{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.11.13","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"gpu","dataSources":[{"sourceId":4521,"databundleVersionId":326986,"sourceType":"competition"}],"dockerImageVersionId":31193,"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":true}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"### Aryan Chharia (102203313), Abhishek Verma (102203309), Ayush Goyal (102383007)\n### 4C17-19 Group","metadata":{}},{"cell_type":"code","source":"# This Python 3 environment comes with many helpful analytics libraries installed\n# It is defined by the kaggle/python Docker image: https://github.com/kaggle/docker-python\n# For example, here's several helpful packages to load\n\nimport numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\nimport matplotlib.pyplot as plt\n# Input data files are available in the read-only \"../input/\" directory\n# For example, running this (by clicking run or pressing Shift+Enter) will list all files under the input directory\n\nimport os\nfor dirname, _, filenames in os.walk('/kaggle/input'):\n    for filename in filenames:\n        print(os.path.join(dirname, filename))\n\n# You can write up to 20GB to the current directory (/kaggle/working/) that gets preserved as output when you create a version using \"Save & Run All\" \n# You can also write temporary files to /kaggle/temp/, but they won't be saved outside of the current session","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:48:52.183403Z","iopub.execute_input":"2025-11-28T13:48:52.183573Z","iopub.status.idle":"2025-11-28T13:48:54.040482Z","shell.execute_reply.started":"2025-11-28T13:48:52.183556Z","shell.execute_reply":"2025-11-28T13:48:54.039614Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# Fast unzip for Kaggle\nimport os\nimport subprocess\nimport time\nfrom zipfile import ZipFile\n\nSRC = \"/kaggle/input/noaa-right-whale-recognition/imgs.zip\"   # zip in read-only input\nDST = \"/kaggle/working/imgs\"                                 # writable extraction target\n\n# Safety: don't re-extract if already present\nif os.path.exists(DST) and os.listdir(DST):\n    print(f\"Already extracted to {DST} (skipping).\")\nelse:\n    os.makedirs(DST, exist_ok=True)\n    t0 = time.time()\n    try:\n        subprocess.run(['unzip', '-qq', SRC, '-d', DST], check=True)\n        print(f\"Unzipped with system 'unzip' into {DST} in {time.time() - t0:.1f}s\")\n    except FileNotFoundError:\n        print(\"System 'unzip' not available — falling back to Python ZipFile.extractall()\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:50:40.511106Z","iopub.execute_input":"2025-11-28T13:50:40.511867Z","iopub.status.idle":"2025-11-28T13:52:51.424786Z","shell.execute_reply.started":"2025-11-28T13:50:40.511838Z","shell.execute_reply":"2025-11-28T13:52:51.424105Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"df = pd.read_csv('/kaggle/input/noaa-right-whale-recognition/train.csv')","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:52:56.096276Z","iopub.execute_input":"2025-11-28T13:52:56.096559Z","iopub.status.idle":"2025-11-28T13:52:56.119454Z","shell.execute_reply.started":"2025-11-28T13:52:56.096537Z","shell.execute_reply":"2025-11-28T13:52:56.118737Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"df.head()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:52:56.455793Z","iopub.execute_input":"2025-11-28T13:52:56.456612Z","iopub.status.idle":"2025-11-28T13:52:56.478227Z","shell.execute_reply.started":"2025-11-28T13:52:56.456583Z","shell.execute_reply":"2025-11-28T13:52:56.477382Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"import random\nfrom pathlib import Path\nfrom PIL import Image","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:52:56.686924Z","iopub.execute_input":"2025-11-28T13:52:56.687243Z","iopub.status.idle":"2025-11-28T13:52:56.691034Z","shell.execute_reply.started":"2025-11-28T13:52:56.687206Z","shell.execute_reply":"2025-11-28T13:52:56.690224Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"imgs_dir = Path(\"/kaggle/working/imgs\")\nimg_paths = list(imgs_dir.rglob(\"*.jpg\"))\nchosen = random.choice(img_paths)\nimg = Image.open(chosen)\n\nplt.figure(figsize=(10,6))\nplt.imshow(img)\nplt.axis(\"off\")\nplt.title(chosen.name)\nplt.show()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:52:58.741124Z","iopub.execute_input":"2025-11-28T13:52:58.741769Z","iopub.status.idle":"2025-11-28T13:52:59.870262Z","shell.execute_reply.started":"2025-11-28T13:52:58.741742Z","shell.execute_reply":"2025-11-28T13:52:59.869251Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"img_paths[:5]","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:52:59.871530Z","iopub.execute_input":"2025-11-28T13:52:59.871805Z","iopub.status.idle":"2025-11-28T13:52:59.878155Z","shell.execute_reply.started":"2025-11-28T13:52:59.871782Z","shell.execute_reply":"2025-11-28T13:52:59.877419Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"imgs_dir = Path(\"/kaggle/working/imgs/imgs\")\nfilename = \"w_7440\" + \".jpg\"\n\nimg_path = next(imgs_dir.rglob(filename))  # returns a single Path\n\nimg = Image.open(img_path)\n\nplt.figure(figsize=(10, 6))\nplt.imshow(img)\nplt.axis(\"off\")\nplt.title(img_path.name)\nplt.show()\n","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:53:06.532430Z","iopub.execute_input":"2025-11-28T13:53:06.533166Z","iopub.status.idle":"2025-11-28T13:53:07.487815Z","shell.execute_reply.started":"2025-11-28T13:53:06.533139Z","shell.execute_reply":"2025-11-28T13:53:07.486997Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"imgs_dir = Path(\"/kaggle/working/imgs/imgs\")\nfilename = \"w_7489\" + \".jpg\"\n\nimg_path = next(imgs_dir.rglob(filename))  # returns a single Path\n\nimg = Image.open(img_path)\n\nplt.figure(figsize=(10, 6))\nplt.imshow(img)\nplt.axis(\"off\")\nplt.title(img_path.name)\nplt.show()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:53:16.622268Z","iopub.execute_input":"2025-11-28T13:53:16.622805Z","iopub.status.idle":"2025-11-28T13:53:16.718074Z","shell.execute_reply.started":"2025-11-28T13:53:16.622780Z","shell.execute_reply":"2025-11-28T13:53:16.716774Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"df = df[df['Image']!='w_7489.jpg']","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:53:18.098768Z","iopub.execute_input":"2025-11-28T13:53:18.099125Z","iopub.status.idle":"2025-11-28T13:53:18.107514Z","shell.execute_reply.started":"2025-11-28T13:53:18.099101Z","shell.execute_reply":"2025-11-28T13:53:18.106751Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"df.info","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:53:19.111866Z","iopub.execute_input":"2025-11-28T13:53:19.112631Z","iopub.status.idle":"2025-11-28T13:53:19.120122Z","shell.execute_reply.started":"2025-11-28T13:53:19.112599Z","shell.execute_reply":"2025-11-28T13:53:19.119345Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"print(len(df), len(df['whaleID'].unique()))","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:53:20.788404Z","iopub.execute_input":"2025-11-28T13:53:20.788735Z","iopub.status.idle":"2025-11-28T13:53:20.796757Z","shell.execute_reply.started":"2025-11-28T13:53:20.788713Z","shell.execute_reply":"2025-11-28T13:53:20.795983Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"counts_per_id = df['whaleID'].value_counts()\ncounts_per_id","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:53:21.026044Z","iopub.execute_input":"2025-11-28T13:53:21.026768Z","iopub.status.idle":"2025-11-28T13:53:21.037667Z","shell.execute_reply.started":"2025-11-28T13:53:21.026740Z","shell.execute_reply":"2025-11-28T13:53:21.036883Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"freq = counts_per_id.value_counts().sort_index()\nfreq","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:53:23.205777Z","iopub.execute_input":"2025-11-28T13:53:23.206143Z","iopub.status.idle":"2025-11-28T13:53:23.218801Z","shell.execute_reply.started":"2025-11-28T13:53:23.206118Z","shell.execute_reply":"2025-11-28T13:53:23.218053Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"fig, ax1 = plt.subplots(1, 1, figsize=(14, 8))\n\nx = freq.index.values\ny = freq.values\n\nax1.bar(x, y, edgecolor='black')\nax1.set_xlabel(\"Number of images per whale ID\")\nax1.set_ylabel(\"Number of whale IDs\")\nax1.set_title(\"Count of whale IDs by number of images\")\nax1.grid(axis='y', linestyle='--', alpha=0.4)\n\n# reduce x-tick clutter: show at most ~20 ticks evenly spaced\nmax_ticks = 20\nif len(x) > max_ticks:\n    step = max(1, int((x.max() - x.min()) / (max_ticks - 1)))\n    ticks = np.arange(x.min(), x.max() + 1, step)\nelse:\n    ticks = x\n\nax1.set_xticks(ticks)\nax1.set_xticklabels([str(t) for t in ticks], rotation=45)\n\nplt.tight_layout()\nplt.show()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:53:28.002190Z","iopub.execute_input":"2025-11-28T13:53:28.002782Z","iopub.status.idle":"2025-11-28T13:53:28.288281Z","shell.execute_reply.started":"2025-11-28T13:53:28.002755Z","shell.execute_reply":"2025-11-28T13:53:28.287450Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"imgs_dir = Path(\"/kaggle/working/imgs\")\nimg_paths = list(imgs_dir.rglob(\"*.jpg\"))\n\nwidths = []\nheights = []\n\nfor p in img_paths:\n    with Image.open(p) as img:\n        w, h = img.size\n        widths.append(w)\n        heights.append(h)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:53:33.122195Z","iopub.execute_input":"2025-11-28T13:53:33.122472Z","iopub.status.idle":"2025-11-28T13:53:33.930860Z","shell.execute_reply.started":"2025-11-28T13:53:33.122452Z","shell.execute_reply":"2025-11-28T13:53:33.930128Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"print(len(widths), len(heights))","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:53:33.932119Z","iopub.execute_input":"2025-11-28T13:53:33.932440Z","iopub.status.idle":"2025-11-28T13:53:33.936896Z","shell.execute_reply.started":"2025-11-28T13:53:33.932415Z","shell.execute_reply":"2025-11-28T13:53:33.935934Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"plt.figure(figsize=(12, 5))\n\n# WIDTH HISTOGRAM\nplt.subplot(1, 2, 1)\nplt.hist(widths, bins=30, edgecolor='black', alpha=0.8)\nplt.title(\"Image Width Distribution\", fontsize=14)\nplt.xlabel(\"Width (pixels)\", fontsize=12)\nplt.ylabel(\"Number of Images\", fontsize=12)\nplt.grid(axis='y', linestyle='--', alpha=0.4)\n\n# stats on width\nw_mean = int(np.mean(widths))\nw_median = int(np.median(widths))\nw_min = min(widths)\nw_max = max(widths)\n\nplt.text(0.98, 0.95,\n         f\"Mean: {w_mean}\\nMedian: {w_median}\\nMin: {w_min}\\nMax: {w_max}\",\n         ha='right', va='top', transform=plt.gca().transAxes,\n         fontsize=10, bbox=dict(facecolor='white', alpha=0.7))\n\n# HEIGHT HISTOGRAM\nplt.subplot(1, 2, 2)\nplt.hist(heights, bins=30, edgecolor='black', alpha=0.8)\nplt.title(\"Image Height Distribution\", fontsize=14)\nplt.xlabel(\"Height (pixels)\", fontsize=12)\nplt.ylabel(\"Number of Images\", fontsize=12)\nplt.grid(axis='y', linestyle='--', alpha=0.4)\n\n# stats on height\nh_mean = int(np.mean(heights))\nh_median = int(np.median(heights))\nh_min = min(heights)\nh_max = max(heights)\n\nplt.text(0.98, 0.95,\n         f\"Mean: {h_mean}\\nMedian: {h_median}\\nMin: {h_min}\\nMax: {h_max}\",\n         ha='right', va='top', transform=plt.gca().transAxes,\n         fontsize=10, bbox=dict(facecolor='white', alpha=0.7))\n\nplt.tight_layout()\nplt.show()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:53:36.681366Z","iopub.execute_input":"2025-11-28T13:53:36.682032Z","iopub.status.idle":"2025-11-28T13:53:37.112741Z","shell.execute_reply.started":"2025-11-28T13:53:36.682003Z","shell.execute_reply":"2025-11-28T13:53:37.111928Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"plt.figure(figsize=(10, 5))\n\n# first histogram (widths)\nplt.hist(widths, bins=30, alpha=0.6, edgecolor='black', label='Widths')\n\n# create a second axis sharing x or y\nax2 = plt.gca().twinx()\n\n# second histogram (heights)\nax2.hist(heights, bins=30, alpha=0.6, edgecolor='black', color='orange', label='Heights')\n\nplt.title(\"Combined Histogram of Image Widths and Heights\", fontsize=14)\nplt.xlabel(\"Pixel Value (Width / Height)\", fontsize=12)\nplt.grid(alpha=0.3, linestyle='--')\n\n# legends (from both axes)\nplt.legend(loc=\"upper left\")\nax2.legend(loc=\"upper right\")\n\nplt.show()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:53:39.302197Z","iopub.execute_input":"2025-11-28T13:53:39.302468Z","iopub.status.idle":"2025-11-28T13:53:39.631349Z","shell.execute_reply.started":"2025-11-28T13:53:39.302446Z","shell.execute_reply":"2025-11-28T13:53:39.630599Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"import shutil\nfrom pathlib import Path\n\nout_dir = Path(\"/kaggle/working/imgs_resized\")\n\n# Delete the directory if it exists\nif out_dir.exists():\n    shutil.rmtree(out_dir)\n\nprint(\"Output directory deleted.\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:53:45.139696Z","iopub.execute_input":"2025-11-28T13:53:45.140005Z","iopub.status.idle":"2025-11-28T13:53:45.144753Z","shell.execute_reply.started":"2025-11-28T13:53:45.139976Z","shell.execute_reply":"2025-11-28T13:53:45.143994Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"from multiprocessing import Pool, cpu_count\nfrom tqdm import tqdm\n\nimgs_dir = Path(\"/kaggle/working/imgs\")\nout_dir = Path(\"/kaggle/working/imgs_resized\")\nout_dir.mkdir(exist_ok=True)\n\nimg_paths = list(imgs_dir.rglob(\"*.jpg\"))\nmax_side = 480\n\n\ndef resize_one(path):\n    try:\n        with Image.open(path) as img:\n            img.thumbnail((max_side, max_side), Image.LANCZOS)\n            img.save(out_dir / path.name)\n        return True\n    except:\n        return False\n\n# multiprocessing with tqdm progress bar\nwith Pool(cpu_count()) as p:\n    list(tqdm(p.imap_unordered(resize_one, img_paths, chunksize=20),\n              total=len(img_paths),\n              desc=\"Resizing images\",\n              ncols=90))\n\nprint(\"Resized images saved to:\", out_dir)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:53:48.395755Z","iopub.execute_input":"2025-11-28T13:53:48.396537Z","iopub.status.idle":"2025-11-28T13:58:12.246005Z","shell.execute_reply.started":"2025-11-28T13:53:48.396511Z","shell.execute_reply":"2025-11-28T13:58:12.245135Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"imgs_dir_resized = Path(\"/kaggle/working/imgs_resized\")\nimg_paths_resized = list(imgs_dir_resized.rglob(\"*.jpg\"))\n\nwidths_resized = []\nheights_resized = []\n\nfor p in img_paths_resized:\n    with Image.open(p) as img:\n        w, h = img.size\n        widths_resized.append(w)\n        heights_resized.append(h)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:58:52.972497Z","iopub.execute_input":"2025-11-28T13:58:52.972798Z","iopub.status.idle":"2025-11-28T13:58:53.756365Z","shell.execute_reply.started":"2025-11-28T13:58:52.972772Z","shell.execute_reply":"2025-11-28T13:58:53.755580Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"plt.figure(figsize=(12, 5))\n\n# WIDTH HISTOGRAM\nplt.subplot(1, 2, 1)\nplt.hist(widths_resized, bins=30, edgecolor='black', alpha=0.8)\nplt.title(\"Image Width Distribution\", fontsize=14)\nplt.xlabel(\"Width (pixels)\", fontsize=12)\nplt.ylabel(\"Number of Images\", fontsize=12)\nplt.grid(axis='y', linestyle='--', alpha=0.4)\n\n# stats on width\nw_mean = int(np.mean(widths_resized))\nw_median = int(np.median(widths_resized))\nw_min = min(widths_resized)\nw_max = max(widths_resized)\n\nplt.text(0.98, 0.95,\n         f\"Mean: {w_mean}\\nMedian: {w_median}\\nMin: {w_min}\\nMax: {w_max}\",\n         ha='right', va='top', transform=plt.gca().transAxes,\n         fontsize=10, bbox=dict(facecolor='white', alpha=0.7))\n\n# HEIGHT HISTOGRAM\nplt.subplot(1, 2, 2)\nplt.hist(heights_resized, bins=30, edgecolor='black', alpha=0.8)\nplt.title(\"Image Height Distribution\", fontsize=14)\nplt.xlabel(\"Height (pixels)\", fontsize=12)\nplt.ylabel(\"Number of Images\", fontsize=12)\nplt.grid(axis='y', linestyle='--', alpha=0.4)\n\n# stats on height\nh_mean = int(np.mean(heights_resized))\nh_median = int(np.median(heights_resized))\nh_min = min(heights_resized)\nh_max = max(heights_resized)\n\nplt.text(0.98, 0.95,\n         f\"Mean: {h_mean}\\nMedian: {h_median}\\nMin: {h_min}\\nMax: {h_max}\",\n         ha='right', va='top', transform=plt.gca().transAxes,\n         fontsize=10, bbox=dict(facecolor='white', alpha=0.7))\n\nplt.tight_layout()\nplt.show()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:58:55.503158Z","iopub.execute_input":"2025-11-28T13:58:55.503442Z","iopub.status.idle":"2025-11-28T13:58:55.941611Z","shell.execute_reply.started":"2025-11-28T13:58:55.503421Z","shell.execute_reply":"2025-11-28T13:58:55.940865Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"plt.figure(figsize=(14, 5))\n\n# first histogram (widths)\nplt.hist(widths_resized, bins=30, alpha=0.6, edgecolor='black', label='Widths')\n\n# create a second axis sharing x or y\nax2 = plt.gca().twinx()\n\n# second histogram (heights)\nax2.hist(heights_resized, bins=30, alpha=0.6, edgecolor='black', color='orange', label='Heights')\n\nplt.title(\"Combined Histogram of Image Widths and Heights\", fontsize=14)\nplt.xlabel(\"Pixel Value (Width / Height)\", fontsize=12)\nplt.grid(alpha=0.3, linestyle='--')\n\n# legends (from both axes)\nplt.legend(loc=\"upper left\")\nax2.legend(loc=\"upper right\")\n\nplt.show()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:59:00.189575Z","iopub.execute_input":"2025-11-28T13:59:00.190327Z","iopub.status.idle":"2025-11-28T13:59:00.526556Z","shell.execute_reply.started":"2025-11-28T13:59:00.190301Z","shell.execute_reply":"2025-11-28T13:59:00.525782Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"from fastai.vision.all import * ","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:59:03.126709Z","iopub.execute_input":"2025-11-28T13:59:03.127010Z","iopub.status.idle":"2025-11-28T13:59:17.360372Z","shell.execute_reply.started":"2025-11-28T13:59:03.126988Z","shell.execute_reply":"2025-11-28T13:59:17.359473Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"import torch\nimport torch.nn.functional as F\nfrom fastai.metrics import Metric\n\nclass MultiLogLoss(Metric):\n    \"Multiclass logarithmic loss (average cross-entropy). Handles logits or probabilities.\"\n    def __init__(self, eps=1e-7):\n        self.eps = eps\n        self._name = 'log_loss'\n        self.reset()\n\n    def reset(self):\n        self.total_loss = 0.0\n        self.count = 0\n\n    def accumulate(self, learn):\n        \"Called for each batch: uses learn.pred and learn.y\"\n        preds = learn.pred\n        targs = learn.y\n\n        # ensure targs is (bs,) of indices\n        if targs.ndim > 1:\n            targs = targs.view(-1)\n\n        # If preds look like probabilities (rows sum to ~1), use -log(p)\n        if preds.ndim > 1 and torch.allclose(preds.sum(dim=1), torch.ones(preds.size(0)).to(preds.device), atol=1e-3):\n            probs = preds.clamp(self.eps, 1.0 - self.eps)\n            batch_loss = -torch.log(probs[range(probs.size(0)), targs]).sum()\n        else:\n            # treat preds as logits -> use numerically-stable cross_entropy\n            batch_loss = F.cross_entropy(preds, targs, reduction='sum')\n\n        # accumulate on CPU\n        self.total_loss += batch_loss.detach().cpu().item()\n        self.count += targs.size(0)\n\n    @property\n    def value(self):\n        if self.count == 0: \n            return None\n        return self.total_loss / self.count","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:59:17.361722Z","iopub.execute_input":"2025-11-28T13:59:17.362019Z","iopub.status.idle":"2025-11-28T13:59:17.372107Z","shell.execute_reply.started":"2025-11-28T13:59:17.361992Z","shell.execute_reply":"2025-11-28T13:59:17.368488Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"import numpy as np\nimport matplotlib.pyplot as plt\n\ndef plot_multilogloss(learner):\n    \"\"\"\n    Plots validation MultiLogLoss per epoch from a FastAI learner \n    that was trained using MultiLogLoss() as a metric.\n    \"\"\"\n    # metric names corresponding to recorder.values columns\n    mnames = learner.recorder.metric_names[1:-1]\n    vals   = np.array(learner.recorder.values)\n\n    idx = mnames.index('multi_log_loss')\n    val_logloss = vals[:, idx]\n\n    epochs = np.arange(1, len(val_logloss) + 1)\n\n    plt.plot(epochs, val_logloss, marker='o')\n    plt.xlabel('epoch')\n    plt.ylabel('validation log_loss')\n    plt.title('Validation MultiLogLoss per epoch')\n    plt.grid(True)\n    plt.show()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:59:17.372933Z","iopub.execute_input":"2025-11-28T13:59:17.373194Z","iopub.status.idle":"2025-11-28T13:59:17.393895Z","shell.execute_reply.started":"2025-11-28T13:59:17.373174Z","shell.execute_reply":"2025-11-28T13:59:17.393088Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"items = df['Image']\nitem = items[0]\nitem","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:59:17.395603Z","iopub.execute_input":"2025-11-28T13:59:17.395796Z","iopub.status.idle":"2025-11-28T13:59:17.417502Z","shell.execute_reply.started":"2025-11-28T13:59:17.395781Z","shell.execute_reply":"2025-11-28T13:59:17.416734Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"image_file = f'imgs_resized/{item}'\nimage = PILImage.create(image_file)\nimage","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:59:17.418268Z","iopub.execute_input":"2025-11-28T13:59:17.418756Z","iopub.status.idle":"2025-11-28T13:59:17.449552Z","shell.execute_reply.started":"2025-11-28T13:59:17.418732Z","shell.execute_reply":"2025-11-28T13:59:17.448835Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"image2whaleID = {o.Image: o.whaleID for o in df.itertuples()}\nlabel = image2whaleID[item]\nlabel","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:59:17.450315Z","iopub.execute_input":"2025-11-28T13:59:17.450574Z","iopub.status.idle":"2025-11-28T13:59:17.458714Z","shell.execute_reply.started":"2025-11-28T13:59:17.450553Z","shell.execute_reply":"2025-11-28T13:59:17.457840Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"def get_image_file(item):\n    return f'imgs_resized/{item}'\n\ndef get_label(item):\n    return image2whaleID[item]","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:59:17.459482Z","iopub.execute_input":"2025-11-28T13:59:17.459737Z","iopub.status.idle":"2025-11-28T13:59:17.472556Z","shell.execute_reply.started":"2025-11-28T13:59:17.459711Z","shell.execute_reply":"2025-11-28T13:59:17.471818Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"x_pipe = [get_image_file, PILImage.create]\ny_pipe = [get_label, Categorize()]","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:59:17.473178Z","iopub.execute_input":"2025-11-28T13:59:17.473394Z","iopub.status.idle":"2025-11-28T13:59:17.488341Z","shell.execute_reply.started":"2025-11-28T13:59:17.473379Z","shell.execute_reply":"2025-11-28T13:59:17.487668Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# make sure at least one of each whale is in training set, then randomly split\nmust_train_whales = df.groupby('whaleID').first()['Image']\nmust_train_ids = pd.Series(df['Image'].index, index=df['Image']).get(must_train_whales)\nmust_train_ids = set(must_train_ids)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:59:17.489048Z","iopub.execute_input":"2025-11-28T13:59:17.489272Z","iopub.status.idle":"2025-11-28T13:59:17.516763Z","shell.execute_reply.started":"2025-11-28T13:59:17.489252Z","shell.execute_reply":"2025-11-28T13:59:17.515987Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"train_ids, valid_ids = RandomSplitter(seed=42)(items)\nprint(f\"Before: train_ids={len(train_ids)}, valid_ids={len(valid_ids)}\")\ntrain_ids = L(set(train_ids).union(must_train_ids))\nvalid_ids = L(set(valid_ids) - must_train_ids)\nprint(f\"After: train_ids={len(train_ids)}, valid_ids={len(valid_ids)}\")\nsplits = (train_ids, valid_ids)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:59:17.518903Z","iopub.execute_input":"2025-11-28T13:59:17.519257Z","iopub.status.idle":"2025-11-28T13:59:17.533596Z","shell.execute_reply.started":"2025-11-28T13:59:17.519235Z","shell.execute_reply":"2025-11-28T13:59:17.532865Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"dss = Datasets(items, [x_pipe, y_pipe], splits=splits)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:59:17.534371Z","iopub.execute_input":"2025-11-28T13:59:17.534820Z","iopub.status.idle":"2025-11-28T13:59:17.599758Z","shell.execute_reply.started":"2025-11-28T13:59:17.534795Z","shell.execute_reply":"2025-11-28T13:59:17.599080Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"dss.show(dss[0])","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:59:17.600529Z","iopub.execute_input":"2025-11-28T13:59:17.600728Z","iopub.status.idle":"2025-11-28T13:59:17.715211Z","shell.execute_reply.started":"2025-11-28T13:59:17.600714Z","shell.execute_reply":"2025-11-28T13:59:17.714524Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"after_item = [ToTensor(), Resize((320, 480))]\nafter_batch = [IntToFloatTensor(), *aug_transforms(size=(224, 336))]","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:59:17.716087Z","iopub.execute_input":"2025-11-28T13:59:17.716374Z","iopub.status.idle":"2025-11-28T13:59:17.723294Z","shell.execute_reply.started":"2025-11-28T13:59:17.716353Z","shell.execute_reply":"2025-11-28T13:59:17.722601Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"dls = dss.dataloaders(32, after_item=after_item, after_batch=after_batch)\ndls.show_batch()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T13:59:17.724185Z","iopub.execute_input":"2025-11-28T13:59:17.724489Z","iopub.status.idle":"2025-11-28T13:59:19.610283Z","shell.execute_reply.started":"2025-11-28T13:59:17.724468Z","shell.execute_reply":"2025-11-28T13:59:19.609514Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"metrics = [error_rate, MultiLogLoss()]\nlearn_resnet26d_w = vision_learner(dls, 'resnet26d', metrics=metrics).to_fp16()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T14:00:55.126910Z","iopub.execute_input":"2025-11-28T14:00:55.127766Z","iopub.status.idle":"2025-11-28T14:00:57.314282Z","shell.execute_reply.started":"2025-11-28T14:00:55.127739Z","shell.execute_reply":"2025-11-28T14:00:57.313499Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"learn_resnet26d_w.lr_find(num_it=100)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T14:00:57.315530Z","iopub.execute_input":"2025-11-28T14:00:57.315858Z","iopub.status.idle":"2025-11-28T14:01:13.839723Z","shell.execute_reply.started":"2025-11-28T14:00:57.315835Z","shell.execute_reply":"2025-11-28T14:01:13.838758Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"learn_resnet26d_w.fine_tune(10, 0.009120108559727669)\nlearn_resnet26d_w.recorder.plot_loss()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T14:01:32.578874Z","iopub.execute_input":"2025-11-28T14:01:32.579199Z","iopub.status.idle":"2025-11-28T14:06:25.669003Z","shell.execute_reply.started":"2025-11-28T14:01:32.579173Z","shell.execute_reply":"2025-11-28T14:06:25.668116Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"plot_multilogloss(learn_resnet26d_w)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T14:06:25.670690Z","iopub.execute_input":"2025-11-28T14:06:25.670921Z","iopub.status.idle":"2025-11-28T14:06:25.834272Z","shell.execute_reply.started":"2025-11-28T14:06:25.670899Z","shell.execute_reply":"2025-11-28T14:06:25.833439Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"def submit(learn, path):\n    test_df = pd.read_csv('/kaggle/input/noaa-right-whale-recognition/sample_submission.csv')\n    test_dl = learn.dls.test_dl(test_df['Image'])\n    \n    preds, targs = learn.get_preds(dl=test_dl)\n    \n    df = pd.DataFrame(preds.numpy(), columns=learn.dls.vocab)\n    df[\"Image\"] = test_df['Image']\n\n    preds_path = path\n    df.to_csv(preds_path, index=False)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T14:06:25.835090Z","iopub.execute_input":"2025-11-28T14:06:25.835530Z","iopub.status.idle":"2025-11-28T14:06:25.840920Z","shell.execute_reply.started":"2025-11-28T14:06:25.835505Z","shell.execute_reply":"2025-11-28T14:06:25.840158Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"submit(learn_resnet26d_w, \"submission_resnet26_w.csv\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T14:06:25.842467Z","iopub.execute_input":"2025-11-28T14:06:25.842673Z","iopub.status.idle":"2025-11-28T14:06:43.813623Z","shell.execute_reply.started":"2025-11-28T14:06:25.842659Z","shell.execute_reply":"2025-11-28T14:06:43.812463Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"### Result (First score - Private Score, Second score - Public Score)\n![image.png](attachment:809843f7-4797-46b6-82f7-302609e69a38.png)","metadata":{},"attachments":{"165f9956-0d53-4b91-a650-6288c2b6912c.png":{"image/png":"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"},"809843f7-4797-46b6-82f7-302609e69a38.png":{"image/png":"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"}}},{"cell_type":"code","source":"metrics = [error_rate, MultiLogLoss()]\nlearn_resnet50d = vision_learner(dls, 'resnet50d', metrics=metrics).to_fp16()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T14:08:38.514003Z","iopub.execute_input":"2025-11-28T14:08:38.514783Z","iopub.status.idle":"2025-11-28T14:08:40.787362Z","shell.execute_reply.started":"2025-11-28T14:08:38.514751Z","shell.execute_reply":"2025-11-28T14:08:40.786573Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"learn_resnet50d.lr_find(num_it=100)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T14:08:46.842344Z","iopub.execute_input":"2025-11-28T14:08:46.842632Z","iopub.status.idle":"2025-11-28T14:09:07.424280Z","shell.execute_reply.started":"2025-11-28T14:08:46.842610Z","shell.execute_reply":"2025-11-28T14:09:07.423411Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"learn_resnet50d.fine_tune(10, 0.002511886414140463)\nlearn_resnet50d.recorder.plot_loss()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T14:10:29.148463Z","iopub.execute_input":"2025-11-28T14:10:29.148789Z","iopub.status.idle":"2025-11-28T14:17:06.536777Z","shell.execute_reply.started":"2025-11-28T14:10:29.148760Z","shell.execute_reply":"2025-11-28T14:17:06.535860Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"plot_multilogloss(learn_resnet50d)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T14:17:06.539056Z","iopub.execute_input":"2025-11-28T14:17:06.539310Z","iopub.status.idle":"2025-11-28T14:17:06.691862Z","shell.execute_reply.started":"2025-11-28T14:17:06.539287Z","shell.execute_reply":"2025-11-28T14:17:06.691124Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"submit(learn_resnet50d, \"submission_resnet50.csv\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T14:17:06.692727Z","iopub.execute_input":"2025-11-28T14:17:06.693190Z","iopub.status.idle":"2025-11-28T14:17:29.034445Z","shell.execute_reply.started":"2025-11-28T14:17:06.693170Z","shell.execute_reply":"2025-11-28T14:17:29.033540Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"### Result (First score - Private Score, Second score - Public Score)\n![image.png](attachment:d05bef0e-9d9f-40c2-b64f-7c9f6ec88b2c.png)","metadata":{},"attachments":{"a02346be-94b1-47d6-ae36-c83e610706e1.png":{"image/png":"iVBORw0KGgoAAAANSUhEUgAABdMAAAB2CAYAAAAjgQrGAAAAAXNSR0IArs4c6QAAAARnQU1BAACxjwv8YQUAAAAJcEhZcwAADsMAAA7DAcdvqGQAAD7OSURBVHhe7d0LXFRl3gfwH4oDggPooCRggqjgDa94A0soDazUNXVLxU3NzWVbpdZ0zUu7QVtoW+Lborut2iZ2UXPFWiHbBUtJDUxF0yFFyERXZbwwijCivufMeWYYhssMFw31930/5/U5z7k958w5rf7Pc/6Pg28Hv1sgIiIiIiIiIiIiIqIaNRN/EhERERERERERERFRDRhMJyIiIiIiIiIiIiKygcF0IiIiIiIiIiIiIiIbGEwnIiIiIiIiIiIiIrKBwXQiIiIiIiIiIiIiIhsYTCciIiIiIiIiIiIisoHBdCIiIiIiIiIiIiIiGxx8O/jdEmWb/AICRImIiIiIiIiIiIiI6O5XkJcnSrWrUzCdiIiIiIiIiIiIiOh+xDQvREREREREREREREQ2MJhORERERERERERERGQDg+lERERERERERERERDYwmE5EREREREREREREZAOD6URERERERERERERENjCYTkRERERERERERERkA4PpREREREREREREREQ2MJhORERERERERERERGQDg+lERERERERERERERDYwmE5EREREREREREREZAOD6URERERERERERERENjCYTkRERERERERERERkA4PpREREREREREREREQ2MJhORERERERERERERGQDg+lERERERERERERERDYwmE5EREREREREREREZAOD6URERERERERERERENjCYTkRERERERERERERkA4PpREREREREREREREQ2MJhORERERERERERERGQDg+lERERERERERERERDYwmE5EREREREREREREZAOD6URERERERERERERENjCYTkRERERERERERERkg4NvB79bokxEdO/rMxExI/yhRhEObl6NtDxRbzd/RM6ciN4eUrEoC/9Ymw6dsoCIiIiIiIiIiO5h7JlOd5ZXEEJCQxDkJeZvK38Eh4Yh2E/MEiEI81+Zj+lPT8SEcH8Y6hxIl+Ujq3kQxsj7eOEPmD9KVBMRERERERER0T2NPdPvR+MXY80TnaRCMbLfm4OkTKX69ovC0u3xiNBIxTOpmPfEIqQrC26DIMSsW4vp3VWAIQdJY6dhzVmxiH5GE7H4/Sj4i7nq6LNXY867u8RcBb/IWMSMDUFQFx+oy4qQf/wgUtcnYePeOvQLD4/HF29FQb4FCzZNxvg3tEq9oBk0ETGTo9C7sz80LgboCo7iYFoykj7Oqtz73CsGa7bMQLB8ex1ajSefTWLvdCIiIiIiIiKiexx7pt+PfIIQ3CtYmrrBz13U3RHe8FSLosZTmrudpP23VilFlQY+nZUi3QHyy5r31xqnxeNFnYmXNwKN917NU5Cf6SYx0WBMwnZsej0aESFB8PZQQ+0lf3UwFvOTPkfygghjcNweE34RJtbNR1aKZSBdg4gF6/FZ0nyMCQ2Gn5d0DLUGfr3CMOblVfjs03iMsfya4mwS0vcbjEVVr1BE35EvLYiIiIiIiIiI6OfEYDrdQavx2p/WIT0zHWsWL0SyqL09duGduASkZO5CyhuLEHfHet9TxcuaYAT6iDqT/j5oaywYoMvLQc6hqpO2QG9cwyR4zgrMf1SEy0sKkZW2ASkZWujL5QoVgsYvQfxke8LpUQgJEoH6giykHFGKRqP+gPnjg6S9ScQxNn6ciqwzImDuF4XYhRONZZPk/VrpLGSdEDzaWCAiIiIiIiIionvYPZTmpRkc2jeHg5hT3MStMzfAPDZW5qxH9tQgqaBD+uKRmLdNqSZqFOb7C9B+0B9TEo1FxeRV+OalEKjsvvcmInHHfITKMXCDFsm/mYzlB5Ql6DMfm1ZPhDEl/tlUzBtlI21QaDy2rogyfhGh+88iPDY/VamXTF+9GzF95FB6PjbOGI8E0zEQjPmfrsUE+SByuqAh07BGWQB4xSJ5WzTkMzXsXY6hMeuUeiIiIiIiIiIiuifdxT3Tm6HZAE+4zA5E61UD0O6DELRL6Ie2lSa5fhDareoJj9k+cBrQwirY3rT5Rc7Hmm27kb1vn3n6Zvt6xE8LqZTWInadWL59GSJFnUltyxTBmPLWJmTsFevt3YGtK2IQUilthZzrXFn+TVI0/MbHY9MOsf6+3dgUN9bYHs2jsViz3aK9/12LWFOPYqOK/WSvixV1CnvPVdYY1yVYTkVSaR+7kfH5Kqv2SuTAsHH5diwdZc+1slflaxE8axW+2K3MJ88Rq0iCp8Yj2bKde3fjiw/jMaWPWMFCva5Ln2gs/XRHxTY7tiLxharXXGa7LeKcRCBdFjRVWdd8Tp5qpfc3zuO0PS9xZkYhRHQm16Wvrgikyw4kYMtepW84vIIREaoUa9TXX6QWMqDweEUgHQhB25Zl0Ov10B/JQorlMZCDg/kiG7pKBTelpDibi9OmRQ8GwtbhLdl9/0ns+V0jXt9u3s/KyaLSQsXyHUi0Tr1DRERERERERER2ufuC6S1aoMWkALRZHYK2swOgHuABlYt1j/TKHFxc4TTAFx6z+6Ht6p5wm+QOhxZiYROlmbkKH74+EcFeSujRoDcYU0qoNEGIfGEFEucEG+sbRgXPifGIDfeH2lFUOarhHToDiW/ForojqLzH4K0FUahIa62C36hYLJk5H39LiEawRuQpl3kEY0rcCsR0F/M1qMu5NsZ1CZ6zHqsWjDXvQ6GCun0IpiR8hKWjqwslA97PvFXttVr6anS1wWe7tY5A/MwQWF46mbGdc6IQZNlORxU0gVGIXbkesRYB9fpdF29MWRqLCMsc5WofhE5bViVtSl3aUptIbyXJCwwGFPtFYfrriUp+9RXxmD46qMp1jOzsI4Lvemj3Vu13nnzkhCj5wG+AKNYg0k8cWw7kW6Z4QRYSJg1H+HBpik6A1ZCk6GRK9C+3WSkJqSgwDWrb3h+9RdGWutx/9v6u6du+xWljSYXeodHGUoUIRPQV+9QfxK5NSpGIiIiIiIiIiOrmLgqmN0Pzx/3R5u/90CbSU46p14tDC1e0jAxCu78Hw+1xpybaUz0ML46RU2FI9DlImtEfQ4cPwdBRLyPtjFypQqfwMYiQiw2iRnCvttBujcOUUf0x/sUkpP8kh+qkI3SPxpIFFT2MzTr4Q703CfMmTcb017egwLi6GqGzJsL7p1QsnzEe42fEISVP2Q9UQRgaWVuouS7n2gjXJTwe8VNFbuxLOUh+cTwG9B+JKctScdrYZHkgyrcwvUpvcw2CugNZqxZhzuw5WLQqC6JTMtSDIhs2AGV7H3hLbUl5W973Iqz8SqqzaKchbwsWPdVfaqf0Gy0U11y6rlNemW9MMVLv66IJkrbPQtLCOcbjJu01nxFCIi1eENjdllTMGykt+6AiHC2neZHXNaV6cVIZWynxR/RH8YiJDFPyq4dGIebV9di0OqbSS5xO3qZWFCJ/qyhaOnHa/Du07RAlStXr5GXalx5FdubQ14z+A8b0Utqsz0qtSPEinDhrOroT1DZeGhnVdP/9yfQsSfffC38QX1LU4XfNTEZmnlyQanuFYopSVIRHmL+e0GWmYKNSJCIiIiIiIiKiOro7gunOrmgZ1weaX7ZDi+aiztKNG7hx4TJKD5/DtUrTBZRduI5bN8R6lpq3RMtfSvuMa4/mzqKuyVDLGSUUJYUoMKWdOJuORS9MNga+h46Nqz0/tJ30e5Mw509boD0LFHy9GvP+uAUFYpnfgDEiWGtBl453YlYjPVeLnM1xeOc/hWKBHllrFiH5QL7U3i2IS/yv6CkLtPUeKErVqcu5Nvy6RD4+UKT60CNz5TQs/zpfKuug/XgR5n0sBpRUBSPi6aovAE5vnYvfvJeKzMxdSHtvFhLSzGFcePcXxfow5GDNpGmIWy/vW5qk8zK3U1qW/Ls4pIkfpSAtDou2ym2WBIRiijG3SH2vSyFS5s3CmrRdxuOuiXkTaaae1l4+CBHFurWldv6mXt4qNZz0+ciUB/pMy8LpEqVa3WcGliVUBMVNp9UYzPu6VATR6tr1iUXiggjlpYJ03y9/fYOxunpu0BiTt9euxvtvq3QtN2uhP5uPrBzpmQqQ16nL76pFyn5xVi69EWqR6iXi0YHixUghsrY1xn81iIiIiIiIiIjuT00/mO7VBm7Le8Kto3W+85u4cep/uLziAM79OhtFsVpcXpqP4krTMVyK/U5anoXzK/JRcqqsymCkzTs+CM3yADg1pGdxo0tFzgljWFc6/ygs3bEVyauWYfHMsQj1KDMGvhuHAfn715l79hodSEGWKZrezh9VMmecLUSaKMoyi/SiZEBZuSjKMousUmLUpC7n2tDrokGQqadziRZZVukutB9nwZQ0pFPgKFEy0UvXKkeUFennzotSA+UfxCeV2m7ZTmlu6nzMf7liGtOqTAn6y6lN+sp/1vO6XMpHjilAa5SOoko3g6yuband8uiRmBIzB4vmzsKTI8djzsIEJCychdHj45Apjq15aCJibufzeEO6V0WxRn1ikbwyGkFyMFse+HTey0hp8HNn4/5bNhnho8bjN3OXI83Yy7xuv6t2TTpyjKtbpnqxSPGSl4lkO3vkExERERERERFRVU07mN6mDdziuqCli5gXbvx4Chdf2YeiV35EaXYZbl0XC2py/SZuZp+D/pUDOP+KFvofr1cKqju4eMIjLgBObURFE5D8agLSCkSgWu2DoJAIjJm1GImrN+Gb7dYDe9aXHkWmjuVmWpSJXsJwUcNTFG+nupxrw67LQHibTuiadO6iaGYRnFSprfdj9bKgMd2Q+yZbsminRzDGPD0REyynSJEmRCa+1KjXdbEnqFyPttROB+3eXUjLqEiTY3R2C1ZmiJ7VKh/4iZ7++eaXNU5wqi6NiqPKfHy9rvb+5iIsDWi8q35xYalSIL0QaW/MrjzwabWKoTO9hKqRjfuvGnX6Xc9+guxc5SxVfSMwXS5YpHgp2J9ilQ+eiIiIiIiIiIjqoukG09UecPtzF7S0TMFy4zpKtxyAbnEhDKduisq6uXXqMkoWS/vYckHODlPB2RPuf/aXs080DWfl3NTD8dikl7F8bSoyD+VDJ6KBKk0wpixaggnKbAOooW4timZBcDK9vCixL+DXYHU51wZdF2ldU2y2uQpOolgdg8G04s/Bop1nUhE3W85pXv20NEWsd9vul3q0pZ60JVVD+7nnTHegDwKrSSMTGuIv3cUyA86frD1UXJHfXA3PGlLSaAZZB9KnYdHWKt31zYJMedil++V8pUFNq2P//WdWp99Vh6QvDyovDVRBCJlskeLFkIP0NQylExERERERERE1RBMNpjeH0287V+6RfuMqrq48iMubq6ZqqbubuLH5GHQrz+C6RUDdwaUd3H/bpokMSupvHJQxqGU6kt9dhDnPjsdjQ8YjKUtE0tTdEGKdiUTucVspPUYUvGtNl6FC0KCJoix0H4MQU+7ny3bmlm6wupxrPa6LmRaFZ8V6HtJ64UrRRDOrNzqJ8unjP2c+DIt2anzgc1zOaW4xHS+CrlAp55h7QzfkutSmPm2pyQys2b0P2fukaV2sEuS1MCHAR5RKYbislLRfa0XufelefSjWqkd5BKL6i20MWmStV4o1ScstrDUljRxIT1wuAunlOqTbCKQDY+GvJECHdBGQK4o1s3H/zUzEpg9XYenCGYgwP7d1/F3XpyPLGLCXU73EIqK3cpUNR3YhqcFpaoiIiIiIiIjoTlCpnPCAtzfa+/g0aJL3Ie+LGk+TDKY7RHaCe3eLnBHGQPpRXPm2upFE6+/Wtydx0Sqg3qx7J7hF2pWv4vYJmI/kfZuwZkU8lv0+FsGiWuZs7qlfkVYiq8CUqyUIYxJiESoHw/3CMD3pRURYRyytqEOfQ+J4f2XGKwSxC8bCFEs/nZWG2x5Srsu51vG6VCf5810ivYgGEb9fhgniZDWDZiB+XLBIGZKPrJSftxdvcvq3UGKiwRjz6gxzqg5IZz3lTyuQ/Ok+fLNzPebLo4Q2wnWpTZ3aIssvUtaXuHlYtmaLOQ0Juv8Sf3s1Stxr/gidsxYxoeKzEMvc3pnJSD+ibKOStln6+kTlhZF0f8euXoJI0RZd5gasUYo1yzQF5oFOfWeIkkLz6Hz8zRRIl1qftfZNpOi6ITQ0rNIUbHo4ZKED4O+hFHXazMrPSvcZWLl9N7J3b8fKWRXXoNb7b0IY/AJDEDEiGBo58F2v33UDUrOVI6j8IhDcXi5J55O62lhHRERERERERE1br+A+mDJtBp4Y8ws8PnpsgyZ5H1OmTUe3nr3E3qmhml4wvYU7Wo2r3Du8LC230QPpJnJA/VLaFYve7s3hPM4XLVqI2Z9D3j+QlqmEI1Xdo7Fm925k7NiBjN2bML2XEu41HEjFP0RaicxPM2GKp6l7RSPx033I/jQRMYNKkXOotp61emhzgdAFm5AtHSN72ypM6a7sH7p0rF61SynfTnU51zpel2ptexNJGSLY2D4C8z/djW927MYXSTEIUfJhQPvBa0iwmbLjNtv0GpaLdmoGxWDlVuVcv9m9FrFKQ6HfuxH/yJIKjXFdalOXtsj25iJfxMy9R6819kTftEDuU65D0vJPoDUuU8FvdDw2yb3U921C4tRgJV2LPNjnnxMscntrsfwvFdt4R85H8jbl/p7SRwTfpXs1aVmqUq7NkRRkGQf2lPbUNQRTlKJR9K8mwk/c+tJThJCZy5C4IrHKNG+MWEUSMWoglI7pOhxMtzr+iAjpfpJ2qNIgJML00kBiz/33r9XYKBfr+bum/ftbJWDv5aO0T38Qu6wGOyUiIiIiIiKipsfVtRUGhYaiWbPGy5vRrFkzDA0bBudKubSpvppcML3ZBF+4WPy2t/LzcfkTWyOMNszNT7Qotsxn4vwAWk34OaPpOiTPno3lGfnQy4NeqlRQq9VQy/Ezgw7atOWYNWO16OEqOZCAuW+kwjROoVFJPtITF2FDrUnPDTj98ZvYmCttKB3DRF+QjuXzXkbKHUkLUZdzreN1qZYOKXOfx6KPc0TuaRVUxh1I9NI1e3c25iTmKPM/K6t2OirnavyZpHPN+XgRnp+7pRGvS23q0hbJ2SQsX7kLp43XV+HdWSQpP7AcU36zHGm5OpFypYJ83yXFVjPYZ43bGKA7tA7zou29V7X4R3qOsg+X3oiYpbwIqJ8IRPQV25/9FmnblKJZjhanxYC1umM55pdddbv/6vm7ZiQj3aK3ui4zRQnOExERERHJRs3HyvfXYk0tU+ILYWLl2vhjwqurKrZbEYMqQxP5hWHKq4lY8+l2Y8eQLz6V9v36DERafvFpzSsEE15ehpWW27warXyBXRN5G4vjZHy+HivfisWYwOr+zh+E6QmVz7fqtLgRxiijJq37DCyt9re3mF61SotbA7/xiy2eqUTEVHkQ7HhWoEHI04uR+P4mfCHfw9s2SffwfEwYVMu/W/2iMP11q+fL1rNiySsa8eZ219QuutPU7m6iBOzJ3IV/b93SoEneh8zBwQEerdsYy9QwDr4d/BqegrzRuML1rz3RyjwI6DXo43NQ8oOYbQwtnOA8wRVlGy/glmWMvqsP2izyhTmEri9E0W9P4fb0h68LDYIGdYPGUSqWF0G7V1s1eGZBExiCoJZFyDxQt2znxu08naTTtif39e1Sl3Ot23WpnsU+rhbW+ZpZi3xrO+LDbQRodelYNPJlpIlZ+9zp61Kbuu3fr08YfFDTtZXzgftAjTLojmdBa09A3CsIIZ09oUIxCjMtg9T2isLS7fFK+qO8DZgy0bIXfB2MT0TGgjCp7QbkvPcEpq+q5ir4BSPUQ1/LfVWX++92/65EREREdN+Ysx7ZUyuPRmRNl7EIj82t+etPOU3hkldjEGpOASmx+reOZnQ81i6IgrfoO1JJuZxacTZ+s6pyRybjOEZvRSPIcvw0E3lco9efwTyrcY3klI2Jr06sfhvopb+vz5b+vm55HIt/E9RIi+T+k7FczNE9aNQyfBEXIf1LqxZH1mFAdC13gVcIpi98HTGhlnuR7tPFIzFPdLiy51mBVwTmv/M6JgRW97AYoF0/G3Pezqr0b8DgWWuROFN84W2thmelMg2mr/4cMX0sjlmveAU1NjnPuZyeRSYHw88UmlI7109j74+aWs/0UE+4WvyX4Nbx07jW2IH0Bb3gHtkFnr9vAwfLzuc/nMHV46IsUz8AlybxSk4H7V4x2KMdATRdbla9gsLG7ewaRPJ2qsu51u26VM9iHw0MpN9ed/q61KZu+y84IK1X47XNR468n0w7A+mys1pkGbepTyBdloqEjVnQ6/XQtwvGGFOe9zrxx/SIbtLfy6V9FGxDcnWBdFlBjo37qi733+3+XYmIiIjofhHqWW34zW7yeEeb5B6slsFBa91jkGgZSNcXQntIi9Omr6kd5dSK8Yi3HJRf7iWbYBFILzcY/95uMH2e6qhBxIIViO0j5mVeM7AsziKQbtCh4JD0bwXlM1CJGsEz38LS0ZbBTm808BLQvcDHs/pAtL0eisWaD1dZBdIrs+tZQRBi3qocSDeUWNz3UCFo8jLET7Y4zqhlWGYRSDdckp+vfPH1s0Q8KzHdxXw1NJNfR7RlIJ2I7NakguktBra1yJV+A6U7iixymTeQKZDeWRlctFl364D6TZTtOFcpd3rLge6iTGRb2tyRGNC/f+0T3/L+7HTvzUL48OHSNBkJpjzvdZKPNTEjlX08FYd0UUtEREREdDdQm3Pmyj1oq/k3izTV1is9ZEAw1PIXk3Jq0a2Ve8uaBI2JEIP7A4bcdZg+fDSmPDsZo4dPQ/IRU8TPByGPR4kyEDE7GiGm6KBuFxJ+OcT4d+6hY2dhY67YRhWEibEx5t7EoS+MQbApHqhLR9zYkRj/7DSMH/kE4sRYRXIP3Igpz6GiL75IWSmTex5Xc/4D2Cv93ueiguk20H5Q3T0gTbX1Su8bgmAPuWBAQcYWZFXzINjzrGD8bzDBNHadIR8pC8dj6DDpvh8yHglfm7ZQI2RMxT08/akw8zOgy4jDk4/Iz9d4PDZkGjaaep1Jz0rEmBq+QPEaiyW/CmnYywSi+1gTCqY7wbGDEug2unERZbtFuaGsAukmN85eq5zqZfdllFrkdXHo4ILKWxARERERERHdvTp5mcJw53F6nyjWkSF3CxZFj8e8LMuBuyoM6OAjSsCJvctRkWQlB8u/1irjGEk0XoGiFIVIZUR+iQE5m1+rCAqezULCe/81ByJVvUIRLXr6hviZjmNA1j8tx1KSxypah6wSMRsQiimmL88jfdBWFPVFP+un2fQzivQ23wXQ1fdD9RItUhZOwvi52dJeqmfrWZkeNdAc1C7YughxaabG5GPjm28i6eMN2ChPe06LALo/Tu/ZotR9vBpJayzGL5Oer4P5FXNqjb8oVRb5UgxMHeoLNm2pX+pTovtYEwqmu6KFpyjKzl3F9cYYd7SGQPr1jBxcWHtNzAnX9bh+TpRlni6QXyISERERERER3QtMvXGNWooBQuXBB1ctw/zJYbA1dqEctH5yUhzSaolDJ8cMMffunZIoKoUgT7W5DeZgdvdgdDLF0nEC2dapFDOyob0kyvCB/yBRNNNDf1EUzYqgN/+T3wd+fUXRIgpiuCG1Z3Qslq5SBmCUB0etftBSuueYQ0QGlCEIY+YsE4OIJiJ+5lgE1ZqaRbJ/NeaNn2wR/K7K9rMShaAOpqchH1kpWmgCx2L6QqUtibODYNi7AQnLEpDw9jpkivXS3pPm5bplSUg5YqwUgtHbv+Kl1Imcar4wCY9HzKNinTOpSHqjptcARFSTphNM79qyUuD6lq4UN0XZWrPOTqJkQ10C6UbXcaPS/2a7onlXUSQiIiIiIiK6q0XB2xQkLGmLqHWJiB0dhuBewQgOicCElxKxadsyjKklkJj5n/Tq01XYwysacx4z9ZbVIevLLUrRzyJ/te40TohihS3IPy2K0pqmDrfF5sTSGvQOjRBlYVQEelvGxU1hgS7e5hQZcm735FejEREinb90DUIjY7D4w8+RPCdYrED3qiDzFxpA71nrsXhqBELk56BXGCJnLUbypvWV8/Nb+zod6TbG/rL9rATC29SMS4U4/9BabPpwMWLGKW0JjZyB2Hc24Yu3Jtb8kis0BonGlwDrsXXHWkwQK+oPrMPKD5RyhQjE/176b4CxrEN60iKmLSWqhyaVM91S+fkyUaqs+S+DoFnSB22mtRQ1NahzIF1x43zNy4iIiIiIiIjuXk5wMv0T2UUDjaMep3NzkFOgM6degVcE5r8VY5FjvJF4jcXStbHmvOi6jCQkbFPK9ZWyxyJlTOTrSH49GpGhYYicuQxbF0dUBM0taCxCBBovDQxn85FjOTiqPODj1LewdJSYpXuQdGeY7wPpOdAYoCuQnoPcQujLRbVLEKYsXYZIMXvbuQzEc/KgolUG0ZVaGP4i4l+q4QWPuz+CjC8BguAtP1slOmRtjcOcV5Is0ispghfEILK9UtZ93fDnj+h+1XSC6S0dbDZGDqS3edzduF6L8OCaA+r1DKRX1QwONmL2RERERERERHeHnVj5yhzErdqAlLR1iBs7HKMnTcP0p0biyT9V9KJVdY+syDHeGIyB9MWIED3eDUfW4eW5lrme60f33jvYYB7QVIWgyFjEr0hE/KwIeEOL9GpGhdRtTsCchcux5uNUpLw3DUNHjcf0KoOjahD21AxRpnuPDluWzsGit1djY9oWrHl+CB57SnoOJo1G+PProDXfBmGYOFOUbzeVCiqrQXTnpRWaFkr3dnT1gf3L+dAeyoH2jB7GDzVcNAgZvRhrrHvW94nFvNHikw59Fta92fDnj+h+1XSC6ddu1ZjWxaiFE1oEtqrU4GoD6nUMpLf18kJgt+4IGTQEDw1/BA8XtkePZA8EfeiGTps94Neyu3EdIiIiIiIiorubDtq9u5DyXgLiFi63GLBTWrL1ZXxywBRF9IHfAFFsqD7RWLmuciB9zlzLQUkllsGAlmp4imKFMHi2FkXZDfGnPKBp9CQs+jgHFR15DdAXpCMpdjYKXE190w0wmAYjLchBZto6JC1bhLhVlq2Q9rXqvzBlk1F1CLpzvZLpjis4sAtp65OQsDAOSQdEpezAcqz8T0UQ2ycwSpRvt6qD6KYv3FgxiK7Gu/qvRTKTMOfZaZjyxHAMHTINy00vkOSe9a/MF9sEIeb3v0SQMT27HlnvLUSyjRQ1RFSzphNMt+LY1iov+vUylL5xCJePm/9X06hSQN3OQHpzR0cEdO5iDJ77dngQZaWlOHH8GLL27sbOjHTj9O1Xmfhh52FcPVtqXEdeV95G3paIiIiIiIjoXqMrrj7dar31icGav8QiRMS05UD6rOjlyLIO5KVpUWgKhrt0QlC4KJt4DYCfSE8BFKJgvyga5SNt2TQ8NkQZ8HRA/yEIf+plrNn7CAI7iFWkbXKV0Rtrd7wIxaJI9y/txTs1KGcBdOaBdW0NomuPHCRvPVjR4zwgGMpIApEY2t000KkaIS9tR/a+fWKKrgjSayIQL9etixUVRFSdphNM/+EaTKmpZA4a56qNqy2g/lwbuwLp3j6+CA17CConJxz4LguHDuzHqZ9O4vLlS7h+3YBbt24ZJ7ks18nL5HXkdeVt5G3lfRARERERERHdVUYnIsMURNsWLwJtJlEI7WEaBtQAfZEoyvyCEdrHNHCofTSDYpG8cgaCPZR5UyDdOo+zYguyc825NRA6IbpSvvPg6WEVAb+zOUgXgfHItyqCghlJVtu8NMacnx15WUg5Ihc0iP1QnP++3Vgzy3ILIOjpEHQSZfkCFIgi3WO8pHvT9BzsXouYSskIghDd13wXQK/LFyWJVxBCBgVVus8aZguyLO77KoPohodWDKJboofySMZUtH3fdsRbvXjSPGgxmC8R3RbN3dw9/ijKP7ObaDbsATi5iFmXclzfdhE3rHO/3LyB8m90uNGjLZzbVITbm3fUVJqXWQfSg7p1R9t2Xsg9egSnC0/h+vXrYolt8roXdEUovnwZfv6doFarUVR0XiwlIiIiIiIiauJy3dBvUhgelD8Eb9UFg4d54ca1m1B3fhjPvzYXIzqIzmm6r7By3pc4LpdHLcMXq17EuDET8VTXc0jdroX5X9ldRiI6wh/Gf8Zfy0f6OmUbzaPz8belk9DZ1Bn2khZfZurwYGgYwiyn4LbQZWtRJO0xyxCEp8S+nHwHY1Q/L6jcO2LkzCWYG+kvxos0QLs5Dn/Zq4QVj18PxBMjuxiDh/I2Y4c+iAe6hCDi6Rfx4uNdoBxej6z3/4B/HpJbfQ2Xuo7EuO5yzpjm8OoVgR6trkHv6Il+417Bkkm9oTaGFaTjbE3A38Vx6B5z9RK6PPoUerSRys290OOhHlAbrqB5m354auESTOkjwtEGLT5f+nfskUM/3WOQ/FE8nh39FH45zAVZ/9qDig8sumBEdAT8lQcB+Rnr8OUx44IKNTwrOSX+5nvYpfNwPNTRgJIyN3R+6Nf448tReLCFvIF0F3+bjN+nfi+XKtou7a1z2Ej08FbDybUnHpn2Iv70VDBcRGjMcCgFC7dkSS1yhIu7Hj8d/h5HqkwlaN3TRwnAl2iR9mk69u/fi8xDJ437oDtP7eaGroHK68NjuVpc0TfsS4nG3h8BDr4d/G6J8s+uxYsD0KavqWf5DVz7RzaKvxaz1mpI6WJiHUjvGdwbzRwccPTI98ae5w3hIO2nW/ceuCnt53DOQVFLRERERERE1LRpRi/Dh69G1Ny7tlyH9NefwbytSrIIufd3fLhYW5eORSNfRpoypwTa48S+LJZV2qY2lfanwZiEj7D40Zq3q9q73fY2uow4TKo02GkwYtf9DVPMaS+qqr0XPd0T+shfTkSLPOLVMUD7wfOYkijugjnrkT3V9H2EFsn9J2O5mJO/6li6PR4RyoOA9MUjMW+bcUGFGp4V4z0ctxbzR/mIlz/VkAcmjbbIp26z7ZISqY2/k9pomQ++WnIvfZHqxfr5pp9Fex8fPD56rLH8761bcKbQlMO/fhp7f9TEcqZf//Y8KsLczeE83BMOYq6KGlK+yKrrkS4H0o98f7jaQLqm63BMjnkNb6/+B94zTn/FGwumYXRvP7FGZfI+5H3J+5T3TURERERERHQ3kAcafX7hBmSdqdo70Thw5+yKQLos6+uKHMy6nExkiXLj0yFl/jP4zbvpKLBumkGHnI8XYVKVALfYZu0unK5um/UvWwXSZTlYPnc2ktK0FoOWCjUeh+45B5ZjTmwS0nJ1qHob5GDjwkkVgXTZV1nQihUNR7KQrhQbgXQPL56GOfJ9f8mqJeV6FGQk4TeWgXSZ1PYpzyzCxqxC6C3zJcukbU5nbcCiaHsC6URUH02qZzrgCte/9kQr8UWN/HmMPj4HJT+I2epY9VCvLke6PIDo/n1ZVQLpmoHTMPf3kxHuZz5gFQbdIWxdHo+3tlfNlib3UO/bP8SYV11OG0NERERERER015BzQHf2hApl0B3PgtZ6YFATOWe6hx6ZByzyR99mmsAQBHk6AVcL7T6uX58w+LhKhfIiaPdqrYLo1TNvU4fj0L3GH8GhSqoTfeEu5NSULF9+XvyAE3beW/UiP2s+bnW4hzUIGtQNGkepyHv4nsCe6U1fEwumA80m9UDbyFZiDriVn4/zr56z6LFeDWNAvQdcfjpaKZDe3NHROGDo94dyjIOJVtAgfP47+NO4QOUzGoMOBdnZyD6cjW+O6qDu2Au9uvfC0CED4C3i7LpvVuAPL67HIWXWzN3dAz16BSNz19e4UW79SpCIiIiIiIiIiIjINgbTm74mF0xHC3eo/xoEF2cxLyn793e49ImNwULlQRmsVgno3AUqJydjgn1LQ+d/gjfH+UEFA3I/j8ercV/UMEq3BgN+tRB/ej4UmuaA/uBavPTrVVUC6l0Cg2AoK0PecesRJoiIiIiIiIiIiIhse8DbG0+M+YWxvCdzF3S6hg2ErNF4YnBomLH8ecq/8L/Tp41lqr+mF0yXOER2QdtJbSrypd8oQ8naQ9B/XTU/em0eGv4IDnyXhZKSElEj3URPJOCDxcOhgQ7Z/zcfLyRbh8ar0gycjbf/MhmBKgNyk2Pwq/+rvI2Liwv69AvB1zv+K2qIiIiIiIiIiIiI7Ofi4opJv3pWzDUeOfX1h/98H9euVcRIqX6a1ACkJrfSTuDyEYvAeXMnuEzrhlYDlbzo9mjr5QX9leJKgXRgOGY/JwfSDcj9qJpA+u8+wJ69e/HvhMdEhUL37Qq8lJAJHVQInDgH09qJBYJ8DPlY8jGJiIiIiIiIiIiI6qqk5KqxR7r1uI8NcfPmTezetZOB9EbSJHumG6k94LYsEC1dxLzsxnWUfvY9ijeX1Z5DXRLYrTvKSkuNg4OajXsbX84PhfrcF1jw5BJkiGozOZg+JRC6HUvw+PwvRKWJBs+v3oJpPVUo2DwVTyfkinqFPMipk7Mzco8eETV0Z6nRbcxEdD6fis++qcMnK47OaN8jBL5qPU5lHcCZMlF/P3NSw8P5Gi5dbsQxANoPw5OPtcXxf23G0cuijoiIiIiIiIiIqnByckYbTRvAwZy3o17koPyFoiIYDAZRQw3VJHumG+kvofiVY7hWKuZlzVvAeWwfaOJ8oPKtvelubu7QFxeLOcXoocHG0ZkLdq2vGki3SYe//ScH8q3n1/MxaJRKM/lY8jEblWNrdBnzAl569c9YEqdMr8T+CiEd5GGam472YxZgydxxaC/m7SIHbD3VaLQz6RqFqAFqXM2rSyA9ACPnLMHMX4xA+MNhCJBvjsZul53qdQ3ra/AM6V6agb5itvKxWyNs5gLMfvFX6GmcbyRncnHVfQCiRgSKCiIiIiIiIiIiqk5ZWSnOnD5tHDC0IZOcI52B9MbVdIPpsgsXULz4GKy/Qmje0Ret/9wfnn/uCOcBTnCQBx+10rKli1WKFw06esnRUj1Of1+5V7ndPspFvvynlx9CjRUV5GPJx2w0cqD3hZfxTF81fvp2A9YkLkfSR6k4ZOiIqBkvYmSnphVQr7O+EzF7zkT0ErMN44hug3qiVeFBfHNWVNnDpzeCPPTYs3YJlr7+LnbJYzo0arvuRhexJ2UDUjen4rCoaRzn8J32HFr1GIxud/mtS0RERERERERE96emHUyXnb2A4tjDKP7xulVql2Zo7vsA3Gf3Qbu/D4Dn8iC4z/OHm5gcVY4oL78u1pUNgLexO7kBZTbGMVV7B9YeTHX3REdRNJGP5dii8aKEHo88icEeF7Hng2XYmHoAp4rOoejITnyW9A4+O6XG4HGjq/Zidm0ND8921fesNve4doRz+x7o2TsAzqaVpGWe3YegW4D1dtK60v6cnaSicfsa9l2Tmtoj17vKb0BawEle7m61x9rOo1o90bOTI05pd8PyQwZj+zXyfqRJ4yzqZKJeOq4jLqL4qjhWQ9olLzMeQ9p3QB/0NHZzr4WjM1yN+2pd8TtUYdF+67aY1XSOlqR1jMcSv6UN5Rd+xLHCC2JOYnlu4lg17qeWa3RpXy7OOAagZw9R0RB1aZP53rW+1sp1cXW1rFTqKu1L/Fb2XDsiIiIiIiIiIrp3Nd2c6VU0Q/PHO8J9fDu0sGMc0h7JHtiZkW6RsD8Uf9ryNh5rr0PGH0dhQaqottR7Nv6ZOBmBLQHD0fX47bMrUHmIUmn5Xmm54RD+Nuw5rBW1MgcHBwwbHoH0L61zrdfHgxg5dxb6nFqPpR9/L+osdB2BZ4aqcWjLZhy+JFc4wvexWZg0xBvGcKp8fUpPI/2jVdh1QuS9llN7PN4C+w+2Rq+gFsYAoeP177FxYzHCJw2BR2k5HOWgom431rz7GU4ZNxuCZ+KehON3u6HuPQSet6RKR2mdy9J27603576W04TM7JKL997ajDPGGov2lJei3Hisk9i+5h/Yc6ZcWX+ARbD5xGd4be1uqVD7djUKnIh5UzriwKpl2F4o6twHYNzMcejpDpRftT435foOrpSVJw9fZ7fDQ/Vsl3INzuFoeQC6yS9tzNtWo1MUpj8zDL7Sj1UuX9LyPOzJa4fBvhbX0DEAYb+ejIj2ziiXfxtnR5T++CXWvJ+BItOlqPUcq1lHPpbDRew5qMfgftfx2eLV2C+tYv37VT9/EnuKAjDYzxHSkaT/K8WxtHfw0R69fBSJPb9dAKLmzkC3vNV4+195oq5+7GsT4NpvMmaM7gEPMS9/mXJ467vY/J28TjuEzYnFQ7oN+HPyAWWx8V7qA+T8E0s3ii9YjM9OS2x/813suapUERERERERERHR/ae5m7vHH0W5ibuFW8cu4dq2czA4tYBjRxc0ryWorjmswpmfCnHzpqkb+k8IjJyJvm0dceWHtfhMjiJaO7sXW/a7YNijwfBqH4zIMBdkb9mLc2Ixhj6OqZGdob50FJ+t/w8sw4EtWqjQvr0PfiwwJoJpGMdgDH2sCy5lfYzDp0SdJd0JHD5wFOdEN2zngdPw3Agf/PTFX/DXdf/GVzu/Q3GnRxEZ6gfdvv04J3fQ9+2Hh7v6AyfeQ+Kaz7Fz9ym49B+BEf1aIWfdG3j/s//iqx/d0XvQIHiWfCUd96a0UQf0ighEQJtb2LP2L1i3VVonRwe3PuF4pMtN7N1XADlMqg4ahv4aHb775iiumNoT4YajG5fjvU+/xM5dcntGYlSoBnm7vsfp3F346pqf1J6L+GzxW/jkgHKStrarnAG/gnOvcDzcUY9vpR9V+a2c0XPCswh1PYxPViUh5Uup3d8Z8ODwcHRTHcJ3eaeR941Ud0I6337X8KXchoz9KGhAu4zXoJMnyr79O95dvRkZYtuqAvHk86MRcDEDq1e8h9T/foXsywEYOawjnA2nxTWU2j/1dxjldgQfvbtCan+GdN2l9oc/hjDPU9h9RCftx9Y5ylFfZ4T8ahZCnb7HR6veRUrql8jMc8WwEX3g5ngRP0jn/D9pLevfr9r5Tg8A+/+O5e9vNZ5/SadQDOrubj5/+367i1AHPoJermeltloMDFwP9rQJHZ/Eb57uBxz8J95etQEZO75CXutBeOzhAXDK/wZ5l6+iuE0vDOmqwklpG/m9VPuw0dK5N4eq1U38IM6/2/BfoAdy8O+dx62+fCAiIiIiIiIiovtJ00/zYu36dVz/MA8XZmTh/Io86LMvwVBywyoFDGBwvgEXl8o5zHccLZD+vwqB/SZXGUDU7OAK/GrOeuRek9bsNhl/fX+2OeVL4LBAeEt/GvJyYN3/XD7WNevk7vXl0xatRdE2Z/TpFwDnE6n46JuLSlX5Rexfl4pjjgEYGGKZ+iMP2aligM6yXPxPjsmezcbXpt7rJ7KhvQK09ZLPssKlQ9uw6yexju4APsvIRXmH3hjgqVRVprQH2m347HvRQ1huz7bdKHINQt8ax5+s73ZAa3c1cOUi5NNRlOJw8mv481824JhOtPvybhwuBNq3l45RJ3Vo1+UD2P7fk8YXDDUK7I1urnrpd/gSZ8rkinJczdmMnT8alypch2BgQDkOp23GMdH7H7qdSN13Dq0694a/scKOc5T206sDcGzXJ+Z1yn/6EqmHKnpu200+tx3i3pHOP+uw1GBXNdoaK+y/Rv+7IC33aFv7QKvdxyHmDy8jqrujqKhBrW0C/Af2hceVbGz8l3S/GmvKcepfn2DPldboM1Bp1KUjJ3HJNQBdjHmb2iHAT42fdu/EKfcHEWC8vwPR2dcRZ7T7jMF2IiIiIiIiIiK6f919wXSzm7iZXYSSFbm4OCsb56Zm4dz873BeTBd/Oge1m5tYV5G7NRO5BkA1YDSm9RaV1TEH1A3I35ehpHppNxm/G+EnFfTIzlgv11QiH6u42BT5bKDC8xBhcTu0g1sr4NIFU34TobwQuitAK/d2oqIGZaU2e9ueP2vVi1inxxW4oXW1bySU9jgHPoV5f1hSMU0bAk84o1WNbwnqu10NHJ3RfuA4jHsuFrPl/bz6Gp6UA6YOymL71a1dNtLxA63V0lbncN4yeI5yJchs4tkaraS1gsa+UumYv+4v/Zau6oqUJbbO0bgfPXRWKXLOnDV/a1EnNZ9b4/52cs5yT9fWaO9pI5guqe16e7Rylu7VMyL1kMlJnNdJx2jVRpn98XvkX1WjY5B8bXugq+c55B/MxanL7dChs3T8jj3g76rHj9r6XTMiIiIiIiIiIrp33MXBdGs3cevMddwUk+7UebTWWEV7j67Ahzvl/st+GP9GAkbXFmeWA+oTx+Kl/5ND6b3wu6WzIKfTNhzdir9tVlaxJB/rwoWKvtENUn4Ol64CHTrU0B3bqwdCwvqgvWWn8zvuOiqN72qltHA/MnZlWExfIvWLVGTbyIJT3+0qc0bPKa9g5qhAuJ7/HrtT1+Pv776Gz06IxfXQOO2qi1L8dMDyeNK0I1U65h4oh2z8c2yoxrpGpV+vwmuLX8Gar+9EUpVcHM4rRXv/HnDt1Qm+l08ir+gkjv1YCv/OPeER9CA8LufiUKWXH0REREREREREdD+6h4LplZ0/exbqVm5VUr18sXwVMuSYt2Y45q58DePlzuY1OaeDrt0APP/3dzC5mwow5GLjOysghiU0k48hH0s+ZuPIRfZRPZx7jLAaJFOmRt8nf4mosK5obow1njb2tPXwCkCl2LprAHxbyddBpMFoAOugvnPHdvCAHherPd1zuFgMOJadQdaunRXTniwc2r8bx2q8RPXdDrh4WQ+5+3PFq5O+6BXgiPzty7Au5UtkHczDJZ0bWlf+UMFO9W9XtS7Ivfp94FvpkjrD18ti4NOii7hywxGlZ3ZXPub+Azi0R8ntbdc5Gvejhq9/5bcuvr4+otRY7L9GD7SRzvPSeave4reHsbe/Vwf4inlFoHStK3/Jkf/Djyj3ehChHXxwpeB7Y9vyc7Qof6ArhnRoh9IflToiIiIiIiIiIrq/3bPBdFnhqZPw6fCgmBPObcWCBSuQrQNUvo9h7vv/wru/H43AKr3U/TB0xmv4Z/JfMa233CX9NL5IeBH/d1AstiAfQz5WYzrz7w3Yo/fGyN/F4smBAfDwbAePDj0w+NkXjKk8ju1IhTLEZTn2f5eL0g7hmPRIAJzlzBiuARg8ORy+5bk4kF1rBm+7OPcYjSeDlVwdjh1GYNJDD6I0bx8OyGNcVlGKrL25KO86GtNN7XFqh26/+B3mvfwcQlyVtXBRL63ZDg90dYSjk7ySndtVozT/NC4194G/OUB9AVektrXt1Aeu8n7kdCiPjMMAe9KNNGK7qvXDDnxb5Iy+T0xEF42xcVLbnkW4ZZr6qzux+zjQbdRzGNxJCYQ7evXBk7MWYN70cPHSxI5zNO6nHL5hz2JwB+VYrsETMbpnY3/SYO81EoHss5ZD994+Z/YcwBnnPhg3oQ88nKQKuU0TRqOv8zl8t8fieT14EMektg3uIT1Xh8SrstwfkO8ahF6epTiaY/36jIiIiIiIiIiI7kfN3dw9/ijK95zi4mIEdeuO4suXUVZmkTLi7CFs+/I4WgYPRrCPJ7x7PIRfTJqJqU8/jUlTp2Lq1JmY+dunMap/Z3g6A4aCHUha8BxW7LwmdlDB3d0Dfv6dcCjnIG7dvClqG8HNi8j77hhKfIMxZNBQDBkyGIP6ByPASYc9m5OQst8ix/bZ75Fb5ov+ESPwSPgjeDisHwJwDJ/9cz0OFYs2+fbDw12BHzL2439KDdr3fQSB+AFfHVDC8tLZIGDoAHheyMbeXDn/ewf0igjEha//A9Uj0zBu5CMY1t8fqvM78eHaDFwQu1YHDUN/jQ7ffXMUV+QK6/Y8NBg93C9K7V6HnYUG4zbQnURJx1CEDx+BYb3V+EHe1p7tqnP5JtoNGoAuLU5h9xH5swMdCi9q0HNYOB6V9zP8YfRw+BbZxQF48JbF+Xp0w6B+LXHS4prUt11VrkGNruKk9jI0/cPxyPBwPBwRjv7qI0jN9UBgG9P2N1F0JBfFHQZj1KMjpXWk6z6wB1pfzMbGj1JxzjhwqT3nKO/nJG50H4JHpfORjzU04Cr++81FBPpV3AvWbbc1b2R9P9nz23V9BE8McMWR1FTkNXB4AbvadPU4cs+7o2voCIyQrmFFm9YiI9/yfiqCU6dwBLrnYdenB6U5U90I9NBY1hERERERERER0f3MwbeD3y1Rvid5+/jCt8OD2L8vC7duVT1Vv2HTMO3Z0Rja1RtqlaiU3TBA91MWMj/9EH/bkI3qsqE7ODigb/8QnPrpJE4XmgLSt4MjnD3bAPpzKDUGUmuirOd8/QIuXW54j3RgCJ6JexL49yv4aI+y7+bXLuDqVXv3LW2jkdqDYlzS1ZD/2skZzjdKUVppl3ZsZ8UjKhaze5/Gmjc3iB77sgZcj0ZqV20c3duhlaONfTk6w9XDDajxutt5jq6t4dGyHFeK9Kjjlaijmq6RI7pFL8GENruxIjFVpKq5Uxr3dyMiIiIiIiIiovvTPR9Ml8m901UqFY58f1jU1KBdIAZ09kTZ6UwcKhB1tejeoycMBgO0R4+ImnuNZTBdVDVVrlJb5z6Jll+9gTU7LHrtU9PgHo7pL4ZD/6/XsPHg7Q3nExERERERERER3Q73RTBd1jO4N5o5OODoke+r7aFeF3KP9G7de+CmtJ/DOdUkUb9n3EXBdCIiIiIiIiIiIqLb6L4JpsvkHupu7h7IO/YDLl+uX6IJOUd6QJeuKJa2v3d7pBMRERERERERERGRpfsqmC6Tc6h37tIV58+fQ+FPJ1FSUiKW1M7FxQU+HR5E27btcPzYD7c5RzoRERERERERERERNSX3XTBd1tzREX5+/vDxfRD6K8W4qNNBX1xsDKyXl183ruPo2MIYQFe7uaG1RgN1KzcUnjqJgoJ83ChnzmciIiIiIiIiIiKi+8l9GUy31NbLC23aaODm5o6WLV3g2MLRWF9+vRzXrpWguPgyLlzQ4fzZs8Z6IiIiIiIiIiIiIrr/3PfBdCIiIiIiIiIiIiIiW5qJP4mIiIiIiIiIiIiIqAYMphMRERERERERERER2cBgOhERERERERERERGRDQymExERERERERERERHZwGA6EREREREREREREZENDKYTEREREREREREREdnAYDoRERERERERERERkQ0MphMRERERERERERER2cBgOhERERERERERERGRDQymExERERERERERERHZwGA6EREREREREREREZENDKYTEREREREREREREdnAYDoRERERERERERERkQ0MphMRERERERERERER2cBgOhERERERERERERGRDQymExERERERERERERHZwGA6EREREREREREREZENDKYTEREREREREREREdnAYDoRERERERERERERkQ0Ovh38bomyTX4BAaJERERERERERERERHT3K8jLE6Xa1SmYTkRERERERERERER0P2KaFyIiIiIiIiIiIiIiGxhMJyIiIiIiIiIiIiKygcF0IiIiIiIiIiIiIiIbGEwnIiIiIiIiIiIiIrKBwXQiIiIiIiIiIiIiIhsYTCciIiIiIiIiIiIisoHBdCIiIiIiIiIiIiKiWgH/D+O933CIZ0OXAAAAAElFTkSuQmCC"},"d05bef0e-9d9f-40c2-b64f-7c9f6ec88b2c.png":{"image/png":"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"}}},{"cell_type":"code","source":"metrics = [error_rate, MultiLogLoss()]\nlearn_effb3 = vision_learner(dls, 'efficientnet_b3', metrics=metrics).to_fp16()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T14:19:36.615367Z","iopub.execute_input":"2025-11-28T14:19:36.615697Z","iopub.status.idle":"2025-11-28T14:19:38.468755Z","shell.execute_reply.started":"2025-11-28T14:19:36.615669Z","shell.execute_reply":"2025-11-28T14:19:38.467976Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"learn_effb3.lr_find(num_it=100)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T14:19:38.469813Z","iopub.execute_input":"2025-11-28T14:19:38.470266Z","iopub.status.idle":"2025-11-28T14:20:00.408756Z","shell.execute_reply.started":"2025-11-28T14:19:38.470244Z","shell.execute_reply":"2025-11-28T14:20:00.407934Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"learn_effb3.fine_tune(10, 0.0020892962347716093)\nlearn_effb3.recorder.plot_loss()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T14:21:33.741985Z","iopub.execute_input":"2025-11-28T14:21:33.742856Z","iopub.status.idle":"2025-11-28T14:29:07.314862Z","shell.execute_reply.started":"2025-11-28T14:21:33.742821Z","shell.execute_reply":"2025-11-28T14:29:07.313997Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"plot_multilogloss(learn_effb3)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T14:29:07.316384Z","iopub.execute_input":"2025-11-28T14:29:07.316615Z","iopub.status.idle":"2025-11-28T14:29:07.467921Z","shell.execute_reply.started":"2025-11-28T14:29:07.316592Z","shell.execute_reply":"2025-11-28T14:29:07.467214Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"submit(learn_effb3, \"submission_efficientnet_b3.csv\")","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T14:29:07.468679Z","iopub.execute_input":"2025-11-28T14:29:07.468887Z","iopub.status.idle":"2025-11-28T14:29:28.397784Z","shell.execute_reply.started":"2025-11-28T14:29:07.468871Z","shell.execute_reply":"2025-11-28T14:29:28.396898Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"### Result (First score - Private Score, Second score - Public Score)\n![image.png](attachment:ce0d7d04-8cde-4df6-88a5-5a915afcf609.png)","metadata":{},"attachments":{"8c21bd2c-2faa-4180-8747-33bfc713239e.png":{"image/png":"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"},"ce0d7d04-8cde-4df6-88a5-5a915afcf609.png":{"image/png":"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"}}},{"cell_type":"code","source":"import pandas as pd\nimport numpy as np\n\na = pd.read_csv('/kaggle/working/submission_resnet50.csv')     # model A\nb = pd.read_csv('/kaggle/working/submission_efficientnet_b3.csv')      # model B\n\n# Ensure same order and identical Image column\nassert (a['Image'].values == b['Image'].values).all(), \"Image orders differ!\"\n\n# columns that are class probs\ncols = [c for c in a.columns if c != 'Image']\n\n# simple (unweighted) average\nprobs = (a[cols].values + b[cols].values) / 2.0\n\n# re-normalize rows just in case\nprobs = probs / probs.sum(axis=1, keepdims=True)\n\nout = pd.DataFrame(probs, columns=cols)\nout['Image'] = a['Image']\nout.to_csv('ensemble_avg.csv', index=False)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T14:32:12.235726Z","iopub.execute_input":"2025-11-28T14:32:12.236091Z","iopub.status.idle":"2025-11-28T14:32:18.810635Z","shell.execute_reply.started":"2025-11-28T14:32:12.236062Z","shell.execute_reply":"2025-11-28T14:32:18.809808Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"### Result (First score - Private Score, Second score - Public Score)\n![image.png](attachment:a58ad7d6-76c8-48b7-a0d3-b7b272b4e267.png)","metadata":{},"attachments":{"6a9b77ee-96d9-4b42-9e3e-3684a1bd0a7e.png":{"image/png":"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"},"a58ad7d6-76c8-48b7-a0d3-b7b272b4e267.png":{"image/png":"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"}}},{"cell_type":"code","source":"import torch\nimport torch.nn as nn\n\nclass BaseCNN(nn.Module):\n    def __init__(self, n_classes):\n        super().__init__()\n        self.net = nn.Sequential(\n            # --- Layer 1: Detect Edges ---\n            # Input: 3 colors -> Output: 16 features\n            nn.Conv2d(3, 16, kernel_size=3, padding=1), \n            nn.ReLU(), \n            nn.MaxPool2d(2), \n\n            # --- Layer 2: Detect Shapes ---\n            # Input: 16 features -> Output: 32 features\n            nn.Conv2d(16, 32, kernel_size=3, padding=1), \n            nn.ReLU(), \n            nn.MaxPool2d(2),\n\n            # --- Layer 3: Detect Objects ---\n            # Input: 32 features -> Output: 64 features\n            nn.Conv2d(32, 64, kernel_size=3, padding=1), \n            nn.ReLU(), \n            \n            # --- Classifier Head ---\n            # Squashes everything to a single list of numbers and predicts\n            nn.AdaptiveAvgPool2d(1),  \n            nn.Flatten(),\n            nn.Linear(64, n_classes)\n        )\n\n    def forward(self, x):\n        return self.net(x)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T14:51:52.067203Z","iopub.execute_input":"2025-11-28T14:51:52.067521Z","iopub.status.idle":"2025-11-28T14:51:52.073661Z","shell.execute_reply.started":"2025-11-28T14:51:52.067490Z","shell.execute_reply":"2025-11-28T14:51:52.072998Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# instantiate and train\nn_classes = dls.c\nmodel = BaseCNN(n_classes)\nlearn_cnn = Learner(dls, model, loss_func=CrossEntropyLossFlat(), metrics=metrics).to_fp16()\nlearn_cnn.lr_find(num_it=100)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T14:54:08.594320Z","iopub.execute_input":"2025-11-28T14:54:08.595176Z","iopub.status.idle":"2025-11-28T14:54:16.701174Z","shell.execute_reply.started":"2025-11-28T14:54:08.595130Z","shell.execute_reply":"2025-11-28T14:54:16.700195Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"learn_cnn.fit_one_cycle(10, 0.005248074419796467)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T14:54:29.979738Z","iopub.execute_input":"2025-11-28T14:54:29.980055Z","iopub.status.idle":"2025-11-28T14:56:11.990557Z","shell.execute_reply.started":"2025-11-28T14:54:29.980029Z","shell.execute_reply":"2025-11-28T14:56:11.989703Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"learn_cnn.recorder.plot_loss()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T14:56:11.992333Z","iopub.execute_input":"2025-11-28T14:56:11.992636Z","iopub.status.idle":"2025-11-28T14:56:12.225932Z","shell.execute_reply.started":"2025-11-28T14:56:11.992608Z","shell.execute_reply":"2025-11-28T14:56:12.224988Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"plot_multilogloss(learn_cnn)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T14:56:12.226811Z","iopub.execute_input":"2025-11-28T14:56:12.227121Z","iopub.status.idle":"2025-11-28T14:56:12.370136Z","shell.execute_reply.started":"2025-11-28T14:56:12.227092Z","shell.execute_reply":"2025-11-28T14:56:12.369381Z"}},"outputs":[],"execution_count":null},{"cell_type":"code","source":"# predict on test and write submission\nimport pandas as pd\ntest_df = pd.read_csv('/kaggle/input/noaa-right-whale-recognition/sample_submission.csv')\ntest_dl = learn_cnn.dls.test_dl(test_df['Image'])\npreds, _ = learn_cnn.get_preds(dl=test_dl)\nprobs = preds.cpu().numpy()\nout = pd.DataFrame(probs, columns=learn_cnn.dls.vocab)\nout['Image'] = test_df['Image']\nout.to_csv('base_cnn_submission.csv', index=False)","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2025-11-28T14:56:12.371644Z","iopub.execute_input":"2025-11-28T14:56:12.371924Z","iopub.status.idle":"2025-11-28T14:56:26.233448Z","shell.execute_reply.started":"2025-11-28T14:56:12.371906Z","shell.execute_reply":"2025-11-28T14:56:26.232459Z"}},"outputs":[],"execution_count":null},{"cell_type":"markdown","source":"### Result (First score - Private Score, Second score - Public Score)\n![image.png](attachment:58fc024c-ba7b-4baf-93a0-b1998e712029.png)","metadata":{},"attachments":{"7208f40d-913b-44d7-98a5-650b48c8c91b.png":{"image/png":"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"},"58fc024c-ba7b-4baf-93a0-b1998e712029.png":{"image/png":"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"}}}]}