0.6755\n",
      "batch 3/5, train_loss: 0.6647\n",
      "batch 4/5, train_loss: 0.6918\n",
      "batch 5/5, train_loss: 0.6242\n",
      "epoch 67 average loss: 0.6594\n",
      "----------\n",
      "epoch 68/200\n",
      "batch 1/5, train_loss: 0.6689\n",
      "batch 2/5, train_loss: 0.6875\n",
      "batch 3/5, train_loss: 0.6482\n",
      "batch 4/5, train_loss: 0.6339\n",
      "batch 5/5, train_loss: 0.6314\n",
      "epoch 68 average loss: 0.6540\n",
      "current epoch: 68 current mean recall per class: 0, 0.3019, 0.3089, 0.2263, 0.332, 0.2726, 0.1412\n",
      "best mean recall: 0.3016 at epoch: 62\n",
      "----------\n",
      "epoch 69/200\n",
      "batch 1/5, train_loss: 0.7058\n",
      "batch 2/5, train_loss: 0.6588\n",
      "batch 3/5, train_loss: 0.6214\n",
      "batch 4/5, train_loss: 0.6372\n",
      "batch 5/5, train_loss: 0.6740\n",
      "epoch 69 average loss: 0.6595\n",
      "----------\n",
      "epoch 70/200\n",
      "batch 1/5, train_loss: 0.6561\n",
      "batch 2/5, train_loss: 0.6795\n",
      "batch 3/5, train_loss: 0.6847\n",
      "batch 4/5, train_loss: 0.6093\n",
      "batch 5/5, train_loss: 0.6278\n",
      "epoch 70 average loss: 0.6515\n",
      "current epoch: 70 current mean recall per class: 0, 0.3849, 0.314, 0.2401, 0.3535, 0.2176, 0.1902\n",
      "best mean recall: 0.3016 at epoch: 62\n",
      "----------\n",
      "epoch 71/200\n",
      "batch 1/5, train_loss: 0.6518\n",
      "batch 2/5, train_loss: 0.6625\n",
      "batch 3/5, train_loss: 0.6767\n",
      "batch 4/5, train_loss: 0.6674\n",
      "batch 5/5, train_loss: 0.6327\n",
      "epoch 71 average loss: 0.6582\n",
      "----------\n",
      "epoch 72/200\n",
      "batch 1/5, train_loss: 0.6581\n",
      "batch 2/5, train_loss: 0.6544\n",
      "batch 3/5, train_loss: 0.6512\n",
      "batch 4/5, train_loss: 0.6924\n",
      "batch 5/5, train_loss: 0.6201\n",
      "epoch 72 average loss: 0.6552\n",
      "current epoch: 72 current mean recall per class: 0, 0.3823, 0.2882, 0.1844, 0.2402, 0.26, 0.2185\n",
      "best mean recall: 0.3016 at epoch: 62\n",
      "----------\n",
      "epoch 73/200\n",
      "batch 1/5, train_loss: 0.6144\n",
      "batch 2/5, train_loss: 0.7004\n",
      "batch 3/5, train_loss: 0.6414\n",
      "batch 4/5, train_loss: 0.6353\n",
      "batch 5/5, train_loss: 0.6294\n",
      "epoch 73 average loss: 0.6442\n",
      "----------\n",
      "epoch 74/200\n",
      "batch 1/5, train_loss: 0.6800\n",
      "batch 2/5, train_loss: 0.6666\n",
      "batch 3/5, train_loss: 0.6674\n",
      "batch 4/5, train_loss: 0.6416\n",
      "batch 5/5, train_loss: 0.6525\n",
      "epoch 74 average loss: 0.6616\n",
      "current epoch: 74 current mean recall per class: 0, 0.3514, 0.3168, 0.2317, 0.4099, 0.4495, 0.2569\n",
      "best mean recall: 0.3016 at epoch: 62\n",
      "----------\n",
      "epoch 75/200\n",
      "batch 1/5, train_loss: 0.6516\n",
      "batch 2/5, train_loss: 0.6831\n",
      "batch 3/5, train_loss: 0.6884\n",
      "batch 4/5, train_loss: 0.6075\n",
      "batch 5/5, train_loss: 0.6529\n",
      "epoch 75 average loss: 0.6567\n",
      "----------\n",
      "epoch 76/200\n",
      "batch 1/5, train_loss: 0.6787\n",
      "batch 2/5, train_loss: 0.6677\n",
      "batch 3/5, train_loss: 0.6423\n",
      "batch 4/5, train_loss: 0.6104\n",
      "batch 5/5, train_loss: 0.6164\n",
      "epoch 76 average loss: 0.6431\n",
      "current epoch: 76 current mean recall per class: 0, 0.3657, 0.3175, 0.3, 0.3766, 0.3784, 0.2266\n",
      "best mean recall: 0.3016 at epoch: 62\n",
      "----------\n",
      "epoch 77/200\n",
      "batch 1/5, train_loss: 0.6333\n",
      "batch 2/5, train_loss: 0.6285\n",
      "batch 3/5, train_loss: 0.6864\n",
      "batch 4/5, train_loss: 0.5926\n",
      "batch 5/5, train_loss: 0.6545\n",
      "epoch 77 average loss: 0.6391\n",
      "----------\n",
      "epoch 78/200\n",
      "batch 1/5, train_loss: 0.6544\n",
      "batch 2/5, train_loss: 0.6592\n",
      "batch 3/5, train_loss: 0.6323\n",
      "batch 4/5, train_loss: 0.6359\n",
      "batch 5/5, train_loss: 0.6131\n",
      "epoch 78 average loss: 0.6390\n",
      "current epoch: 78 current mean recall per class: 0, 0.3111, 0.3292, 0.2056, 0.4229, 0.3804, 0.06854\n",
      "best mean recall: 0.3016 at epoch: 62\n",
      "----------\n",
      "epoch 79/200\n",
      "batch 1/5, train_loss: 0.6513\n",
      "batch 2/5, train_loss: 0.6489\n",
      "batch 3/5, train_loss: 0.7048\n",
      "batch 4/5, train_loss: 0.6935\n",
      "batch 5/5, train_loss: 0.6048\n",
      "epoch 79 average loss: 0.6607\n",
      "----------\n",
      "epoch 80/200\n",
      "batch 1/5, train_loss: 0.6485\n",
      "batch 2/5, train_loss: 0.6360\n",
      "batch 3/5, train_loss: 0.7053\n",
      "batch 4/5, train_loss: 0.6684\n",
      "batch 5/5, train_loss: 0.6254\n",
      "epoch 80 average loss: 0.6567\n",
      "saved new best metric model\n",
      "current epoch: 80 current mean recall per class: 4.963e-06, 0.3418, 0.2905, 0.3487, 0.4683, 0.3642, 0.3127\n",
      "best mean recall: 0.3038 at epoch: 80\n",
      "----------\n",
      "epoch 81/200\n",
      "batch 1/5, train_loss: 0.6028\n",
      "batch 2/5, train_loss: 0.6367\n",
      "batch 3/5, train_loss: 0.5899\n",
      "batch 4/5, train_loss: 0.6576\n",
      "batch 5/5, train_loss: 0.6852\n",
      "epoch 81 average loss: 0.6345\n",
      "----------\n",
      "epoch 82/200\n",
      "batch 1/5, train_loss: 0.6436\n",
      "batch 2/5, train_loss: 0.5947\n",
      "batch 3/5, train_loss: 0.6870\n",
      "batch 4/5, train_loss: 0.6156\n",
      "batch 5/5, train_loss: 0.6312\n",
      "epoch 82 average loss: 0.6344\n",
      "current epoch: 82 current mean recall per class: 0, 0.3983, 0.3009, 0.2765, 0.4413, 0.3979, 0.2621\n",
      "best mean recall: 0.3038 at epoch: 80\n",
      "----------\n",
      "epoch 83/200\n",
      "batch 1/5, train_loss: 0.6332\n",
      "batch 2/5, train_loss: 0.6171\n",
      "batch 3/5, train_loss: 0.6491\n",
      "batch 4/5, train_loss: 0.6565\n",
      "batch 5/5, train_loss: 0.6585\n",
      "epoch 83 average loss: 0.6429\n",
      "----------\n",
      "epoch 84/200\n",
      "batch 1/5, train_loss: 0.6518\n",
      "batch 2/5, train_loss: 0.6773\n",
      "batch 3/5, train_loss: 0.6689\n",
      "batch 4/5, train_loss: 0.6438\n",
      "batch 5/5, train_loss: 0.6026\n",
      "epoch 84 average loss: 0.6489\n",
      "current epoch: 84 current mean recall per class: 0, 0.3907, 0.2661, 0.2209, 0.3423, 0.3244, 0.3258\n",
      "best mean recall: 0.3038 at epoch: 80\n",
      "----------\n",
      "epoch 85/200\n",
      "batch 1/5, train_loss: 0.6746\n",
      "batch 2/5, train_loss: 0.6590\n",
      "batch 3/5, train_loss: 0.6108\n",
      "batch 4/5, train_loss: 0.6034\n",
      "batch 5/5, train_loss: 0.6223\n",
      "epoch 85 average loss: 0.6340\n",
      "----------\n",
      "epoch 86/200\n",
      "batch 1/5, train_loss: 0.7054\n",
      "batch 2/5, train_loss: 0.6183\n",
      "batch 3/5, train_loss: 0.6381\n",
      "batch 4/5, train_loss: 0.6123\n",
      "batch 5/5, train_loss: 0.6199\n",
      "epoch 86 average loss: 0.6388\n",
      "current epoch: 86 current mean recall per class: 0, 0.2623, 0.2968, 0.2285, 0.4533, 0.3993, 0.2957\n",
      "best mean recall: 0.3038 at epoch: 80\n",
      "----------\n",
      "epoch 87/200\n",
      "batch 1/5, train_loss: 0.6780\n",
      "batch 2/5, train_loss: 0.6196\n",
      "batch 3/5, train_loss: 0.5594\n",
      "batch 4/5, train_loss: 0.6608\n",
      "batch 5/5, train_loss: 0.6112\n",
      "epoch 87 average loss: 0.6258\n",
      "----------\n",
      "epoch 88/200\n",
      "batch 1/5, train_loss: 0.6553\n",
      "batch 2/5, train_loss: 0.6316\n",
      "batch 3/5, train_loss: 0.6289\n",
      "batch 4/5, train_loss: 0.6288\n",
      "batch 5/5, train_loss: 0.6621\n",
      "epoch 88 average loss: 0.6413\n",
      "current epoch: 88 current mean recall per class: 0, 0.3382, 0.2853, 0.1588, 0.5014, 0.3074, 0.2861\n",
      "best mean recall: 0.3038 at epoch: 80\n",
      "----------\n",
      "epoch 89/200\n",
      "batch 1/5, train_loss: 0.6021\n",
      "batch 2/5, train_loss: 0.6796\n",
      "batch 3/5, train_loss: 0.6117\n",
      "batch 4/5, train_loss: 0.6938\n",
      "batch 5/5, train_loss: 0.6628\n",
      "epoch 89 average loss: 0.6500\n",
      "----------\n",
      "epoch 90/200\n",
      "batch 1/5, train_loss: 0.6356\n",
      "batch 2/5, train_loss: 0.5882\n",
      "batch 3/5, train_loss: 0.5992\n",
      "batch 4/5, train_loss: 0.6643\n",
      "batch 5/5, train_loss: 0.6479\n",
      "epoch 90 average loss: 0.6270\n",
      "saved new best metric model\n",
      "current epoch: 90 current mean recall per class: 0, 0.3159, 0.327, 0.3393, 0.4971, 0.4685, 0.262\n",
      "best mean recall: 0.3157 at epoch: 90\n",
      "----------\n",
      "epoch 91/200\n",
      "batch 1/5, train_loss: 0.6470\n",
      "batch 2/5, train_loss: 0.6566\n",
      "batch 3/5, train_loss: 0.6395\n",
      "batch 4/5, train_loss: 0.6966\n",
      "batch 5/5, train_loss: 0.6263\n",
      "epoch 91 average loss: 0.6532\n",
      "----------\n",
      "epoch 92/200\n",
      "batch 1/5, train_loss: 0.5979\n",
      "batch 2/5, train_loss: 0.6103\n",
      "batch 3/5, train_loss: 0.5663\n",
      "batch 4/5, train_loss: 0.6656\n",
      "batch 5/5, train_loss: 0.6748\n",
      "epoch 92 average loss: 0.6230\n",
      "current epoch: 92 current mean recall per class: 0, 0.3967, 0.2765, 0.2737, 0.548, 0.415, 0.2779\n",
      "best mean recall: 0.3157 at epoch: 90\n",
      "----------\n",
      "epoch 93/200\n",
      "batch 1/5, train_loss: 0.6861\n",
      "batch 2/5, train_loss: 0.6139\n",
      "batch 3/5, train_loss: 0.6567\n",
      "batch 4/5, train_loss: 0.5885\n",
      "batch 5/5, train_loss: 0.6082\n",
      "epoch 93 average loss: 0.6307\n",
      "----------\n",
      "epoch 94/200\n",
      "batch 1/5, train_loss: 0.6917\n",
      "batch 2/5, train_loss: 0.6724\n",
      "batch 3/5, train_loss: 0.5952\n",
      "batch 4/5, train_loss: 0.6314\n",
      "batch 5/5, train_loss: 0.6130\n",
      "epoch 94 average loss: 0.6407\n",
      "current epoch: 94 current mean recall per class: 0, 0.3639, 0.249, 0.2323, 0.39, 0.476, 0.3637\n",
      "best mean recall: 0.3157 at epoch: 90\n",
      "----------\n",
      "epoch 95/200\n",
      "batch 1/5, train_loss: 0.6251\n",
      "batch 2/5, train_loss: 0.6412\n",
      "batch 3/5, train_loss: 0.6277\n",
      "batch 4/5, train_loss: 0.6337\n",
      "batch 5/5, train_loss: 0.5918\n",
      "epoch 95 average loss: 0.6239\n",
      "----------\n",
      "epoch 96/200\n",
      "batch 1/5, train_loss: 0.6270\n",
      "batch 2/5, train_loss: 0.6076\n",
      "batch 3/5, train_loss: 0.6356\n",
      "batch 4/5, train_loss: 0.6911\n",
      "batch 5/5, train_loss: 0.6762\n",
      "epoch 96 average loss: 0.6475\n",
      "current epoch: 96 current mean recall per class: 0, 0.2094, 0.3064, 0.2475, 0.5958, 0.4154, 0.4203\n",
      "best mean recall: 0.3157 at epoch: 90\n",
      "----------\n",
      "epoch 97/200\n",
      "batch 1/5, train_loss: 0.5998\n",
      "batch 2/5, train_loss: 0.6135\n",
      "batch 3/5, train_loss: 0.7041\n",
      "batch 4/5, train_loss: 0.5868\n",
      "batch 5/5, train_loss: 0.6470\n",
      "epoch 97 average loss: 0.6302\n",
      "----------\n",
      "epoch 98/200\n",
      "batch 1/5, train_loss: 0.6627\n",
      "batch 2/5, train_loss: 0.6073\n",
      "batch 3/5, train_loss: 0.6108\n",
      "batch 4/5, train_loss: 0.6607\n",
      "batch 5/5, train_loss: 0.6145\n",
      "epoch 98 average loss: 0.6312\n",
      "current epoch: 98 current mean recall per class: 0, 0.3278, 0.3262, 0.3325, 0.456, 0.3861, 0.3221\n",
      "best mean recall: 0.3157 at epoch: 90\n",
      "----------\n",
      "epoch 99/200\n",
      "batch 1/5, train_loss: 0.6212\n",
      "batch 2/5, train_loss: 0.6678\n",
      "batch 3/5, train_loss: 0.6529\n",
      "batch 4/5, train_loss: 0.6485\n",
      "batch 5/5, train_loss: 0.6432\n",
      "epoch 99 average loss: 0.6467\n",
      "----------\n",
      "epoch 100/200\n",
      "batch 1/5, train_loss: 0.6696\n",
      "batch 2/5, train_loss: 0.6789\n",
      "batch 3/5, train_loss: 0.6216\n",
      "batch 4/5, train_loss: 0.6029\n",
      "batch 5/5, train_loss: 0.5742\n",
      "epoch 100 average loss: 0.6294\n",
      "current epoch: 100 current mean recall per class: 0, 0.3709, 0.3502, 0.2583, 0.3975, 0.445, 0.3263\n",
      "best mean recall: 0.3157 at epoch: 90\n",
      "----------\n",
      "epoch 101/200\n",
      "batch 1/5, train_loss: 0.6269\n",
      "batch 2/5, train_loss: 0.6499\n",
      "batch 3/5, train_loss: 0.6299\n",
      "batch 4/5, train_loss: 0.6339\n",
      "batch 5/5, train_loss: 0.6789\n",
      "epoch 101 average loss: 0.6439\n",
      "----------\n",
      "epoch 102/200\n",
      "batch 1/5, train_loss: 0.6065\n",
      "batch 2/5, train_loss: 0.5879\n",
      "batch 3/5, train_loss: 0.6648\n",
      "batch 4/5, train_loss: 0.6112\n",
      "batch 5/5, train_loss: 0.6582\n",
      "epoch 102 average loss: 0.6257\n",
      "current epoch: 102 current mean recall per class: 0, 0.336, 0.3066, 0.2584, 0.5292, 0.388, 0.3661\n",
      "best mean recall: 0.3157 at epoch: 90\n",
      "----------\n",
      "epoch 103/200\n",
      "batch 1/5, train_loss: 0.6653\n",
      "batch 2/5, train_loss: 0.5915\n",
      "batch 3/5, train_loss: 0.6374\n",
      "batch 4/5, train_loss: 0.5875\n",
      "batch 5/5, train_loss: 0.6324\n",
      "epoch 103 average loss: 0.6228\n",
      "----------\n",
      "epoch 104/200\n",
      "batch 1/5, train_loss: 0.6420\n",
      "batch 2/5, train_loss: 0.6348\n",
      "batch 3/5, train_loss: 0.6234\n",
      "batch 4/5, train_loss: 0.5786\n",
      "batch 5/5, train_loss: 0.6596\n",
      "epoch 104 average loss: 0.6277\n",
      "current epoch: 104 current mean recall per class: 0, 0.3601, 0.2541, 0.2493, 0.398, 0.351, 0.3607\n",
      "best mean recall: 0.3157 at epoch: 90\n",
      "----------\n",
      "epoch 105/200\n",
      "batch 1/5, train_loss: 0.5962\n",
      "batch 2/5, train_loss: 0.6750\n",
      "batch 3/5, train_loss: 0.6078\n",
      "batch 4/5, train_loss: 0.6660\n",
      "batch 5/5, train_loss: 0.6280\n",
      "epoch 105 average loss: 0.6346\n",
      "----------\n",
      "epoch 106/200\n",
      "batch 1/5, train_loss: 0.6806\n",
      "batch 2/5, train_loss: 0.6059\n",
      "batch 3/5, train_loss: 0.6637\n",
      "batch 4/5, train_loss: 0.6011\n",
      "batch 5/5, train_loss: 0.6062\n",
      "epoch 106 average loss: 0.6315\n",
      "saved new best metric model\n",
      "current epoch: 106 current mean recall per class: 0, 0.3328, 0.2817, 0.258, 0.5415, 0.3886, 0.5459\n",
      "best mean recall: 0.3355 at epoch: 106\n",
      "----------\n",
      "epoch 107/200\n",
      "batch 1/5, train_loss: 0.6111\n",
      "batch 2/5, train_loss: 0.5778\n",
      "batch 3/5, train_loss: 0.6731\n",
      "batch 4/5, train_loss: 0.5984\n",
      "batch 5/5, train_loss: 0.6627\n",
      "epoch 107 average loss: 0.6246\n",
      "----------\n",
      "epoch 108/200\n",
      "batch 1/5, train_loss: 0.6037\n",
      "batch 2/5, train_loss: 0.6593\n",
      "batch 3/5, train_loss: 0.6464\n",
      "batch 4/5, train_loss: 0.6711\n",
      "batch 5/5, train_loss: 0.5811\n",
      "epoch 108 average loss: 0.6323\n",
      "current epoch: 108 current mean recall per class: 0, 0.2945, 0.302, 0.3039, 0.5016, 0.4251, 0.118\n",
      "best mean recall: 0.3355 at epoch: 106\n",
      "----------\n",
      "epoch 109/200\n",
      "batch 1/5, train_loss: 0.6082\n",
      "batch 2/5, train_loss: 0.6604\n",
      "batch 3/5, train_loss: 0.5660\n",
      "batch 4/5, train_loss: 0.6556\n",
      "batch 5/5, train_loss: 0.6184\n",
      "epoch 109 average loss: 0.6217\n",
      "----------\n",
      "epoch 110/200\n",
      "batch 1/5, train_loss: 0.6145\n",
      "batch 2/5, train_loss: 0.6269\n",
      "batch 3/5, train_loss: 0.6463\n",
      "batch 4/5, train_loss: 0.6020\n",
      "batch 5/5, train_loss: 0.6604\n",
      "epoch 110 average loss: 0.6300\n",
      "current epoch: 110 current mean recall per class: 0, 0.4078, 0.2926, 0.2217, 0.4972, 0.4222, 0.2669\n",
      "best mean recall: 0.3355 at epoch: 106\n",
      "----------\n",
      "epoch 111/200\n",
      "batch 1/5, train_loss: 0.5758\n",
      "batch 2/5, train_loss: 0.6017\n",
      "batch 3/5, train_loss: 0.6434\n",
      "batch 4/5, train_loss: 0.6160\n",
      "batch 5/5, train_loss: 0.6747\n",
      "epoch 111 average loss: 0.6223\n",
      "----------\n",
      "epoch 112/200\n",
      "batch 1/5, train_loss: 0.5856\n",
      "batch 2/5, train_loss: 0.5914\n",
      "batch 3/5, train_loss: 0.5775\n",
      "batch 4/5, train_loss: 0.6777\n",
      "batch 5/5, train_loss: 0.6364\n",
      "epoch 112 average loss: 0.6137\n",
      "current epoch: 112 current mean recall per class: 0, 0.3357, 0.3223, 0.2835, 0.513, 0.3687, 0.298\n",
      "best mean recall: 0.3355 at epoch: 106\n",
      "----------\n",
      "epoch 113/200\n",
      "batch 1/5, train_loss: 0.6374\n",
      "batch 2/5, train_loss: 0.6193\n",
      "batch 3/5, train_loss: 0.6183\n",
      "batch 4/5, train_loss: 0.5985\n",
      "batch 5/5, train_loss: 0.5920\n",
      "epoch 113 average loss: 0.6131\n",
      "----------\n",
      "epoch 114/200\n",
      "batch 1/5, train_loss: 0.6529\n",
      "batch 2/5, train_loss: 0.5846\n",
      "batch 3/5, train_loss: 0.6571\n",
      "batch 4/5, train_loss: 0.6388\n",
      "batch 5/5, train_loss: 0.5986\n",
      "epoch 114 average loss: 0.6264\n",
      "current epoch: 114 current mean recall per class: 0, 0.3385, 0.3388, 0.2084, 0.3526, 0.4251, 0.2709\n",
      "best mean recall: 0.3355 at epoch: 106\n",
      "----------\n",
      "epoch 115/200\n",
      "batch 1/5, train_loss: 0.6582\n",
      "batch 2/5, train_loss: 0.6372\n",
      "batch 3/5, train_loss: 0.6328\n",
      "batch 4/5, train_loss: 0.6257\n",
      "batch 5/5, train_loss: 0.6125\n",
      "epoch 115 average loss: 0.6333\n",
      "----------\n",
      "epoch 116/200\n",
      "batch 1/5, train_loss: 0.6124\n",
      "batch 2/5, train_loss: 0.5965\n",
      "batch 3/5, train_loss: 0.6510\n",
      "batch 4/5, train_loss: 0.6500\n",
      "batch 5/5, train_loss: 0.5429\n",
      "epoch 116 average loss: 0.6106\n",
      "saved new best metric model\n",
      "current epoch: 116 current mean recall per class: 0, 0.3716, 0.3465, 0.219, 0.5732, 0.4681, 0.4108\n",
      "best mean recall: 0.3413 at epoch: 116\n",
      "----------\n",
      "epoch 117/200\n",
      "batch 1/5, train_loss: 0.6442\n",
      "batch 2/5, train_loss: 0.6095\n",
      "batch 3/5, train_loss: 0.6336\n",
      "batch 4/5, train_loss: 0.5780\n",
      "batch 5/5, train_loss: 0.6177\n",
      "epoch 117 average loss: 0.6166\n",
      "----------\n",
      "epoch 118/200\n",
      "batch 1/5, train_loss: 0.5833\n",
      "batch 2/5, train_loss: 0.6374\n",
      "batch 3/5, train_loss: 0.6225\n",
      "batch 4/5, train_loss: 0.6523\n",
      "batch 5/5, train_loss: 0.6202\n",
      "epoch 118 average loss: 0.6231\n",
      "saved new best metric model\n",
      "current epoch: 118 current mean recall per class: 0, 0.3192, 0.3113, 0.3163, 0.5694, 0.4986, 0.4555\n",
      "best mean recall: 0.3529 at epoch: 118\n",
      "----------\n",
      "epoch 119/200\n",
      "batch 1/5, train_loss: 0.6494\n",
      "batch 2/5, train_loss: 0.6184\n",
      "batch 3/5, train_loss: 0.6406\n",
      "batch 4/5, train_loss: 0.5873\n",
      "batch 5/5, train_loss: 0.6684\n",
      "epoch 119 average loss: 0.6328\n",
      "----------\n",
      "epoch 120/200\n",
      "batch 1/5, train_loss: 0.6233\n",
      "batch 2/5, train_loss: 0.6198\n",
      "batch 3/5, train_loss: 0.6495\n",
      "batch 4/5, train_loss: 0.6166\n",
      "batch 5/5, train_loss: 0.6383\n",
      "epoch 120 average loss: 0.6295\n",
      "current epoch: 120 current mean recall per class: 0, 0.3419, 0.2946, 0.2272, 0.5694, 0.4486, 0.4487\n",
      "best mean recall: 0.3529 at epoch: 118\n",
      "----------\n",
      "epoch 121/200\n",
      "batch 1/5, train_loss: 0.5958\n",
      "batch 2/5, train_loss: 0.6351\n",
      "batch 3/5, train_loss: 0.6260\n",
      "batch 4/5, train_loss: 0.6505\n",
      "batch 5/5, train_loss: 0.5753\n",
      "epoch 121 average loss: 0.6166\n",
      "----------\n",
      "epoch 122/200\n",
      "batch 1/5, train_loss: 0.6614\n",
      "batch 2/5, train_loss: 0.5991\n",
      "batch 3/5, train_loss: 0.6044\n",
      "batch 4/5, train_loss: 0.5881\n",
      "batch 5/5, train_loss: 0.6483\n",
      "epoch 122 average loss: 0.6203\n",
      "current epoch: 122 current mean recall per class: 0, 0.3518, 0.3312, 0.2941, 0.4974, 0.3931, 0.3866\n",
      "best mean recall: 0.3529 at epoch: 118\n",
      "----------\n",
      "epoch 123/200\n",
      "batch 1/5, train_loss: 0.6607\n",
      "batch 2/5, train_loss: 0.6498\n",
      "batch 3/5, train_loss: 0.6736\n",
      "batch 4/5, train_loss: 0.6030\n",
      "batch 5/5, train_loss: 0.6059\n",
      "epoch 123 average loss: 0.6386\n",
      "----------\n",
      "epoch 124/200\n",
      "batch 1/5, train_loss: 0.6270\n",
      "batch 2/5, train_loss: 0.6372\n",
      "batch 3/5, train_loss: 0.6149\n",
      "batch 4/5, train_loss: 0.6501\n",
      "batch 5/5, train_loss: 0.5466\n",
      "epoch 124 average loss: 0.6152\n",
      "current epoch: 124 current mean recall per class: 0, 0.4062, 0.324, 0.2323, 0.6421, 0.4249, 0.3341\n",
      "best mean recall: 0.3529 at epoch: 118\n",
      "----------\n",
      "epoch 125/200\n",
      "batch 1/5, train_loss: 0.6045\n",
      "batch 2/5, train_loss: 0.6575\n",
      "batch 3/5, train_loss: 0.6341\n",
      "batch 4/5, train_loss: 0.5958\n",
      "batch 5/5, train_loss: 0.5896\n",
      "epoch 125 average loss: 0.6163\n",
      "----------\n",
      "epoch 126/200\n",
      "batch 1/5, train_loss: 0.6088\n",
      "batch 2/5, train_loss: 0.6095\n",
      "batch 3/5, train_loss: 0.6239\n",
      "batch 4/5, train_loss: 0.6348\n",
      "batch 5/5, train_loss: 0.6877\n",
      "epoch 126 average loss: 0.6329\n",
      "current epoch: 126 current mean recall per class: 0, 0.4057, 0.3082, 0.2261, 0.5045, 0.4756, 0.3705\n",
      "best mean recall: 0.3529 at epoch: 118\n",
      "----------\n",
      "epoch 127/200\n",
      "batch 1/5, train_loss: 0.6751\n",
      "batch 2/5, train_loss: 0.6662\n",
      "batch 3/5, train_loss: 0.5867\n",
      "batch 4/5, train_loss: 0.6122\n",
      "batch 5/5, train_loss: 0.5718\n",
      "epoch 127 average loss: 0.6224\n",
      "----------\n",
      "epoch 128/200\n",
      "batch 1/5, train_loss: 0.5848\n",
      "batch 2/5, train_loss: 0.6226\n",
      "batch 3/5, train_loss: 0.6579\n",
      "batch 4/5, train_loss: 0.5885\n",
      "batch 5/5, train_loss: 0.6194\n",
      "epoch 128 average loss: 0.6146\n",
      "current epoch: 128 current mean recall per class: 0, 0.3855, 0.3065, 0.264, 0.4208, 0.3141, 0.29\n",
      "best mean recall: 0.3529 at epoch: 118\n",
      "----------\n",
      "epoch 129/200\n",
      "batch 1/5, train_loss: 0.5819\n",
      "batch 2/5, train_loss: 0.6279\n",
      "batch 3/5, train_loss: 0.6494\n",
      "batch 4/5, train_loss: 0.6561\n",
      "batch 5/5, train_loss: 0.5927\n",
      "epoch 129 average loss: 0.6216\n",
      "----------\n",
      "epoch 130/200\n",
      "batch 1/5, train_loss: 0.6214\n",
      "batch 2/5, train_loss: 0.6183\n",
      "batch 3/5, train_loss: 0.6884\n",
      "batch 4/5, train_loss: 0.6411\n",
      "batch 5/5, train_loss: 0.5760\n",
      "epoch 130 average loss: 0.6290\n",
      "saved new best metric model\n",
      "current epoch: 130 current mean recall per class: 8.131e-06, 0.3998, 0.2964, 0.2903, 0.5448, 0.5067, 0.4806\n",
      "best mean recall: 0.3598 at epoch: 130\n",
      "----------\n",
      "epoch 131/200\n",
      "batch 1/5, train_loss: 0.6310\n",
      "batch 2/5, train_loss: 0.6324\n",
      "batch 3/5, train_loss: 0.5805\n",
      "batch 4/5, train_loss: 0.5952\n",
      "batch 5/5, train_loss: 0.6323\n",
      "epoch 131 average loss: 0.6143\n",
      "----------\n",
      "epoch 132/200\n",
      "batch 1/5, train_loss: 0.5985\n",
      "batch 2/5, train_loss: 0.6489\n",
      "batch 3/5, train_loss: 0.6487\n",
      "batch 4/5, train_loss: 0.5909\n",
      "batch 5/5, train_loss: 0.6079\n",
      "epoch 132 average loss: 0.6190\n",
      "current epoch: 132 current mean recall per class: 0, 0.3311, 0.3414, 0.323, 0.5439, 0.4248, 0.4232\n",
      "best mean recall: 0.3598 at epoch: 130\n",
      "----------\n",
      "epoch 133/200\n",
      "batch 1/5, train_loss: 0.6928\n",
      "batch 2/5, train_loss: 0.6366\n",
      "batch 3/5, train_loss: 0.6128\n",
      "batch 4/5, train_loss: 0.6286\n",
      "batch 5/5, train_loss: 0.6071\n",
      "epoch 133 average loss: 0.6356\n",
      "----------\n",
      "epoch 134/200\n",
      "batch 1/5, train_loss: 0.6230\n",
      "batch 2/5, train_loss: 0.6645\n",
      "batch 3/5, train_loss: 0.6203\n",
      "batch 4/5, train_loss: 0.6043\n",
      "batch 5/5, train_loss: 0.5903\n",
      "epoch 134 average loss: 0.6204\n",
      "current epoch: 134 current mean recall per class: 0, 0.3962, 0.3395, 0.2725, 0.3838, 0.4432, 0.3746\n",
      "best mean recall: 0.3598 at epoch: 130\n",
      "----------\n",
      "epoch 135/200\n",
      "batch 1/5, train_loss: 0.5880\n",
      "batch 2/5, train_loss: 0.6623\n",
      "batch 3/5, train_loss: 0.5724\n",
      "batch 4/5, train_loss: 0.5813\n",
      "batch 5/5, train_loss: 0.6726\n",
      "epoch 135 average loss: 0.6153\n",
      "----------\n",
      "epoch 136/200\n",
      "batch 1/5, train_loss: 0.6280\n",
      "batch 2/5, train_loss: 0.5910\n",
      "batch 3/5, train_loss: 0.5756\n",
      "batch 4/5, train_loss: 0.6623\n",
      "batch 5/5, train_loss: 0.5713\n",
      "epoch 136 average loss: 0.6056\n",
      "saved new best metric model\n",
      "current epoch: 136 current mean recall per class: 0, 0.4146, 0.3255, 0.3313, 0.4184, 0.4983, 0.5514\n",
      "best mean recall: 0.3628 at epoch: 136\n",
      "----------\n",
      "epoch 137/200\n",
      "batch 1/5, train_loss: 0.5857\n",
      "batch 2/5, train_loss: 0.6865\n",
      "batch 3/5, train_loss: 0.5867\n",
      "batch 4/5, train_loss: 0.6536\n",
      "batch 5/5, train_loss: 0.6040\n",
      "epoch 137 average loss: 0.6233\n",
      "----------\n",
      "epoch 138/200\n",
      "batch 1/5, train_loss: 0.6463\n",
      "batch 2/5, train_loss: 0.6007\n",
      "batch 3/5, train_loss: 0.6735\n",
      "batch 4/5, train_loss: 0.5835\n",
      "batch 5/5, train_loss: 0.5911\n",
      "epoch 138 average loss: 0.6190\n",
      "current epoch: 138 current mean recall per class: 0, 0.4428, 0.3236, 0.2985, 0.5312, 0.4273, 0.3741\n",
      "best mean recall: 0.3628 at epoch: 136\n",
      "----------\n",
      "epoch 139/200\n",
      "batch 1/5, train_loss: 0.5943\n",
      "batch 2/5, train_loss: 0.5748\n",
      "batch 3/5, train_loss: 0.5970\n",
      "batch 4/5, train_loss: 0.6175\n",
      "batch 5/5, train_loss: 0.6657\n",
      "epoch 139 average loss: 0.6099\n",
      "----------\n",
      "epoch 140/200\n",
      "batch 1/5, train_loss: 0.6546\n",
      "batch 2/5, train_loss: 0.5815\n",
      "batch 3/5, train_loss: 0.6015\n",
      "batch 4/5, train_loss: 0.6472\n",
      "batch 5/5, train_loss: 0.5934\n",
      "epoch 140 average loss: 0.6156\n",
      "current epoch: 140 current mean recall per class: 0, 0.4253, 0.3535, 0.314, 0.3549, 0.3589, 0.4672\n",
      "best mean recall: 0.3628 at epoch: 136\n",
      "----------\n",
      "epoch 141/200\n",
      "batch 1/5, train_loss: 0.6594\n",
      "batch 2/5, train_loss: 0.5875\n",
      "batch 3/5, train_loss: 0.6057\n",
      "batch 4/5, train_loss: 0.5878\n",
      "batch 5/5, train_loss: 0.6473\n",
      "epoch 141 average loss: 0.6175\n",
      "----------\n",
      "epoch 142/200\n",
      "batch 1/5, train_loss: 0.6042\n",
      "batch 2/5, train_loss: 0.6049\n",
      "batch 3/5, train_loss: 0.6275\n",
      "batch 4/5, train_loss: 0.6137\n",
      "batch 5/5, train_loss: 0.5995\n",
      "epoch 142 average loss: 0.6100\n",
      "current epoch: 142 current mean recall per class: 0, 0.4303, 0.3198, 0.2795, 0.5028, 0.3904, 0.4839\n",
      "best mean recall: 0.3628 at epoch: 136\n",
      "----------\n",
      "epoch 143/200\n",
      "batch 1/5, train_loss: 0.6518\n",
      "batch 2/5, train_loss: 0.5939\n",
      "batch 3/5, train_loss: 0.6114\n",
      "batch 4/5, train_loss: 0.5561\n",
      "batch 5/5, train_loss: 0.6211\n",
      "epoch 143 average loss: 0.6069\n",
      "----------\n",
      "epoch 144/200\n",
      "batch 1/5, train_loss: 0.6058\n",
      "batch 2/5, train_loss: 0.5960\n",
      "batch 3/5, train_loss: 0.6395\n",
      "batch 4/5, train_loss: 0.6367\n",
      "batch 5/5, train_loss: 0.5322\n",
      "epoch 144 average loss: 0.6020\n",
      "current epoch: 144 current mean recall per class: 0, 0.3671, 0.3325, 0.2734, 0.5621, 0.5197, 0.3205\n",
      "best mean recall: 0.3628 at epoch: 136\n",
      "----------\n",
      "epoch 145/200\n",
      "batch 1/5, train_loss: 0.6801\n",
      "batch 2/5, train_loss: 0.5586\n",
      "batch 3/5, train_loss: 0.5807\n",
      "batch 4/5, train_loss: 0.6072\n",
      "batch 5/5, train_loss: 0.5915\n",
      "epoch 145 average loss: 0.6036\n",
      "----------\n",
      "epoch 146/200\n",
      "batch 1/5, train_loss: 0.5416\n",
      "batch 2/5, train_loss: 0.5950\n",
      "batch 3/5, train_loss: 0.6639\n",
      "batch 4/5, train_loss: 0.6853\n",
      "batch 5/5, train_loss: 0.5783\n",
      "epoch 146 average loss: 0.6128\n",
      "current epoch: 146 current mean recall per class: 0, 0.3385, 0.3149, 0.2878, 0.508, 0.5016, 0.2995\n",
      "best mean recall: 0.3628 at epoch: 136\n",
      "----------\n",
      "epoch 147/200\n",
      "batch 1/5, train_loss: 0.6078\n",
      "batch 2/5, train_loss: 0.6486\n",
      "batch 3/5, train_loss: 0.6112\n",
      "batch 4/5, train_loss: 0.5784\n",
      "batch 5/5, train_loss: 0.6473\n",
      "epoch 147 average loss: 0.6187\n",
      "----------\n",
      "epoch 148/200\n",
      "batch 1/5, train_loss: 0.6642\n",
      "batch 2/5, train_loss: 0.5770\n",
      "batch 3/5, train_loss: 0.6223\n",
      "batch 4/5, train_loss: 0.5984\n",
      "batch 5/5, train_loss: 0.5782\n",
      "epoch 148 average loss: 0.6080\n",
      "current epoch: 148 current mean recall per class: 0, 0.4215, 0.3018, 0.2686, 0.5095, 0.4629, 0.4036\n",
      "best mean recall: 0.3628 at epoch: 136\n",
      "----------\n",
      "epoch 149/200\n",
      "batch 1/5, train_loss: 0.5832\n",
      "batch 2/5, train_loss: 0.5774\n",
      "batch 3/5, train_loss: 0.5862\n",
      "batch 4/5, train_loss: 0.6813\n",
      "batch 5/5, train_loss: 0.6456\n",
      "epoch 149 average loss: 0.6147\n",
      "----------\n",
      "epoch 150/200\n",
      "batch 1/5, train_loss: 0.6587\n",
      "batch 2/5, train_loss: 0.6224\n",
      "batch 3/5, train_loss: 0.6289\n",
      "batch 4/5, train_loss: 0.5961\n",
      "batch 5/5, train_loss: 0.6169\n",
      "epoch 150 average loss: 0.6246\n",
      "current epoch: 150 current mean recall per class: 4.432e-06, 0.3687, 0.3699, 0.3122, 0.6161, 0.533, 0.3091\n",
      "best mean recall: 0.3628 at epoch: 136\n",
      "----------\n",
      "epoch 151/200\n",
      "batch 1/5, train_loss: 0.6426\n",
      "batch 2/5, train_loss: 0.5962\n",
      "batch 3/5, train_loss: 0.6132\n",
      "batch 4/5, train_loss: 0.5534\n",
      "batch 5/5, train_loss: 0.6745\n",
      "epoch 151 average loss: 0.6160\n",
      "----------\n",
      "epoch 152/200\n",
      "batch 1/5, train_loss: 0.6050\n",
      "batch 2/5, train_loss: 0.6161\n",
      "batch 3/5, train_loss: 0.6691\n",
      "batch 4/5, train_loss: 0.6251\n",
      "batch 5/5, train_loss: 0.5797\n",
      "epoch 152 average loss: 0.6190\n",
      "current epoch: 152 current mean recall per class: 0, 0.3619, 0.3195, 0.3147, 0.4635, 0.5423, 0.3826\n",
      "best mean recall: 0.3628 at epoch: 136\n",
      "----------\n",
      "epoch 153/200\n",
      "batch 1/5, train_loss: 0.6641\n",
      "batch 2/5, train_loss: 0.6090\n",
      "batch 3/5, train_loss: 0.5949\n",
      "batch 4/5, train_loss: 0.5767\n",
      "batch 5/5, train_loss: 0.6018\n",
      "epoch 153 average loss: 0.6093\n",
      "----------\n",
      "epoch 154/200\n",
      "batch 1/5, train_loss: 0.6460\n",
      "batch 2/5, train_loss: 0.5608\n",
      "batch 3/5, train_loss: 0.5974\n",
      "batch 4/5, train_loss: 0.5829\n",
      "batch 5/5, train_loss: 0.6345\n",
      "epoch 154 average loss: 0.6043\n",
      "current epoch: 154 current mean recall per class: 0, 0.3676, 0.3011, 0.3433, 0.3644, 0.5891, 0.3228\n",
      "best mean recall: 0.3628 at epoch: 136\n",
      "----------\n",
      "epoch 155/200\n",
      "batch 1/5, train_loss: 0.5596\n",
      "batch 2/5, train_loss: 0.6173\n",
      "batch 3/5, train_loss: 0.5704\n",
      "batch 4/5, train_loss: 0.5717\n",
      "batch 5/5, train_loss: 0.6701\n",
      "epoch 155 average loss: 0.5978\n",
      "----------\n",
      "epoch 156/200\n",
      "batch 1/5, train_loss: 0.6042\n",
      "batch 2/5, train_loss: 0.6590\n",
      "batch 3/5, train_loss: 0.6007\n",
      "batch 4/5, train_loss: 0.5792\n",
      "batch 5/5, train_loss: 0.5649\n",
      "epoch 156 average loss: 0.6016\n",
      "saved new best metric model\n",
      "current epoch: 156 current mean recall per class: 0, 0.3926, 0.3306, 0.302, 0.3807, 0.6596, 0.4832\n",
      "best mean recall: 0.3641 at epoch: 156\n",
      "----------\n",
      "epoch 157/200\n",
      "batch 1/5, train_loss: 0.6200\n",
      "batch 2/5, train_loss: 0.6460\n",
      "batch 3/5, train_loss: 0.6188\n",
      "batch 4/5, train_loss: 0.5883\n",
      "batch 5/5, train_loss: 0.5854\n",
      "epoch 157 average loss: 0.6117\n",
      "----------\n",
      "epoch 158/200\n",
      "batch 1/5, train_loss: 0.6297\n",
      "batch 2/5, train_loss: 0.5530\n",
      "batch 3/5, train_loss: 0.6265\n",
      "batch 4/5, train_loss: 0.6311\n",
      "batch 5/5, train_loss: 0.5933\n",
      "epoch 158 average loss: 0.6067\n",
      "current epoch: 158 current mean recall per class: 0, 0.3995, 0.3349, 0.3325, 0.4389, 0.5179, 0.4511\n",
      "best mean recall: 0.3641 at epoch: 156\n",
      "----------\n",
      "epoch 159/200\n",
      "batch 1/5, train_loss: 0.5851\n",
      "batch 2/5, train_loss: 0.6224\n",
      "batch 3/5, train_loss: 0.5772\n",
      "batch 4/5, train_loss: 0.6567\n",
      "batch 5/5, train_loss: 0.5735\n",
      "epoch 159 average loss: 0.6030\n",
      "----------\n",
      "epoch 160/200\n",
      "batch 1/5, train_loss: 0.5562\n",
      "batch 2/5, train_loss: 0.6274\n",
      "batch 3/5, train_loss: 0.5700\n",
      "batch 4/5, train_loss: 0.6340\n",
      "batch 5/5, train_loss: 0.6048\n",
      "epoch 160 average loss: 0.5985\n",
      "current epoch: 160 current mean recall per class: 0, 0.4415, 0.3775, 0.2806, 0.5817, 0.4348, 0.3765\n",
      "best mean recall: 0.3641 at epoch: 156\n",
      "----------\n",
      "epoch 161/200\n",
      "batch 1/5, train_loss: 0.6231\n",
      "batch 2/5, train_loss: 0.5850\n",
      "batch 3/5, train_loss: 0.5691\n",
      "batch 4/5, train_loss: 0.5944\n",
      "batch 5/5, train_loss: 0.6407\n",
      "epoch 161 average loss: 0.6025\n",
      "----------\n",
      "epoch 162/200\n",
      "batch 1/5, train_loss: 0.6328\n",
      "batch 2/5, train_loss: 0.5837\n",
      "batch 3/5, train_loss: 0.5690\n",
      "batch 4/5, train_loss: 0.5545\n",
      "batch 5/5, train_loss: 0.6557\n",
      "epoch 162 average loss: 0.5991\n",
      "current epoch: 162 current mean recall per class: 0, 0.3933, 0.3635, 0.2649, 0.4187, 0.4472, 0.3558\n",
      "best mean recall: 0.3641 at epoch: 156\n",
      "----------\n",
      "epoch 163/200\n",
      "batch 1/5, train_loss: 0.5783\n",
      "batch 2/5, train_loss: 0.6320\n",
      "batch 3/5, train_loss: 0.5895\n",
      "batch 4/5, train_loss: 0.6368\n",
      "batch 5/5, train_loss: 0.6210\n",
      "epoch 163 average loss: 0.6115\n",
      "----------\n",
      "epoch 164/200\n",
      "batch 1/5, train_loss: 0.5770\n",
      "batch 2/5, train_loss: 0.6283\n",
      "batch 3/5, train_loss: 0.5851\n",
      "batch 4/5, train_loss: 0.5722\n",
      "batch 5/5, train_loss: 0.6527\n",
      "epoch 164 average loss: 0.6031\n",
      "current epoch: 164 current mean recall per class: 0, 0.3421, 0.3579, 0.258, 0.5276, 0.532, 0.465\n",
      "best mean recall: 0.3641 at epoch: 156\n",
      "----------\n",
      "epoch 165/200\n",
      "batch 1/5, train_loss: 0.5867\n",
      "batch 2/5, train_loss: 0.6207\n",
      "batch 3/5, train_loss: 0.6151\n",
      "batch 4/5, train_loss: 0.6370\n",
      "batch 5/5, train_loss: 0.6395\n",
      "epoch 165 average loss: 0.6198\n",
      "----------\n",
      "epoch 166/200\n",
      "batch 1/5, train_loss: 0.5789\n",
      "batch 2/5, train_loss: 0.6042\n",
      "batch 3/5, train_loss: 0.6313\n",
      "batch 4/5, train_loss: 0.6306\n",
      "batch 5/5, train_loss: 0.5781\n",
      "epoch 166 average loss: 0.6046\n",
      "current epoch: 166 current mean recall per class: 0, 0.3792, 0.3522, 0.2393, 0.5681, 0.5986, 0.3816\n",
      "best mean recall: 0.3641 at epoch: 156\n",
      "----------\n",
      "epoch 167/200\n",
      "batch 1/5, train_loss: 0.5698\n",
      "batch 2/5, train_loss: 0.5830\n",
      "batch 3/5, train_loss: 0.6036\n",
      "batch 4/5, train_loss: 0.6404\n",
      "batch 5/5, train_loss: 0.5966\n",
      "epoch 167 average loss: 0.5987\n",
      "----------\n",
      "epoch 168/200\n",
      "batch 1/5, train_loss: 0.5924\n",
      "batch 2/5, train_loss: 0.6421\n",
      "batch 3/5, train_loss: 0.5444\n",
      "batch 4/5, train_loss: 0.5992\n",
      "batch 5/5, train_loss: 0.5475\n",
      "epoch 168 average loss: 0.5851\n",
      "saved new best metric model\n",
      "current epoch: 168 current mean recall per class: 0, 0.427, 0.3024, 0.233, 0.6329, 0.5499, 0.4753\n",
      "best mean recall: 0.3744 at epoch: 168\n",
      "----------\n",
      "epoch 169/200\n",
      "batch 1/5, train_loss: 0.5867\n",
      "batch 2/5, train_loss: 0.6524\n",
      "batch 3/5, train_loss: 0.6045\n",
      "batch 4/5, train_loss: 0.6322\n",
      "batch 5/5, train_loss: 0.6065\n",
      "epoch 169 average loss: 0.6164\n",
      "----------\n",
      "epoch 170/200\n",
      "batch 1/5, train_loss: 0.6437\n",
      "batch 2/5, train_loss: 0.5848\n",
      "batch 3/5, train_loss: 0.5717\n",
      "batch 4/5, train_loss: 0.6288\n",
      "batch 5/5, train_loss: 0.5785\n",
      "epoch 170 average loss: 0.6015\n",
      "current epoch: 170 current mean recall per class: 0, 0.3336, 0.3269, 0.2793, 0.5618, 0.4072, 0.3525\n",
      "best mean recall: 0.3744 at epoch: 168\n",
      "----------\n",
      "epoch 171/200\n",
      "batch 1/5, train_loss: 0.6052\n",
      "batch 2/5, train_loss: 0.6306\n",
      "batch 3/5, train_loss: 0.5914\n",
      "batch 4/5, train_loss: 0.5451\n",
      "batch 5/5, train_loss: 0.6572\n",
      "epoch 171 average loss: 0.6059\n",
      "----------\n",
      "epoch 172/200\n",
      "batch 1/5, train_loss: 0.5886\n",
      "batch 2/5, train_loss: 0.5788\n",
      "batch 3/5, train_loss: 0.6622\n",
      "batch 4/5, train_loss: 0.6086\n",
      "batch 5/5, train_loss: 0.6360\n",
      "epoch 172 average loss: 0.6148\n",
      "current epoch: 172 current mean recall per class: 0, 0.3779, 0.3328, 0.2053, 0.531, 0.5649, 0.3993\n",
      "best mean recall: 0.3744 at epoch: 168\n",
      "----------\n",
      "epoch 173/200\n",
      "batch 1/5, train_loss: 0.6201\n",
      "batch 2/5, train_loss: 0.5522\n",
      "batch 3/5, train_loss: 0.6046\n",
      "batch 4/5, train_loss: 0.6497\n",
      "batch 5/5, train_loss: 0.6163\n",
      "epoch 173 average loss: 0.6086\n",
      "----------\n",
      "epoch 174/200\n",
      "batch 1/5, train_loss: 0.6061\n",
      "batch 2/5, train_loss: 0.6449\n",
      "batch 3/5, train_loss: 0.5657\n",
      "batch 4/5, train_loss: 0.5973\n",
      "batch 5/5, train_loss: 0.6584\n",
      "epoch 174 average loss: 0.6145\n",
      "current epoch: 174 current mean recall per class: 0, 0.4315, 0.3353, 0.2712, 0.4708, 0.4089, 0.2196\n",
      "best mean recall: 0.3744 at epoch: 168\n",
      "----------\n",
      "epoch 175/200\n",
      "batch 1/5, train_loss: 0.6143\n",
      "batch 2/5, train_loss: 0.6581\n",
      "batch 3/5, train_loss: 0.6160\n",
      "batch 4/5, train_loss: 0.6123\n",
      "batch 5/5, train_loss: 0.6227\n",
      "epoch 175 average loss: 0.6247\n",
      "----------\n",
      "epoch 176/200\n",
      "batch 1/5, train_loss: 0.6018\n",
      "batch 2/5, train_loss: 0.5947\n",
      "batch 3/5, train_loss: 0.6447\n",
      "batch 4/5, train_loss: 0.6068\n",
      "batch 5/5, train_loss: 0.6058\n",
      "epoch 176 average loss: 0.6107\n",
      "current epoch: 176 current mean recall per class: 0, 0.4777, 0.3325, 0.2822, 0.4592, 0.4672, 0.3941\n",
      "best mean recall: 0.3744 at epoch: 168\n",
      "----------\n",
      "epoch 177/200\n",
      "batch 1/5, train_loss: 0.6033\n",
      "batch 2/5, train_loss: 0.6523\n",
      "batch 3/5, train_loss: 0.6052\n",
      "batch 4/5, train_loss: 0.5900\n",
      "batch 5/5, train_loss: 0.5955\n",
      "epoch 177 average loss: 0.6092\n",
      "----------\n",
      "epoch 178/200\n",
      "batch 1/5, train_loss: 0.6177\n",
      "batch 2/5, train_loss: 0.6132\n",
      "batch 3/5, train_loss: 0.5805\n",
      "batch 4/5, train_loss: 0.6485\n",
      "batch 5/5, train_loss: 0.5888\n",
      "epoch 178 average loss: 0.6097\n",
      "current epoch: 178 current mean recall per class: 0, 0.3876, 0.3191, 0.3565, 0.5149, 0.5141, 0.4546\n",
      "best mean recall: 0.3744 at epoch: 168\n",
      "----------\n",
      "epoch 179/200\n",
      "batch 1/5, train_loss: 0.6277\n",
      "batch 2/5, train_loss: 0.5374\n",
      "batch 3/5, train_loss: 0.6001\n",
      "batch 4/5, train_loss: 0.6288\n",
      "batch 5/5, train_loss: 0.6445\n",
      "epoch 179 average loss: 0.6077\n",
      "----------\n",
      "epoch 180/200\n",
      "batch 1/5, train_loss: 0.6232\n",
      "batch 2/5, train_loss: 0.6412\n",
      "batch 3/5, train_loss: 0.6221\n",
      "batch 4/5, train_loss: 0.6102\n",
      "batch 5/5, train_loss: 0.6477\n",
      "epoch 180 average loss: 0.6289\n",
      "current epoch: 180 current mean recall per class: 0, 0.4039, 0.3208, 0.28, 0.3763, 0.4919, 0.3705\n",
      "best mean recall: 0.3744 at epoch: 168\n",
      "----------\n",
      "epoch 181/200\n",
      "batch 1/5, train_loss: 0.6056\n",
      "batch 2/5, train_loss: 0.6553\n",
      "batch 3/5, train_loss: 0.5901\n",
      "batch 4/5, train_loss: 0.5650\n",
      "batch 5/5, train_loss: 0.6370\n",
      "epoch 181 average loss: 0.6106\n",
      "----------\n",
      "epoch 182/200\n",
      "batch 1/5, train_loss: 0.5693\n",
      "batch 2/5, train_loss: 0.5649\n",
      "batch 3/5, train_loss: 0.5731\n",
      "batch 4/5, train_loss: 0.6380\n",
      "batch 5/5, train_loss: 0.5806\n",
      "epoch 182 average loss: 0.5852\n",
      "current epoch: 182 current mean recall per class: 0, 0.3878, 0.3308, 0.2914, 0.3586, 0.5259, 0.352\n",
      "best mean recall: 0.3744 at epoch: 168\n",
      "----------\n",
      "epoch 183/200\n",
      "batch 1/5, train_loss: 0.5795\n",
      "batch 2/5, train_loss: 0.5880\n",
      "batch 3/5, train_loss: 0.6492\n",
      "batch 4/5, train_loss: 0.6607\n",
      "batch 5/5, train_loss: 0.5500\n",
      "epoch 183 average loss: 0.6055\n",
      "----------\n",
      "epoch 184/200\n",
      "batch 1/5, train_loss: 0.6151\n",
      "batch 2/5, train_loss: 0.5818\n",
      "batch 3/5, train_loss: 0.6768\n",
      "batch 4/5, train_loss: 0.5167\n",
      "batch 5/5, train_loss: 0.6418\n",
      "epoch 184 average loss: 0.6064\n",
      "current epoch: 184 current mean recall per class: 0, 0.4211, 0.2991, 0.2771, 0.4521, 0.4612, 0.4351\n",
      "best mean recall: 0.3744 at epoch: 168\n",
      "----------\n",
      "epoch 185/200\n",
      "batch 1/5, train_loss: 0.5974\n",
      "batch 2/5, train_loss: 0.6349\n",
      "batch 3/5, train_loss: 0.6424\n",
      "batch 4/5, train_loss: 0.5781\n",
      "batch 5/5, train_loss: 0.6494\n",
      "epoch 185 average loss: 0.6204\n",
      "----------\n",
      "epoch 186/200\n",
      "batch 1/5, train_loss: 0.6391\n",
      "batch 2/5, train_loss: 0.5997\n",
      "batch 3/5, train_loss: 0.5414\n",
      "batch 4/5, train_loss: 0.6172\n",
      "batch 5/5, train_loss: 0.5833\n",
      "epoch 186 average loss: 0.5961\n",
      "current epoch: 186 current mean recall per class: 0, 0.4647, 0.3811, 0.1821, 0.4514, 0.496, 0.4758\n",
      "best mean recall: 0.3744 at epoch: 168\n",
      "----------\n",
      "epoch 187/200\n",
      "batch 1/5, train_loss: 0.5996\n",
      "batch 2/5, train_loss: 0.6608\n",
      "batch 3/5, train_loss: 0.6084\n",
      "batch 4/5, train_loss: 0.5887\n",
      "batch 5/5, train_loss: 0.6398\n",
      "epoch 187 average loss: 0.6195\n",
      "----------\n",
      "epoch 188/200\n",
      "batch 1/5, train_loss: 0.5597\n",
      "batch 2/5, train_loss: 0.5837\n",
      "batch 3/5, train_loss: 0.5583\n",
      "batch 4/5, train_loss: 0.6334\n",
      "batch 5/5, train_loss: 0.6209\n",
      "epoch 188 average loss: 0.5912\n",
      "current epoch: 188 current mean recall per class: 0, 0.4213, 0.3308, 0.2326, 0.4255, 0.5068, 0.3163\n",
      "best mean recall: 0.3744 at epoch: 168\n",
      "----------\n",
      "epoch 189/200\n",
      "batch 1/5, train_loss: 0.6014\n",
      "batch 2/5, train_loss: 0.6320\n",
      "batch 3/5, train_loss: 0.6638\n",
      "batch 4/5, train_loss: 0.5570\n",
      "batch 5/5, train_loss: 0.5853\n",
      "epoch 189 average loss: 0.6079\n",
      "----------\n",
      "epoch 190/200\n",
      "batch 1/5, train_loss: 0.6464\n",
      "batch 2/5, train_loss: 0.6175\n",
      "batch 3/5, train_loss: 0.5603\n",
      "batch 4/5, train_loss: 0.5754\n",
      "batch 5/5, train_loss: 0.6328\n",
      "epoch 190 average loss: 0.6065\n",
      "current epoch: 190 current mean recall per class: 0, 0.383, 0.3641, 0.3043, 0.5037, 0.4928, 0.4459\n",
      "best mean recall: 0.3744 at epoch: 168\n",
      "----------\n",
      "epoch 191/200\n",
      "batch 1/5, train_loss: 0.6296\n",
      "batch 2/5, train_loss: 0.6027\n",
      "batch 3/5, train_loss: 0.6276\n",
      "batch 4/5, train_loss: 0.5800\n",
      "batch 5/5, train_loss: 0.6387\n",
      "epoch 191 average loss: 0.6157\n",
      "----------\n",
      "epoch 192/200\n",
      "batch 1/5, train_loss: 0.6474\n",
      "batch 2/5, train_loss: 0.6333\n",
      "batch 3/5, train_loss: 0.6195\n",
      "batch 4/5, train_loss: 0.5779\n",
      "batch 5/5, train_loss: 0.5779\n",
      "epoch 192 average loss: 0.6112\n",
      "saved new best metric model\n",
      "current epoch: 192 current mean recall per class: 0, 0.4241, 0.3459, 0.3157, 0.5804, 0.5758, 0.3976\n",
      "best mean recall: 0.3771 at epoch: 192\n",
      "----------\n",
      "epoch 193/200\n",
      "batch 1/5, train_loss: 0.5451\n",
      "batch 2/5, train_loss: 0.5879\n",
      "batch 3/5, train_loss: 0.6029\n",
      "batch 4/5, train_loss: 0.5754\n",
      "batch 5/5, train_loss: 0.6054\n",
      "epoch 193 average loss: 0.5833\n",
      "----------\n",
      "epoch 194/200\n",
      "batch 1/5, train_loss: 0.5611\n",
      "batch 2/5, train_loss: 0.5669\n",
      "batch 3/5, train_loss: 0.6597\n",
      "batch 4/5, train_loss: 0.6281\n",
      "batch 5/5, train_loss: 0.5840\n",
      "epoch 194 average loss: 0.6000\n",
      "current epoch: 194 current mean recall per class: 0, 0.4343, 0.3403, 0.3411, 0.5967, 0.4002, 0.3217\n",
      "best mean recall: 0.3771 at epoch: 192\n",
      "----------\n",
      "epoch 195/200\n",
      "batch 1/5, train_loss: 0.5989\n",
      "batch 2/5, train_loss: 0.6013\n",
      "batch 3/5, train_loss: 0.5603\n",
      "batch 4/5, train_loss: 0.5965\n",
      "batch 5/5, train_loss: 0.6466\n",
      "epoch 195 average loss: 0.6007\n",
      "----------\n",
      "epoch 196/200\n",
      "batch 1/5, train_loss: 0.6294\n",
      "batch 2/5, train_loss: 0.5948\n",
      "batch 3/5, train_loss: 0.5625\n",
      "batch 4/5, train_loss: 0.5832\n",
      "batch 5/5, train_loss: 0.5586\n",
      "epoch 196 average loss: 0.5857\n",
      "current epoch: 196 current mean recall per class: 1.326e-05, 0.5234, 0.3215, 0.367, 0.5597, 0.4415, 0.3703\n",
      "best mean recall: 0.3771 at epoch: 192\n",
      "----------\n",
      "epoch 197/200\n",
      "batch 1/5, train_loss: 0.5646\n",
      "batch 2/5, train_loss: 0.5963\n",
      "batch 3/5, train_loss: 0.6608\n",
      "batch 4/5, train_loss: 0.6053\n",
      "batch 5/5, train_loss: 0.5719\n",
      "epoch 197 average loss: 0.5998\n",
      "----------\n",
      "epoch 198/200\n",
      "batch 1/5, train_loss: 0.5583\n",
      "batch 2/5, train_loss: 0.5586\n",
      "batch 3/5, train_loss: 0.6236\n",
      "batch 4/5, train_loss: 0.5782\n",
      "batch 5/5, train_loss: 0.6139\n",
      "epoch 198 average loss: 0.5865\n",
      "current epoch: 198 current mean recall per class: 0, 0.402, 0.3045, 0.31, 0.5168, 0.4688, 0.3588\n",
      "best mean recall: 0.3771 at epoch: 192\n",
      "----------\n",
      "epoch 199/200\n",
      "batch 1/5, train_loss: 0.6193\n",
      "batch 2/5, train_loss: 0.5808\n",
      "batch 3/5, train_loss: 0.5365\n",
      "batch 4/5, train_loss: 0.5934\n",
      "batch 5/5, train_loss: 0.6445\n",
      "epoch 199 average loss: 0.5949\n",
      "----------\n",
      "epoch 200/200\n",
      "batch 1/5, train_loss: 0.6020\n",
      "batch 2/5, train_loss: 0.5568\n",
      "batch 3/5, train_loss: 0.6023\n",
      "batch 4/5, train_loss: 0.6117\n",
      "batch 5/5, train_loss: 0.6146\n",
      "epoch 200 average loss: 0.5975\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024/10/07 20:07:12 WARNING mlflow.utils.requirements_utils: Found torch version (2.4.0+cu124) contains a local version label (+cu124). MLflow logged a pip requirement for this package as 'torch==2.4.0' without the local version label to make it installable from PyPI. To specify pip requirements containing local version labels, please use `conda_env` or `pip_requirements`.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "current epoch: 200 current mean recall per class: 0, 0.3304, 0.3373, 0.3223, 0.4577, 0.5082, 0.4501\n",
      "best mean recall: 0.3771 at epoch: 192\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024/10/07 20:07:29 WARNING mlflow.utils.requirements_utils: Found torchvision version (0.19.0+cu124) contains a local version label (+cu124). MLflow logged a pip requirement for this package as 'torchvision==0.19.0' without the local version label to make it installable from PyPI. To specify pip requirements containing local version labels, please use `conda_env` or `pip_requirements`.\n",
      "2024/10/07 20:07:29 WARNING mlflow.utils.requirements_utils: Found torchaudio version (2.4.0+cu124) contains a local version label (+cu124). MLflow logged a pip requirement for this package as 'torchaudio==2.4.0' without the local version label to make it installable from PyPI. To specify pip requirements containing local version labels, please use `conda_env` or `pip_requirements`.\n",
      "/hpc/mydata/kevin.zhao/miniconda3/envs/mlchallenge/lib/python3.9/site-packages/_distutils_hack/__init__.py:32: UserWarning: Setuptools is replacing distutils. Support for replacing an already imported distutils is deprecated. In the future, this condition will fail. Register concerns at https://github.com/pypa/setuptools/issues/new?template=distutils-deprecation.yml\n",
      "  warnings.warn(\n",
      "2024/10/07 20:07:29 WARNING mlflow.models.model: Input example should be provided to infer model signature if the model signature is not provided when logging the model.\n"
     ]
    }
   ],
   "source": [
    "from torchinfo import summary\n",
    "\n",
    "mlflow.end_run()\n",
    "mlflow.set_experiment('training 3D U-Net model for the cryoET ML Challenge')\n",
    "epochs = 200\n",
    "with mlflow.start_run():\n",
    "    params = {\n",
    "        \"epochs\": epochs,\n",
    "        \"learning_rate\": lr,\n",
    "        \"loss_function\": loss_function.__class__.__name__,\n",
    "        \"metric_function\": recall_metric.__class__.__name__,\n",
    "        \"optimizer\": \"Adam\",\n",
    "    }\n",
    "    # Log training parameters.\n",
    "    mlflow.log_params(params)\n",
    "\n",
    "    # Log model summary.\n",
    "    with open(\"model_summary.txt\", \"w\") as f:\n",
    "        f.write(str(summary(model)))\n",
    "    mlflow.log_artifact(\"model_summary.txt\")\n",
    "\n",
    "    train(train_loader, model, loss_function, dice_metric, optimizer, max_epochs=epochs)\n",
    "\n",
    "    # Save the trained model to MLflow.\n",
    "    mlflow.pytorch.log_model(model, \"model\")\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6b72796c-6468-4211-aaf5-52d7a1b6fd4b",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "mlchallenge",
   "language": "python",
   "name": "mlchallenge"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.19"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}