diff --git "a/Instadeep_NT_500M_CPT/Neucletide_transformer.ipynb" "b/Instadeep_NT_500M_CPT/Neucletide_transformer.ipynb" new file mode 100644--- /dev/null +++ "b/Instadeep_NT_500M_CPT/Neucletide_transformer.ipynb" @@ -0,0 +1,3338 @@ +{ + "cells": [ + { + "cell_type": "code", + "execution_count": 1, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "E: Could not open lock file /var/lib/dpkg/lock-frontend - open (13: Permission denied)\n", + "E: Unable to acquire the dpkg frontend lock (/var/lib/dpkg/lock-frontend), are you root?\n", + "Requirement already satisfied: tqdm in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (4.67.1)\n" + ] + } + ], + "source": [ + "!pip install -q transformers datasets accelerate bitsandbytes pysam pandas pyarrow fastparquet\n", + "!apt-get install -q samtools\n", + "!pip install tqdm\n" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Hello\n" + ] + } + ], + "source": [ + "print(\"Hello\")" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Collecting evaluate\n", + " Downloading evaluate-0.4.6-py3-none-any.whl.metadata (9.5 kB)\n", + "Requirement already satisfied: datasets>=2.0.0 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from evaluate) (4.4.2)\n", + "Requirement already satisfied: numpy>=1.17 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from evaluate) (2.2.6)\n", + "Requirement already satisfied: dill in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from evaluate) (0.4.0)\n", + "Requirement already satisfied: pandas in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from evaluate) (2.3.3)\n", + "Requirement already satisfied: requests>=2.19.0 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from evaluate) (2.32.5)\n", + "Requirement already satisfied: tqdm>=4.62.1 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from evaluate) (4.67.1)\n", + "Requirement already satisfied: xxhash in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from evaluate) (3.6.0)\n", + "Requirement already satisfied: multiprocess in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from evaluate) (0.70.18)\n", + "Requirement already satisfied: fsspec>=2021.05.0 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from fsspec[http]>=2021.05.0->evaluate) (2025.10.0)\n", + "Requirement already satisfied: huggingface-hub>=0.7.0 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from evaluate) (0.36.0)\n", + "Requirement already satisfied: packaging in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from evaluate) (25.0)\n", + "Requirement already satisfied: filelock in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from datasets>=2.0.0->evaluate) (3.20.3)\n", + "Requirement already satisfied: pyarrow>=21.0.0 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from datasets>=2.0.0->evaluate) (22.0.0)\n", + "Requirement already satisfied: httpx<1.0.0 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from datasets>=2.0.0->evaluate) (0.28.1)\n", + "Requirement already satisfied: pyyaml>=5.1 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from datasets>=2.0.0->evaluate) (6.0.3)\n", + "Requirement already satisfied: aiohttp!=4.0.0a0,!=4.0.0a1 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from fsspec[http]>=2021.05.0->evaluate) (3.13.3)\n", + "Requirement already satisfied: anyio in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from httpx<1.0.0->datasets>=2.0.0->evaluate) (4.12.1)\n", + "Requirement already satisfied: certifi in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from httpx<1.0.0->datasets>=2.0.0->evaluate) (2026.1.4)\n", + "Requirement already satisfied: httpcore==1.* in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from httpx<1.0.0->datasets>=2.0.0->evaluate) (1.0.9)\n", + "Requirement already satisfied: idna in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from httpx<1.0.0->datasets>=2.0.0->evaluate) (3.11)\n", + "Requirement already satisfied: h11>=0.16 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from httpcore==1.*->httpx<1.0.0->datasets>=2.0.0->evaluate) (0.16.0)\n", + "Requirement already satisfied: typing-extensions>=3.7.4.3 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from huggingface-hub>=0.7.0->evaluate) (4.15.0)\n", + "Requirement already satisfied: hf-xet<2.0.0,>=1.1.3 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from huggingface-hub>=0.7.0->evaluate) (1.2.0)\n", + "Requirement already satisfied: aiohappyeyeballs>=2.5.0 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]>=2021.05.0->evaluate) (2.6.1)\n", + "Requirement already satisfied: aiosignal>=1.4.0 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]>=2021.05.0->evaluate) (1.4.0)\n", + "Requirement already satisfied: async-timeout<6.0,>=4.0 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]>=2021.05.0->evaluate) (5.0.1)\n", + "Requirement already satisfied: attrs>=17.3.0 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]>=2021.05.0->evaluate) (25.4.0)\n", + "Requirement already satisfied: frozenlist>=1.1.1 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]>=2021.05.0->evaluate) (1.8.0)\n", + "Requirement already satisfied: multidict<7.0,>=4.5 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]>=2021.05.0->evaluate) (6.7.0)\n", + "Requirement already satisfied: propcache>=0.2.0 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]>=2021.05.0->evaluate) (0.4.1)\n", + "Requirement already satisfied: yarl<2.0,>=1.17.0 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]>=2021.05.0->evaluate) (1.22.0)\n", + "Requirement already satisfied: charset_normalizer<4,>=2 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from requests>=2.19.0->evaluate) (3.4.4)\n", + "Requirement already satisfied: urllib3<3,>=1.21.1 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from requests>=2.19.0->evaluate) (2.6.3)\n", + "Requirement already satisfied: exceptiongroup>=1.0.2 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from anyio->httpx<1.0.0->datasets>=2.0.0->evaluate) (1.3.1)\n", + "Requirement already satisfied: python-dateutil>=2.8.2 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from pandas->evaluate) (2.9.0.post0)\n", + "Requirement already satisfied: pytz>=2020.1 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from pandas->evaluate) (2025.2)\n", + "Requirement already satisfied: tzdata>=2022.7 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from pandas->evaluate) (2025.3)\n", + "Requirement already satisfied: six>=1.5 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from python-dateutil>=2.8.2->pandas->evaluate) (1.17.0)\n", + "Downloading evaluate-0.4.6-py3-none-any.whl (84 kB)\n", + "Installing collected packages: evaluate\n", + "Successfully installed evaluate-0.4.6\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Collecting scikit-learn\n", + " Downloading scikit_learn-1.7.2-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (11 kB)\n", + "Requirement already satisfied: numpy>=1.22.0 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from scikit-learn) (2.2.6)\n", + "Collecting scipy>=1.8.0 (from scikit-learn)\n", + " Downloading scipy-1.15.3-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (61 kB)\n", + "Collecting joblib>=1.2.0 (from scikit-learn)\n", + " Downloading joblib-1.5.3-py3-none-any.whl.metadata (5.5 kB)\n", + "Collecting threadpoolctl>=3.1.0 (from scikit-learn)\n", + " Downloading threadpoolctl-3.6.0-py3-none-any.whl.metadata (13 kB)\n", + "Downloading scikit_learn-1.7.2-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (9.7 MB)\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m9.7/9.7 MB\u001b[0m \u001b[31m186.6 MB/s\u001b[0m \u001b[33m0:00:00\u001b[0m\n", + "\u001b[?25hDownloading joblib-1.5.3-py3-none-any.whl (309 kB)\n", + "Downloading scipy-1.15.3-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (37.7 MB)\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m37.7/37.7 MB\u001b[0m \u001b[31m207.9 MB/s\u001b[0m \u001b[33m0:00:00\u001b[0m\n", + "\u001b[?25hDownloading threadpoolctl-3.6.0-py3-none-any.whl (18 kB)\n", + "Installing collected packages: threadpoolctl, scipy, joblib, scikit-learn\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━���━━\u001b[0m \u001b[32m4/4\u001b[0m [scikit-learn][0m [scikit-learn]\n", + "\u001b[1A\u001b[2KSuccessfully installed joblib-1.5.3 scikit-learn-1.7.2 scipy-1.15.3 threadpoolctl-3.6.0\n", + "Collecting matplotlib\n", + " Downloading matplotlib-3.10.8-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (52 kB)\n", + "Collecting contourpy>=1.0.1 (from matplotlib)\n", + " Downloading contourpy-1.3.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (5.5 kB)\n", + "Collecting cycler>=0.10 (from matplotlib)\n", + " Downloading cycler-0.12.1-py3-none-any.whl.metadata (3.8 kB)\n", + "Collecting fonttools>=4.22.0 (from matplotlib)\n", + " Downloading fonttools-4.61.1-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (114 kB)\n", + "Collecting kiwisolver>=1.3.1 (from matplotlib)\n", + " Downloading kiwisolver-1.4.9-cp310-cp310-manylinux_2_12_x86_64.manylinux2010_x86_64.whl.metadata (6.3 kB)\n", + "Requirement already satisfied: numpy>=1.23 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from matplotlib) (2.2.6)\n", + "Requirement already satisfied: packaging>=20.0 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from matplotlib) (25.0)\n", + "Collecting pillow>=8 (from matplotlib)\n", + " Downloading pillow-12.1.0-cp310-cp310-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl.metadata (8.8 kB)\n", + "Collecting pyparsing>=3 (from matplotlib)\n", + " Downloading pyparsing-3.3.1-py3-none-any.whl.metadata (5.6 kB)\n", + "Requirement already satisfied: python-dateutil>=2.7 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from matplotlib) (2.9.0.post0)\n", + "Requirement already satisfied: six>=1.5 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from python-dateutil>=2.7->matplotlib) (1.17.0)\n", + "Downloading matplotlib-3.10.8-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (8.7 MB)\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m8.7/8.7 MB\u001b[0m \u001b[31m151.4 MB/s\u001b[0m \u001b[33m0:00:00\u001b[0m\n", + "\u001b[?25hDownloading contourpy-1.3.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (325 kB)\n", + "Downloading cycler-0.12.1-py3-none-any.whl (8.3 kB)\n", + "Downloading fonttools-4.61.1-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (4.9 MB)\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m4.9/4.9 MB\u001b[0m \u001b[31m169.2 MB/s\u001b[0m \u001b[33m0:00:00\u001b[0m\n", + "\u001b[?25hDownloading kiwisolver-1.4.9-cp310-cp310-manylinux_2_12_x86_64.manylinux2010_x86_64.whl (1.6 MB)\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m1.6/1.6 MB\u001b[0m \u001b[31m129.4 MB/s\u001b[0m \u001b[33m0:00:00\u001b[0m\n", + "\u001b[?25hDownloading pillow-12.1.0-cp310-cp310-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl (7.0 MB)\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m7.0/7.0 MB\u001b[0m \u001b[31m153.0 MB/s\u001b[0m \u001b[33m0:00:00\u001b[0m\n", + "\u001b[?25hDownloading pyparsing-3.3.1-py3-none-any.whl (121 kB)\n", + "Installing collected packages: pyparsing, pillow, kiwisolver, fonttools, cycler, contourpy, matplotlib\n", + "\u001b[2K \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m7/7\u001b[0m [matplotlib]7\u001b[0m [matplotlib]\n", + "\u001b[1A\u001b[2KSuccessfully installed contourpy-1.3.2 cycler-0.12.1 fonttools-4.61.1 kiwisolver-1.4.9 matplotlib-3.10.8 pillow-12.1.0 pyparsing-3.3.1\n" + ] + } + ], + "source": [ + "!pip install evaluate\n", + "!pip install scikit-learn\n", + "!pip install matplotlib\n" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/home/zeus/miniconda3/envs/cloudspace/lib/python3.10/site-packages/tqdm/auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n", + " from .autonotebook import tqdm as notebook_tqdm\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "NUM_PROCS: 8 MAX_LEN: 1000\n", + "Loaded: 299999 train | 14073 test\n", + "Feature means/std computed.\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Setting TOKENIZERS_PARALLELISM=false for forked processes.\n", + "Tokenizing train (num_proc=8): 100%|██████████| 299999/299999 [01:30<00:00, 3321.33 examples/s]\n", + "Setting TOKENIZERS_PARALLELISM=false for forked processes.\n", + "Packing/normalizing features train (num_proc=8): 100%|██████████| 299999/299999 [00:30<00:00, 9687.11 examples/s] \n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[train] sample keys: ['labels', 'input_ids', 'attention_mask', 'features']\n", + "[train] shapes/dtypes: input_ids=torch.Size([1000])/torch.int64, features=9/torch.float32, labels=torch.int64\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Setting TOKENIZERS_PARALLELISM=false for forked processes.\n", + "Tokenizing test (num_proc=8): 100%|██████████| 14073/14073 [00:05<00:00, 2406.31 examples/s]\n", + "Setting TOKENIZERS_PARALLELISM=false for forked processes.\n", + "Packing/normalizing features test (num_proc=8): 100%|██████████| 14073/14073 [00:02<00:00, 4779.46 examples/s]" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "[test] sample keys: ['labels', 'input_ids', 'attention_mask', 'features']\n", + "[test] shapes/dtypes: input_ids=torch.Size([1000])/torch.int64, features=9/torch.float32, labels=torch.int64\n", + "✅ Cell 3 complete. Train size: 299999 Test size: 14073\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "\n" + ] + } + ], + "source": [ + "# ==== CELL 3: Re-tokenize + normalize extra features + build torch-format HF Datasets ====\n", + "import os\n", + "import numpy as np\n", + "import pandas as pd\n", + "import torch\n", + "from datasets import Dataset\n", + "from transformers import AutoTokenizer\n", + "\n", + "os.environ[\"TOKENIZERS_PARALLELISM\"] = \"true\"\n", + "NUM_PROCS = min(8, max(1, (os.cpu_count() or 8) - 2))\n", + "MAX_LEN = 1000\n", + "MODEL_NAME = \"InstaDeepAI/nucleotide-transformer-500m-human-ref\"\n", + "\n", + "# --------------- feature list (same order everywhere) ---------------\n", + "feature_cols = [\n", + " \"gnomad_af\",\n", + " \"GERP++_RS_rankscore\",\n", + " \"GERP_91_mammals_rankscore\",\n", + " \"phyloP100way_vertebrate_rankscore\",\n", + " \"phyloP470way_mammalian_rankscore\",\n", + " \"phyloP17way_primate_rankscore\",\n", + " \"phastCons100way_vertebrate_rankscore\",\n", + " \"phastCons470way_mammalian_rankscore\",\n", + " \"phastCons17way_primate_rankscore\",\n", + "]\n", + "\n", + "print(\"NUM_PROCS:\", NUM_PROCS, \"MAX_LEN:\", MAX_LEN)\n", + "\n", + "# 1) load\n", + "train_df = pd.read_parquet(\"Balanced_300k_SEQUENCES.parquet\")\n", + "test_df = pd.read_parquet(\"test_enriched_SEQUENCES.parquet\")\n", + "print(\"Loaded:\", len(train_df), \"train |\", len(test_df), \"test\")\n", + "\n", + "# basic check\n", + "for c in feature_cols:\n", + " if c not in train_df.columns:\n", + " raise RuntimeError(f\"Missing column in train dataframe: {c}\")\n", + "\n", + "# 2) compute train mean/std for normalization (stable)\n", + "train_feat = train_df[feature_cols].astype(float).fillna(0.0)\n", + "feat_means = train_feat.mean(axis=0).astype(np.float32).values\n", + "feat_stds = train_feat.std(axis=0).replace(0, 1).astype(np.float32).values\n", + "\n", + "print(\"Feature means/std computed.\")\n", + "\n", + "# 3) tokenizer\n", + "tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME, trust_remote_code=True)\n", + "\n", + "# 4) helper to build dataset\n", + "from datasets import Dataset\n", + "\n", + "def build_and_tokenize(df, split_name=\"train\"):\n", + " # keep only required columns\n", + " sub = df[[\"raw_sequence\", \"clean_label\"] + feature_cols].copy()\n", + " sub[\"labels\"] = (sub[\"clean_label\"] == \"Pathogenic\").astype(np.int64)\n", + " sub = sub.drop(columns=[\"clean_label\"])\n", + " ds = Dataset.from_pandas(sub, preserve_index=False)\n", + "\n", + " # tokenization\n", + " def tok_fn(batch):\n", + " enc = tokenizer(\n", + " batch[\"raw_sequence\"],\n", + " truncation=True,\n", + " padding=\"max_length\",\n", + " max_length=MAX_LEN\n", + " )\n", + " return enc\n", + "\n", + " ds = ds.map(\n", + " tok_fn,\n", + " batched=True,\n", + " batch_size=128,\n", + " num_proc=NUM_PROCS,\n", + " remove_columns=[\"raw_sequence\"],\n", + " desc=f\"Tokenizing {split_name}\"\n", + " )\n", + "\n", + " # pack & normalize features\n", + " def pack_norm_features(batch):\n", + " # stack columns in correct order and normalize by train stats\n", + " arrs = [np.array(batch[c], dtype=np.float32) for c in feature_cols]\n", + " stacked = np.stack(arrs, axis=1) # (batch, n_features)\n", + " # normalize\n", + " stacked = (stacked - feat_means) / feat_stds\n", + " return {\"features\": stacked.tolist()}\n", + "\n", + " ds = ds.map(\n", + " pack_norm_features,\n", + " batched=True,\n", + " batch_size=512,\n", + " num_proc=NUM_PROCS,\n", + " remove_columns=feature_cols,\n", + " desc=f\"Packing/normalizing features {split_name}\"\n", + " )\n", + "\n", + " # set torch format (ensures Trainer receives torch tensors)\n", + " ds.set_format(type=\"torch\", columns=[\"input_ids\", \"attention_mask\", \"features\", \"labels\"])\n", + "\n", + " # sample sanity\n", + " s = ds[0]\n", + " print(f\"[{split_name}] sample keys: {list(s.keys())}\")\n", + " print(f\"[{split_name}] shapes/dtypes: input_ids={s['input_ids'].shape}/{s['input_ids'].dtype}, features={len(s['features'])}/{s['features'].dtype}, labels={s['labels'].dtype}\")\n", + " return ds\n", + "\n", + "train_fast = build_and_tokenize(train_df, \"train\")\n", + "test_fast = build_and_tokenize(test_df, \"test\")\n", + "\n", + "print(\"✅ Cell 3 complete. Train size:\", len(train_fast), \"Test size:\", len(test_fast))\n" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "🔍 RUNNING PRE-FLIGHT CHECKS...\n", + "✅ GPU DETECTED: NVIDIA H200\n", + " VRAM: 150.12 GB\n", + " 🚀 GOD MODE HARDWARE CONFIRMED.\n", + "✅ Train Set: 299999 samples\n", + "✅ Test Set: 14073 samples\n", + " Structure: OK (Torch tensors present)\n", + "✅ Detected modern Transformers: Using 'eval_strategy'\n", + "\n", + "System Ready. Using strategy key: 'eval_strategy'\n" + ] + } + ], + "source": [ + "# ===== CELL 3.5 — SYSTEM & VARIABLE VERIFICATION =====\n", + "import torch\n", + "import transformers\n", + "from transformers import TrainingArguments\n", + "import psutil\n", + "\n", + "print(\"🔍 RUNNING PRE-FLIGHT CHECKS...\")\n", + "\n", + "# 1. HARDWARE VERIFICATION (Confirm H200)\n", + "if torch.cuda.is_available():\n", + " gpu_name = torch.cuda.get_device_name(0)\n", + " vram_gb = torch.cuda.get_device_properties(0).total_memory / 1e9\n", + " print(f\"✅ GPU DETECTED: {gpu_name}\")\n", + " print(f\" VRAM: {vram_gb:.2f} GB\")\n", + " \n", + " if \"H200\" in gpu_name or \"H100\" in gpu_name:\n", + " print(\" 🚀 GOD MODE HARDWARE CONFIRMED.\")\n", + " else:\n", + " print(\" ⚠️ WARNING: You are not on H100/H200. Adjust batch sizes!\")\n", + "else:\n", + " raise RuntimeError(\"❌ NO GPU DETECTED! Training will fail.\")\n", + "\n", + "# 2. DATASET VERIFICATION\n", + "try:\n", + " print(f\"✅ Train Set: {len(train_fast)} samples\")\n", + " print(f\"✅ Test Set: {len(test_fast)} samples\")\n", + " \n", + " # Check Column Names (Must match model expectations)\n", + " sample = train_fast[0]\n", + " required_keys = [\"input_ids\", \"attention_mask\", \"features\", \"labels\"]\n", + " missing = [k for k in required_keys if k not in sample.keys()]\n", + " \n", + " if missing:\n", + " raise ValueError(f\"❌ DATASET MISSING KEYS: {missing}\")\n", + " print(\" Structure: OK (Torch tensors present)\")\n", + " \n", + "except NameError:\n", + " raise RuntimeError(\"❌ Datasets 'train_fast' or 'test_fast' not found. Did Cell 3 run?\")\n", + "\n", + "# 3. HUGGING FACE ARGUMENT CHECK (The \"eval_strategy\" Bug Fix)\n", + "# Newer transformers use 'eval_strategy', older use 'evaluation_strategy'\n", + "import inspect\n", + "args_sig = inspect.signature(TrainingArguments.__init__)\n", + "valid_params = args_sig.parameters.keys()\n", + "\n", + "if \"eval_strategy\" in valid_params:\n", + " EVAL_STRATEGY_KEY = \"eval_strategy\"\n", + " print(\"✅ Detected modern Transformers: Using 'eval_strategy'\")\n", + "else:\n", + " EVAL_STRATEGY_KEY = \"evaluation_strategy\"\n", + " print(\"⚠️ Detected older Transformers: Using 'evaluation_strategy'\")\n", + "\n", + "print(f\"\\nSystem Ready. Using strategy key: '{EVAL_STRATEGY_KEY}'\")" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Some weights of EsmModel were not initialized from the model checkpoint at InstaDeepAI/nucleotide-transformer-500m-human-ref and are newly initialized: ['pooler.dense.bias', 'pooler.dense.weight']\n", + "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "🔥 Encoder gradient checkpointing enabled\n", + "🚀 NT-500M biological fine-tuning started (A100, fresh run)\n" + ] + }, + { + "data": { + "text/html": [ + "\n", + "
\n", + " \n", + " \n", + " [4688/4688 2:54:57, Epoch 2/2]\n", + "
\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
StepTraining LossValidation LossRoc Auc
5000.5270000.5226920.821170
10000.3965000.5082160.858456
15000.3657000.5876820.874515
20000.3274000.4427120.891527
25000.3001000.4356390.907509
30000.2371000.4246970.911374
35000.2282000.4147450.915350
40000.2243000.4234890.917674
45000.2172000.4222780.918179

" + ], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "💾 SAVING FINAL H200 MODEL...\n", + "✅ H200 MODEL SAVED → ./PathoPreter_Final_A100\n", + " (Config updated with 'custom_feature_dim' for easier inference)\n", + "🚀 A100 RUN COMPLETE.\n" + ] + } + ], + "source": [ + "# ===== CELL 4 — SAFE BIOLOGICAL FINE-TUNE (NT-500M, A100) — CORRECTED =====\n", + "import os\n", + "import torch\n", + "import torch.nn as nn\n", + "import numpy as np\n", + "from transformers import AutoModel, TrainingArguments, Trainer, EarlyStoppingCallback\n", + "from transformers.modeling_outputs import SequenceClassifierOutput\n", + "import evaluate\n", + "from safetensors.torch import save_file\n", + "\n", + "# 1. Optimizations for A100 (Ampere)\n", + "os.environ[\"PYTORCH_CUDA_ALLOC_CONF\"] = \"expandable_segments:True\"\n", + "torch.backends.cuda.matmul.allow_tf32 = True\n", + "torch.backends.cudnn.allow_tf32 = True\n", + "\n", + "# -----------------------------\n", + "# Metrics\n", + "# -----------------------------\n", + "auc_metric = evaluate.load(\"roc_auc\")\n", + "\n", + "def compute_metrics(eval_pred):\n", + " logits, labels = eval_pred\n", + " # Softmax implementation for stability\n", + " exp = np.exp(logits - np.max(logits, axis=1, keepdims=True))\n", + " probs = exp[:, 1] / exp.sum(axis=1)\n", + " return auc_metric.compute(prediction_scores=probs, references=labels)\n", + "\n", + "# -----------------------------\n", + "# Load NT-500M Encoder\n", + "# -----------------------------\n", + "MODEL_NAME = \"InstaDeepAI/nucleotide-transformer-500m-human-ref\"\n", + "FEATURE_DIM = len(feature_cols) # Ensure feature_cols is defined in previous cells\n", + "NUM_LABELS = 2\n", + "\n", + "encoder = AutoModel.from_pretrained(MODEL_NAME, trust_remote_code=True)\n", + "encoder.gradient_checkpointing_enable()\n", + "print(\"🔥 Encoder gradient checkpointing enabled\")\n", + "\n", + "hidden_size = encoder.config.hidden_size\n", + "\n", + "# -----------------------------\n", + "# Hybrid Classifier (Corrected Pooling)\n", + "# -----------------------------\n", + "class SequenceClassificationWithFeatures(nn.Module):\n", + " def __init__(self, encoder, hidden_size, feature_dim, num_labels):\n", + " super().__init__()\n", + " self.encoder = encoder\n", + " self.feature_dim = feature_dim\n", + " self.classifier = nn.Sequential(\n", + " nn.Linear(hidden_size + feature_dim, hidden_size),\n", + " nn.GELU(),\n", + " nn.Dropout(0.1),\n", + " nn.Linear(hidden_size, num_labels)\n", + " )\n", + "\n", + " def forward(self, input_ids=None, attention_mask=None, features=None, labels=None):\n", + " out = self.encoder(input_ids=input_ids, attention_mask=attention_mask, return_dict=True)\n", + "\n", + " # Robust pooling strategy\n", + " if hasattr(out, \"pooler_output\") and out.pooler_output is not None:\n", + " pooled = out.pooler_output\n", + " else:\n", + " # Mean pooling if pooler_output is missing\n", + " last = out.last_hidden_state\n", + " mask = attention_mask.unsqueeze(-1)\n", + " pooled = (last * mask).sum(1) / mask.sum(1).clamp(min=1e-9)\n", + "\n", + " # Ensure features match device and dtype\n", + " features = features.to(pooled.device).to(pooled.dtype)\n", + " \n", + " # Concatenate and Classify\n", + " x = torch.cat([pooled, features], dim=1)\n", + " logits = self.classifier(x)\n", + "\n", + " loss = None\n", + " if labels is not None:\n", + " loss = nn.CrossEntropyLoss()(logits, labels)\n", + "\n", + " return SequenceClassifierOutput(loss=loss, logits=logits)\n", + "\n", + "model = SequenceClassificationWithFeatures(\n", + " encoder=encoder,\n", + " hidden_size=hidden_size,\n", + " feature_dim=FEATURE_DIM,\n", + " num_labels=NUM_LABELS\n", + ")\n", + "\n", + "# -----------------------------\n", + "# Data collator\n", + "# -----------------------------\n", + "def data_collator(batch):\n", + " return {\n", + " \"input_ids\": torch.stack([b[\"input_ids\"] for b in batch]),\n", + " \"attention_mask\": torch.stack([b[\"attention_mask\"] for b in batch]),\n", + " # Improved tensor creation to handle varying input types safely\n", + " \"features\": torch.stack([torch.as_tensor(b[\"features\"], dtype=torch.float32) for b in batch]),\n", + " \"labels\": torch.tensor([b[\"labels\"] for b in batch], dtype=torch.long),\n", + " }\n", + "\n", + "# -----------------------------\n", + "# Training config\n", + "# -----------------------------\n", + "# Effective Batch Size = 64 * 2 = 128 (Ideal for NT-500M stability)\n", + "training_args = TrainingArguments(\n", + " output_dir=\"./PathoPreter_v1_A100\",\n", + "\n", + " per_device_train_batch_size=64,\n", + " gradient_accumulation_steps=2,\n", + " per_device_eval_batch_size=32,\n", + "\n", + " learning_rate=2e-5,\n", + " num_train_epochs=2,\n", + " weight_decay=0.02,\n", + " lr_scheduler_type=\"cosine\",\n", + " warmup_ratio=0.05,\n", + "\n", + " bf16=True, # Essential for A100\n", + " fp16=False,\n", + "\n", + " optim=\"adamw_torch_fused\",\n", + " max_grad_norm=0.5,\n", + "\n", + " dataloader_num_workers=8,\n", + " dataloader_pin_memory=True,\n", + "\n", + " eval_strategy=\"steps\",\n", + " eval_steps=500,\n", + " save_strategy=\"steps\",\n", + " save_steps=500,\n", + " logging_steps=200,\n", + "\n", + " load_best_model_at_end=True,\n", + " metric_for_best_model=\"roc_auc\",\n", + " greater_is_better=True,\n", + " save_total_limit=2,\n", + "\n", + " report_to=\"none\"\n", + ")\n", + "\n", + "torch.cuda.empty_cache()\n", + "\n", + "trainer = Trainer(\n", + " model=model,\n", + " args=training_args,\n", + " train_dataset=train_fast,\n", + " eval_dataset=test_fast,\n", + " data_collator=data_collator,\n", + " compute_metrics=compute_metrics,\n", + " callbacks=[EarlyStoppingCallback(early_stopping_patience=2)]\n", + ")\n", + "\n", + "print(\"🚀 NT-500M biological fine-tuning started (A100, fresh run)\")\n", + "trainer.train()\n", + "\n", + "# ========================================================\n", + "# 💾 TITANIUM SAVE — A100 MODEL\n", + "# ========================================================\n", + "\n", + "print(\"\\n💾 SAVING FINAL A100 MODEL...\")\n", + "\n", + "save_path = \"./PathoPreter_Final_A100\"\n", + "os.makedirs(save_path, exist_ok=True)\n", + "\n", + "# 1. Get the base encoder config\n", + "raw_model = model.encoder\n", + "if hasattr(raw_model, \"_orig_mod\"):\n", + " raw_model = raw_model._orig_mod\n", + "\n", + "# 2. INJECT CUSTOM DIMENSIONS INTO CONFIG\n", + "# This is crucial for easy inference loading later!\n", + "raw_model.config.custom_feature_dim = FEATURE_DIM\n", + "raw_model.config.custom_num_labels = NUM_LABELS\n", + "raw_model.config.architectures = [\"SequenceClassificationWithFeatures\"]\n", + "\n", + "# 3. Save Config and Tokenizer\n", + "raw_model.config.save_pretrained(save_path)\n", + "tokenizer.save_pretrained(save_path)\n", + "\n", + "# 4. Save Weights (Cleaning prefixes)\n", + "raw_state = model.state_dict()\n", + "clean_state = {}\n", + "\n", + "for k, v in raw_state.items():\n", + " k = k.replace(\"_orig_mod.\", \"\")\n", + " # Remove 'encoder.' prefix to match HF format, but KEEP 'classifier' keys\n", + " if k.startswith(\"encoder.\"):\n", + " clean_state[k.replace(\"encoder.\", \"\", 1)] = v\n", + " else:\n", + " clean_state[k] = v\n", + "\n", + "save_file(clean_state, os.path.join(save_path, \"model.safetensors\"))\n", + "\n", + "# 5. Save Scaling Factors\n", + "np.save(os.path.join(save_path, \"feat_means.npy\"), feat_means)\n", + "np.save(os.path.join(save_path, \"feat_stds.npy\"), feat_stds)\n", + "\n", + "print(f\"✅ A100 MODEL SAVED → {save_path}\")\n", + "print(\" (Config updated with 'custom_feature_dim' for easier inference)\")\n", + "print(\"🚀 A100 RUN COMPLETE.\")" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# # ==========================================\n", + "# # 🛑 EMERGENCY SAVE: RECOVER BEST CHECKPOINT\n", + "# # ==========================================\n", + "# # Run this ONLY if you interrupted training manually!\n", + "\n", + "# import os\n", + "# import shutil\n", + "# import torch\n", + "# import numpy as np\n", + "# from safetensors.torch import load_file, save_file\n", + "\n", + "# print(\"🚨 INTERRUPT DETECTED. INITIATING RECOVERY...\")\n", + "\n", + "# # 1. FIND THE BEST CHECKPOINT\n", + "# # The trainer object tracks this automatically\n", + "# if hasattr(trainer, \"state\") and trainer.state.best_model_checkpoint:\n", + "# best_ckpt_path = trainer.state.best_model_checkpoint\n", + "# print(f\"🏆 Best Model found at: {best_ckpt_path}\")\n", + "# else:\n", + "# # Fallback: If run was too short to find a 'best', use the latest output dir\n", + "# print(\"⚠️ No 'Best' checkpoint recorded (maybe too early?). Checking output dir...\")\n", + "# best_ckpt_path = trainer.args.output_dir\n", + "# # Find latest checkpoint folder manually if needed\n", + "# if not os.path.exists(os.path.join(best_ckpt_path, \"model.safetensors\")):\n", + "# # Sort folders by number\n", + "# subdirs = [x for x in os.listdir(best_ckpt_path) if \"checkpoint\" in x]\n", + "# if subdirs:\n", + "# subdirs.sort(key=lambda x: int(x.split(\"-\")[-1]))\n", + "# best_ckpt_path = os.path.join(best_ckpt_path, subdirs[-1])\n", + "# print(f\"👉 Fallback: Using latest checkpoint {best_ckpt_path}\")\n", + "\n", + "# # 2. LOAD THE BEST WEIGHTS\n", + "# print(f\"📥 Loading weights from: {best_ckpt_path}\")\n", + "# weights_file = os.path.join(best_ckpt_path, \"model.safetensors\")\n", + "\n", + "# if not os.path.exists(weights_file):\n", + "# # Try pytorch_model.bin if safetensors missing\n", + "# weights_file = os.path.join(best_ckpt_path, \"pytorch_model.bin\")\n", + "# if not os.path.exists(weights_file):\n", + "# raise FileNotFoundError(f\"❌ No weights found in {best_ckpt_path}\")\n", + "# # Load bin\n", + "# state_dict = torch.load(weights_file, map_location=\"cpu\")\n", + "# else:\n", + "# # Load safetensors\n", + "# state_dict = load_file(weights_file)\n", + "\n", + "# # 3. CLEAN & STRIP PREFIXES (The Titanium Fix)\n", + "# clean_state_dict = {}\n", + "# print(\"✂️ Cleaning weights for deployment...\")\n", + "\n", + "# for key, value in state_dict.items():\n", + "# # Remove compiler prefix\n", + "# key = key.replace(\"_orig_mod.\", \"\")\n", + " \n", + "# if key.startswith(\"encoder.\"):\n", + "# # \"encoder.embeddings...\" -> \"embeddings...\" (For AutoModel)\n", + "# new_key = key.replace(\"encoder.\", \"\", 1)\n", + "# clean_state_dict[new_key] = value\n", + "# else:\n", + "# # Keep classifier keys\n", + "# clean_state_dict[key] = value\n", + "\n", + "# # 4. SAVE FINAL MODEL\n", + "# final_path = \"./PathoPreter_Final_H200_Best\"\n", + "# os.makedirs(final_path, exist_ok=True)\n", + "\n", + "# print(f\"💾 Saving cleaned model to: {final_path}\")\n", + "# save_file(clean_state_dict, os.path.join(final_path, \"model.safetensors\"))\n", + "\n", + "# # Save Config & Tokenizer (From memory is fine)\n", + "# raw_model = model.encoder if hasattr(model, \"encoder\") else model\n", + "# if hasattr(raw_model, \"_orig_mod\"): raw_model = raw_model._orig_mod\n", + "# raw_model.config.save_pretrained(final_path)\n", + "# tokenizer.save_pretrained(final_path)\n", + "\n", + "# # Save Stats (Global variables)\n", + "# np.save(os.path.join(final_path, \"feat_means.npy\"), feat_means)\n", + "# np.save(os.path.join(final_path, \"feat_stds.npy\"), feat_stds)\n", + "\n", + "# print(\"\\n✅ SUCCESS!\")\n", + "# print(f\" Saved Best AUC Checkpoint from: {best_ckpt_path}\")\n", + "# print(f\" Location: {final_path}\")\n", + "# print(\" You can now download this folder.\")" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "🚀 Running final evaluation to get probabilities...\n" + ] + }, + { + "data": { + "text/html": [], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "✅ Extracted 14073 predictions.\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/home/zeus/miniconda3/envs/cloudspace/lib/python3.10/site-packages/IPython/core/pylabtools.py:170: UserWarning: Glyph 129516 (\\N{DNA DOUBLE HELIX}) missing from font(s) DejaVu Sans.\n", + " fig.canvas.print_figure(bytes_io, **kw)\n" + ] + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAA1IAAAJ1CAYAAADT+ME9AAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjgsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvwVt1zgAAAAlwSFlzAAAPYQAAD2EBqD+naQAA5VVJREFUeJzs3Wd4FGX/9vHvpvcQIECooffepdfQIaAiKIqKCLcFu6KiICrorYDKrShFUEFAqvQmoXcBpUPoPbQkpCc7z4v82ceYBJKQMJvk/ByHh9PnnN3ZJb+da66xGIZhICIiIiIiIhnmYHYAERERERGR3EaFlIiIiIiISCapkBIREREREckkFVIiIiIiIiKZpEJKREREREQkk1RIiYiIiIiIZJIKKRERERERkUxSISUiIiIiIpJJKqREREREREQySYWUiDwQI0eOxGKxEBISkmK6xWKhdevW97XtkJAQLBYLI0eOvK/tZNTp06exWCwMHDjwgewvI1q3bo3FYkkxLbtel8y+R5l5r9Nb1t6l9XpL7vCgvy9EJO9SISWST9y6deue/92+fTtT29yzZw/PPvssFStWxNPTE3d3d8qXL8+AAQNYs2ZNDh1J7jN9+nQsFkuK/9zd3alUqRIvvfQSly9fNjuiXXlQherAgQNTvCdOTk74+flRrVo1Hn/8cebNm0d8fHyOZsjP0vtcVKlShddee41r165leduBgYEEBgZmX9gs7P+fx+Xo6EjhwoXp2LEjixcvztF9Z8ePUyKSMU5mBxCRB8PPz++ey7Rq1SpDVwasVitvvPEG48ePx8nJibZt29KjRw+cnZ05efIky5Yt45dffuGjjz5ixIgRd93W4cOH8fDwyOhhpKlRo0YcPnyYwoUL39d2clq7du1o3rw5ANevX2fdunVMnDiRRYsW8eeff+Lv75+t+zPrdXnxxRd57LHHKF26dLYum1OeffZZSpYsiWEYREREcPz4cZYsWcKsWbOoWrUqs2fPplatWinW+emnn4iOjjYpcd7yz89FWFgYq1atYvz48SxYsIA9e/ZQqFAhkxNmjaOjI++//z4A8fHxHDlyhN9//501a9bwxRdf8Prrr5ucUETulwopkXzk0qVLFCtWLM158+bNY+LEiRnazvvvv8/48eOpU6cO8+bNo3z58inmx8TEMHHiRK5fv37PbVWpUiVD+7wbDw+PbNlOTmvfvj3vvPOObdxqtdK9e3eWL1/OxIkTGTVqVLbuz6zXpXDhwhku3jKzbE4ZNGgQTZo0STEtMjKSDz/8kPHjx9OxY0f+/PNPihcvbptvZuGX1/z7c5GQkEBQUBDr16/nm2++ybVN8JycnFJlX716NZ06deKDDz5g6NCh9/0jkoiYS037RCRTTpw4weeff06hQoVYuXJlqiIKwN3dnTfffDNDhUFazVDuNLk6deoUX3/9NVWqVMHV1ZUyZcowatQorFZriuXvds/D1atXef3116lcuTLu7u4ULFiQxo0b88UXX6RYbtq0afTs2ZPAwEDc3NwoWLCg7Y+5nOLg4GBrvrZnz540s7/66qtUqFABV1dXChcuTJ8+fThw4ECGtp/e67J+/XqeeeYZKleujJeXF15eXjRo0IAffvjhrts7f/48/fr1o3Dhwnh4eNCsWTPWrl2barnM3Pf072WnT59O2bJlAZgxY0aK5lEhISG8//77WCwW5s6dm+b2pk2bhsViYcyYMffc9914e3szbtw4Bg4cyJUrV/j4449TzE/rHqnw8HA+++wzWrVqRfHixXFxcaF48eI8+eSThIaGprmfa9euMXjwYIoUKYKHhwcNGzZk4cKFtmZv06dPty37zyaPhw8fJjg4mEKFCmGxWDh9+jQACxcupF+/flSoUAEPDw98fX1p0aIF8+fPT7Xvf2+vW7duFChQAD8/P/r162drWrdt2zbatWuHj48Pfn5+DBo0iKioqPt4de/O2dmZ559/HoBdu3YByZ+PF198kRo1auDr64u7uzs1a9Zk7NixJCQkpDqmM2fOcObMmRTnT1rfD7t376ZDhw54e3vj6+tLcHCw7bX8ty1bttC1a1cKFiyIm5sbVapU4cMPP8zUlcmOHTtSuXJloqOjOXjwIPD/z6XY2Fjef/99ypcvj7Ozc4q8p06dYtCgQZQuXRpXV1cCAgIYOHAgZ86csS1z5/MOsGHDhhTH/s/zCGDx4sW0a9cOPz8/3NzcqFGjBl988QVJSUkplvvnebhkyRKaNWuGt7e3qc0mReyJrkiJSKZMnz6dpKQknn/+eYoWLXrXZV1dXe9rX2+++SYbNmygW7duBAUFsWjRIkaOHEl8fDyffPLJPdc/evQobdq04dKlSzRv3pxevXoRFRXFwYMH+fTTT3njjTdsy77wwgvUrl2b9u3b4+/vz4ULF1i0aBHt27dnwYIF9OzZ876O5V6cnFJ+HYeGhtK6dWvOnz9Px44d6dWrF1evXmX+/PmsWrWKdevW0bhx4yzt67PPPuPEiRM0adKE4OBgbt26xcqVK3n++ec5evQoX375Zap1bt68SbNmzfD392fQoEGEhYUxZ84cOnXqxLx58+jVq1eWsvxbnTp1GDZsGF999RW1a9dOsd3AwECee+45xowZw5QpU3j00UdTrT958mScnJx4+umnsyXPiBEjmD59OnPnzuV///vfXTuYOHz4MB988AFt2rQhODgYT09Pjhw5wqxZs1i2bBl//vknZcqUsS1/+/ZtWrVqxaFDh3jooYdo2bIl58+f57HHHiMoKCjd/dx572rWrMnAgQO5fv06Li4uAAwfPhwXFxeaN29OQEAAYWFh/P777zz88MN8/fXXvPTSS6m2d+rUKR566CEaNGjAoEGD2L17N7Nnz+bcuXOMHTuWjh070qFDBwYPHkxISAhTp07FarUybdq0FNtp3bo1GzZsYP369dl2j86d13vy5MksWbKEli1b0qVLF6KjowkJCWH48OHs2rXLVigWKFCADz/8kAkTJgDwyiuvpMj3T7t27eLzzz+nTZs2PP/88+zdu5dFixbx999/c+DAAdzc3GzL/vbbb/Tr1w9XV1f69u1LkSJFWL16NR999BGrVq0iJCQkxfKZObY7+vTpw/79++nUqRMFChSw/aCwY8cOgoKCiIqKolu3blSsWJHTp08zc+ZMVqxYwbZt2yhXrhyBgYF8+OGHjBo1ijJlyqS4x7BOnTq24eHDhzN27FhKlChB79698fX1ZdOmTbz55pvs2LGD3377LVXW3377jdWrV9OtWzf+85//EBERkaljFcmzDBHJFwDj0qVL6c7/7bffjFatWt1zO61btzYAY+3atZna/4cffmgAxvr161Pl+vd+n3rqKQMwypYta1y8eNE2PSwszChQoIDh7e1txMXF2aavX7/eAIwPP/wwxXYaNGhgAMYPP/yQKs+5c+dSjJ88eTLVMhcvXjSKFy9uVKxYMcX0U6dOGYDx1FNP3eWI/78ff/zRAIwxY8akmJ6UlGR07tzZAIz//ve/KeY99NBDhqOjo7Fy5coU048ePWp4e3sbNWvWTDG9VatWxr+/0tN7XdI61oSEBKNDhw6Go6OjcebMmRTzAAMw+vfvb1itVtv0/fv3Gy4uLoa/v78RHR1tm56Z9zqtZe/1+nbu3NmwWCzGqVOnUkw/cOCAARi9evVKc71/u3Oebdu27a7LlSpVygCM0NBQ27S0Xu9bt24Z169fT7X+H3/8YTg4OBiDBg1KMf399983AGPw4MEppq9du9b2mv/444+26XdeF8D44IMP0sz6z4x3REZGGjVr1jR8fX2NqKioNLc3YcIE23Sr1Wp06dLFAIwCBQoYixYtss2Lj483atWqZTg5ORmXL19OsZ87r8m/3/f0pPe5SEhIMNq2bWsAxqhRowzDMIwzZ84YiYmJKZazWq3GM888YwDG5s2bU8wrU6aMUaZMmTT3e+dzARizZ89OMW/AgAEGYPz666+2aeHh4Yavr6/h6upq7N+/3zY9KSnJ6Nu3rwEYH330Uar9u7q6ptr32rVrDYvFYnh6eto+M3detzp16qQ6f+Lj443AwEDD29vb+PPPP1PM27Rpk+Ho6Gh069YtxfS0Pmd3rF692gCMoKAg4/bt27bpVqvVGDJkiAEY8+bNs02/8x45ODgYa9asSXObIvmZmvaJSKbc6WGuZMmSOb6vESNGEBAQYBsvXLgwPXv2JDIykqNHj9513Z07d7J7925atmzJc889l2r+v/Pf+fX3nwICAujTpw/Hjx9P0YQmq9auXcvIkSMZOXIkL7/8MjVq1GDFihU89NBDDB061Lbc3r172bp1K0899VSqKxOVKlXiueees/1qnhVpHauTkxNDhgwhKSkpzeaMjo6OfPrppyl+Ra9VqxYDBgwgLCyM5cuXZylLVgwZMgTDMJg6dWqK6VOmTAFI8/2+H3fujbpXL3K+vr4ULFgw1fQ2bdpQvXr1VM0gf/nlF1xcXPjoo49STG/Xrh0dO3ZMdz/FihXjvffeS3NeuXLlUk3z8vJi4MCBhIeH25rK/VP58uV5+eWXbeMWi4XHHnsMgLp166a4Guvs7MzDDz9MYmIihw4dSrGdn376icOHD9OoUaN0s6fln5+Ll156iWrVqvHHH39QtmxZXnzxRSD5njRHR8cU61ksFl544QXbNjKrZcuW9O3bN8W0Z555BiDF67R48WLCw8N55plnUnQ64uDgwOeff46Tk1OqpnMAiYmJtuN67733ePjhh+nUqROGYTB69Gjc3d1TLD9q1KhU58/SpUs5ffo0b775JnXr1k0xr3nz5vTs2ZPly5dn+ArRnftgf/jhBzw9PW3TLRYLY8eOxWKx8Ouvv6Zar2fPnrRv3z5D+xDJT9S0T0TsVv369VNNu1MA3bp1667r7ty5E+Cuf5D+08mTJxkzZgx//PEHFy5cIC4uLsX8ixcvpmiWlRXr1q1j3bp1KaY1a9aMdevWpWgGuX37dgCuXLmS5n0dR44csf2/Ro0amc4RGRnJF198waJFiwgNDU11v8vFixdTrVO6dOk0j79FixZMnTqVvXv30qdPn0xnyYquXbtSokQJfvzxR0aOHImjoyPx8fH8/PPPlCpVik6dOj2QHGkJCQlhwoQJ7Nixg2vXrpGYmGibd6f5HUBERASnT5+mWrVqaTaRbdasGatXr05zH7Vr106xrX+6evUqY8eOZcWKFZw5c4aYmJgU89N6b2vVqpWqmdmdHzD+2STs3/P+va2sdsDxz8+Fq6srgYGBvPbaawwfPtxWWMTHxzNx4kRmz57NkSNHuH37NoZh3PW47iWj3y979+4FUjcNhORjLleuHMeOHSMyMhJvb2/bvKSkJNt9og4ODvj5+dG2bVteeOEFevTokWpbaRWgd74Ljh49muZ3weXLl7FarRw7dowGDRqkf7D/2J6np2eqZpl3uLu7275f7pVNRFRIiUgmFStWjCNHjnDhwgUqV66co/vy8fFJNe3OvUT/vin638LDwwEoUaLEPfdz4sQJGjVqREREBG3atKF79+74+Pjg4OBASEgIGzZsSFVYZcWYMWN45513sFqtnD59mpEjR/Lzzz/z3HPP8dNPP9mWu3HjBgDLli1j2bJl6W4vKzf8x8fH07p1a/7880/q1q3LgAEDKFSoEE5OTpw+fZoZM2akeazp3Q93Z/qd1/tBcHR0ZNCgQYwaNYoVK1bQrVs3Fi5cyPXr13nxxRdxcMjexhZ3/ki/V/f0v/32G3379sXLy4ugoCACAwPx8PCw3az/z6uad64gFClSJM1t3e3+w/Tm3bhxg4YNG3L27FmaNWtG+/btKVCgAI6Ojuzbt4/Fixen+d7e7XN2t3n/7OThftz5XNzNww8/zJIlS6hUqZLtHiVnZ2du3brFV199laXPZ0a/X+68V+m97gEBARw7doyIiIgUhZSrqyuxsbEZzpPW9u98F8ycOfOu62b0u+DGjRskJibetSOgtLZ1r/thRfIrFVIikinNmjUjJCSEdevW0bZtW7PjpKtAgQIAXLhw4Z7Ljh8/nps3b/Lzzz/zxBNPpJg3ZMgQNmzYkK3ZHBwcKFeuHDNmzODMmTP8/PPP9O7d29axwp0/8L755htb06bssnjxYv7880+effZZW1O4O2bPns2MGTPSXO/KlSt3ne7r65utOe9l0KBBfPzxx0yePJlu3boxZcoUHBwcbE2zssvJkyc5d+4c/v7+9+ypbOTIkbi5ubFnzx4qVqyYYt7s2bNTjN95j69evZrmttJ7vSF1JwV3TJ06lbNnzzJ69Gjb84vuGDt2bI4/CDan7Nq1iyVLlhAUFMSyZctSNPHbvn07X331VY7u/857ld57cqe5c1qFWWak9b7e2eaSJUvo1q3bfW3/zvYsFkumH3Z8t05WRPIz3SMlIpkycOBAHB0d+eGHHwgLC7vrstlxFSer7jRFSa951D/d6Zr63z3zGYbBli1bsj/c/7FYLHz11VdYLBaGDx9u69b9Tm9827Zty/Z9pnesAJs2bUp3vbNnz6Z5n9iddf59/8b9uPOH8t2uOpYsWZKuXbuyfPlytm7dyrp16wgKCsr25zuNHj0agL59+97zj8nQ0FCqVq2aqoi6dOkSJ0+eTDHNx8eHwMBATpw4kWYxtXXr1kxnzep7a+/uHFfXrl1T3SeV3nE5Ojre86p1Rt05t9Pqzv/cuXOEhoZSrly5FFejsktWvgscHBzSPfbGjRtz/fp1jh8/ni35RPI7FVIikikVKlTgrbfe4tq1a3Tu3JlTp06lWiY2NpZx48aZ+iDNhg0b0rBhQzZu3MjkyZNTzf/nlao79/5s3rw5xTJjx47NcocOGVWnTh169erFkSNHbM13GjVqROPGjfn111+ZM2dOqnWsVmuWr5Kld6wbNmxI83W6IykpiXfffTfFfSl//fUXP//8M/7+/nTp0iVLedLi5+eHxWLh3Llzd13u+eefJzExkUceeQTDMLK1k4nbt2/z+uuvM336dAICAnj33XfvuU6ZMmU4ceJEiisXsbGxDB06NM1mcI8//jjx8fF8+OGHKaaHhISwatWqTGdO772dNWvWA+kM5OzZsxw5ciRTz1XKiPSO6+DBg+k+L6xgwYJcu3YtU03r0tOzZ098fX358ccfbc9+guQfWt5++20SExNTdDWenXr27Enp0qUZN24cGzduTDU/ISEh1etSsGBBzp8/n+b27nQq8swzz6T5wPTLly9z+PDhbEgukj+oaZ+IZNrHH39MbGws48ePp3LlyrRt25YaNWrg7OzMqVOnWLt2LdevX0/1ENMHbebMmbRu3ZrBgwfz888/07RpU2JjYzl48CB79+61/SExZMgQfvzxR/r06cOjjz5KoUKF2L59O3/++Sddu3a9631K2eHDDz9k0aJFfPTRR/Tr1w8nJyd+/fVX2rRpw2OPPcaECROoV68e7u7unD17lm3bthEWFpalPxK7d+9OYGAgn3/+OQcOHKBGjRocPXqUpUuXEhwczLx589Jcr1atWmzevJmGDRvSvn1723OkEhMT+eGHH1L1QHY/vLy8bEXwgAEDqFixIg4ODgwYMCBFhxedOnWiTJkynDlzhmLFitG9e/cs7W/KlCmsXLkSwzCIjIzk+PHjbNiwgcjISKpXr87s2bNT9B6ZnpdeeomXXnqJunXr2nq2W7NmDYZhULt2bfbv359i+bfffpv58+czadIkDhw4QIsWLTh//jxz586le/fuLFmyJFP3ew0YMIDPPvuMl156ifXr11OmTBn279/PunXr6N27NwsWLMj0a5MZTz75ZLY/RwqSf1ho1KgRc+fO5dKlSzRp0oSzZ8/y+++/07Vr1zTP2bZt27J79246d+5MixYtcHFxoWXLlrRs2TLT+/fx8WHy5Mn069ePxo0b07dvX/z9/Vm7di179uyhUaNGvPnmm9lxqKm4uroyb948OnfuTKtWrWjbti01a9a0PXR406ZNFCpUKEUHEW3btmXu3Ln06tWLunXr4ujoSI8ePahVqxadOnVixIgRjB49mgoVKtg+Q9evX+fEiRNs2rSJjz/+mKpVq+bI8YjkNSqkRCTTHBwcGDduHP379+e7775j48aNbNy4EavVSkBAAEFBQTz99NOmd5dbsWJF/vzzT8aMGcOSJUuYMGECXl5eVKxYMcU9JHXr1mX16tW8//77LFiwAEdHRx566CG2bNnC77//nuOFVO3atenduzfz58/np59+4plnnqFs2bLs3buXcePGsWjRIn788UccHR0JCAigZcuWPPzww1nal5eXF3/88QdvvvkmGzduJCQkhOrVqzNz5kyKFi2abiHl5+fHsmXLeOONN5g8eTLR0dHUrVuXUaNG0aFDh/s5/DT9/PPPvPrqqyxdupTw8HAMw6B58+YpCqk7xdXHH3/MwIEDUz3UOKPudKPu6OiIt7c3xYsXp0ePHvTq1YuePXvi7Oycoe288MILODs788033zB58mQKFChA165dGTNmDI888kiq5b29vdm4cSPDhw9n8eLF7N69m+rVq/Prr79y8uRJlixZkqn7bkqWLMmGDRt46623WLt2LYmJidSrV4/Vq1dz7ty5HC+kcoqjoyNLly7lnXfeYeXKlezatYuKFSvyxRdf0Llz5zTP2REjRnDz5k2WLl3Kpk2bSEpK4sMPP8xSIQXwyCOPUKxYMcaMGcOCBQuIjo4mMDCQESNG8Pbbb2f6YbyZ0bBhQ/bv389///tfli9fzpYtW3B1daVEiRL06tWLfv36pVj+zj1jf/zxB0uWLMFqtVKyZElb1+0fffQRLVu25Ouvv2bdunXcunWLQoUKUbZsWUaOHMnjjz+eY8ciktdYjH+20xCRPMtisXDp0iWKFSuW5vx58+YxceLENO8DELFX3bp1Y/ny5Rw7dowKFSqYHSfbPPHEE8ycOZNDhw7p6oCIiJ3SPVIiIpIrHTp0iOXLl9OhQ4dcW0RdunQp1bQNGzYwe/ZsKleurCJKRMSOqWmfSD5yr/s8WrVq9YCSiGTdrFmzOHr0qO3ZW//urCE36dKlC+7u7tSpUwdPT08OHTrEypUrcXR05JtvvjE7noiI3IWa9onkE7du3brnMk5OTnh5eeV8GJH70Lp1azZt2kSZMmUYMWIETz/9tNmRsmzChAnMnDmT0NBQIiMjKVCgAM2aNWP48OG2rq9FRMQ+qZASERERERHJJN0jJSIiIiIikkkqpERERERERDIp33c2YbVauXjxIt7e3lgsFrPjiIiIiIiISe48nL148eL3fCh6vi+kLl68SKlSpcyOISIiIiIiduLcuXOULFnyrsvk+0LK29sbSH6xMvME+ZxitVoJCwvD39//nlWwiM4XySydM5JZOmcks3TOSGbZ0zkTERFBqVKlbDXC3eT7QupOcz4fHx+7KaRiY2Px8fEx/UQS+6fzRTJL54xkls4ZySydM5JZ9njOZOSWH/tIKiIiIiIikouokBIREREREckkFVIiIiIiIiKZpEJKREREREQkk1RIiYiIiIiIZJIKKRERERERkUxSISUiIiIiIpJJKqREREREREQySYWUiIiIiIhIJqmQEhERERERySQVUiIiIiIiIpmkQkpERERERCSTVEiJiIiIiIhkkgopERERERGRTFIhJSIiIiIikkkqpERERERERDJJhZSIiIiIiEgm2VUhtXHjRrp3707x4sWxWCwsWrTonuuEhIRQr149XF1dqVChAtOnT8/xnCIiIiIikr/ZVSEVFRVF7dq1+d///peh5U+dOkXXrl1p06YN+/bt45VXXmHQoEGsWrUqh5OKiIiIiEh+5mR2gH/q3LkznTt3zvDykyZNomzZsnz55ZcAVK1alc2bNzN+/HiCgoJyKqaIiIiIiORzdlVIZda2bdto3759imlBQUG88sor6a4TFxdHXFycbTwiIgIAq9WK1WrNkZyZYbVaMQzDLrKI/dP5Ipmlc0YyS+eMZJbOGcmM6OhoIiIisFgsdnHOZCZDri6kLl++TNGiRVNMK1q0KBEREcTExODu7p5qnTFjxjBq1KhU08PCwoiNjc2xrBlltVoJDw/HMAwcHOyq5aXYIZ0vklk6ZySzdM5IZumckYw6e/YsGzZswNHRkfbt29vFORMZGZnhZXN1IZUVw4cP57XXXrONR0REUKpUKfz9/fHx8TExWTKr1YrFYsHf39/0E0nsn84XySydM5JZOmcks3TOyL0kJCSwdu1adu/eDUDhwoVxcXGhSJEipp8zbm5uGV42VxdSxYoV48qVKymmXblyBR8fnzSvRgG4urri6uqaarqDg4Ppb9wdFovFrvKIfdP5Ipmlc0YyS+eMZJbOGUnPhQsXWLhwIdevXwegcePGtGnThps3b9rFOZOZ/efqQqpp06YsX748xbQ1a9bQtGlTkxKJiIiIiMi/Wa1WNm3axIYNGzAMA29vb3r16kW5cuXs4t6orLCrQur27ducOHHCNn7q1Cn27dtHwYIFKV26NMOHD+fChQv89NNPAAwZMoSJEyfy1ltv8cwzz/DHH38wd+5cli1bZtYhiIiIiIjIv1gsFi5evIhhGNSoUYMuXbqk24Ist7CrQmr37t20adPGNn7nXqannnqK6dOnc+nSJc6ePWubX7ZsWZYtW8arr77KV199RcmSJZkyZYq6PhcRERERMZlhGCQlJeHk5ITFYqF79+6cPn2aGjVqmB0tW9hVIdW6dWsMw0h3/vTp09NcZ+/evTmYSkREREREMiMyMpLff/8dDw8PgoODAfDy8sozRRTYWSElIiIiIiK526FDh1i6dCkxMTE4OjrSqlUrChYsaHasbKdCSkRERERE7ltsbCwrV65k//79QHIP2717986TRRSokBIRERERkft0+vRpFi1aRHh4OBaLhWbNmtG6dWscHR3NjpZjVEiJiIiIiEiWJSYmsmDBAiIjI/Hz86NXr16ULl3a7Fg5ToWUiIiIiIhkmZOTEz169ODQoUMEBQXh6upqdqQHQoWUiIiIiIhkmNVqZfv27Xh7e1OzZk0AKlSoQIUKFUxO9mCpkBIRERERkQy5desWixYt4syZM7i6ulK2bFm8vLzMjmUKFVIiIiIiInJXhmHw119/sWLFCuLi4nB2dqZjx454enqaHc00KqQEgPDwcKKjo82OIZlktVq5du0aSUlJODg4mB1H0uHh4YGvr6/ZMURERLIkOjqapUuXcvjwYQBKlixJcHBwnu3WPKNUSAnh4eFMnDiRhIQEs6NIJhmGQWxsLG5ublgsFrPjSDqcnZ158cUXVUyJiEiuExcXx6RJk4iMjMTBwYHWrVvTrFkz/YCLCikh+VeGhIQEevfujb+/v9lxJBMMwyA6OhoPDw8VUnYqLCyMBQsWEB0drUJKRERyHVdXV6pXr86JEyfo3bs3AQEBZkeyGyqkxMbf318fjlzGMAyioqLw9PRUISUiIiLZ4vz583h6euLn5wdAu3btaNu2Lc7OziYnsy8qpEREREREhKSkJDZt2sTGjRspWbIkAwcOxMHBAScnlQxp0asiIiIiIpLPXbt2jYULF3Lx4kUAfH19SUxMxMXFxeRk9kuFlIiIiIhIPmUYBrt372b16tUkJibi5uZGly5dbA/alfSpkBIRERERyYdiYmJYsGABJ06cAKBs2bL07NlTnSNlkPotlGwREhKCxWLh1q1bZkdJpXXr1rzyyiu28cDAQCZMmGBaHnswcuRI6tSpY3aMLPvhhx8oVaoUDg4OTJgwIUPHM3DgQHr16vVA8omIiOQGLi4u3L59G0dHR4KCghgwYICKqExQISUZtm3bNhwdHenatavZUe7Lrl27GDx4sNkx7mr69OkUKFAgzXkWi4VFixY90Dz2JCIighdffJG3336bCxcuMHjwYN544w3WrVv3wLPk9oJURETyn9jYWJKSkgBwdHSkT58+PP/88zRp0kQ9AGeSCinJsKlTp/LSSy+xceNG242IuZG/vz8eHh5mx8j1zHqA89mzZ0lISKBr164EBATg4eGBl5cXhQoVMiWPiIhIbnH69GkmTZrEpk2bbNMKFy6s54hmkQopyZDbt28zZ84chg4dSteuXZk+fXqay23ZsoVatWrh5uZGkyZNOHDgQIr58+fPp3r16ri6uhIYGMiXX35pm/fuu+/SuHHjVNusXbs2H330kW18ypQpVK1aFTc3N6pUqcK3336bqWP5d9M+i8XClClTCA4OxsPDg4oVK/L777+nWOfAgQN07twZLy8vihYtyoABA7h27Vqm9psTQkJC8Pb2Zt26dTRo0AAPDw8eeughjh49mmK5sWPHUrRoUby9vXn22WeJjY1Nta27va6nT5/GYrEwZ84cWrVqhZubGzNnzkwz05YtW2jdujUeHh74+fkRFBTEzZs3geSno7/88ssUKVIENzc3mjdvzq5du1Icj8ViSfd4pk+fbrv5tVy5clgsFk6fPp3qylBSUhKvvfYaBQoUoFChQrz11lsYhpEip9VqZcyYMZQtWxZ3d3dq167NvHnzMpVl1KhR7N+/H4vFgsViSfdzISIiYqbExERWr17NjBkzCA8P58CBAyQmJpodK9dTISUZMnfuXKpUqULlypV54oknmDZtWqo/TAHefPNNvvzyS3bt2oW/vz/du3e3XbnYs2cPjz76KI899hh///03I0eOZMSIEbY/Ph9//HF27txJaGiobXsHDx7kr7/+on///gDMnDmTDz74gE8++YTDhw/z6aefMmLECGbMmHFfxzdq1CgeffRR/vrrL7p06cLjjz/OjRs3ALh16xZt27albt267N69m5UrV3LlyhUeffTRdLd39uxZvLy87vrfp59+el+Z/+n999/nyy+/ZPfu3Tg5OfHMM8/Y5s2dO5eRI0fy6aefsnv3bgICAlIVnxl9Xd955x2GDRvG4cOHCQoKSpVj3759tGvXjmrVqrFt2zY2b95M9+7dbU0I3nrrLebPn8+MGTP4888/qVChAkFBQbbX+o733nsvzePp27cva9euBWDnzp1cunSJUqVKpcrx5ZdfMn36dKZNm8bmzZu5ceMGCxcuTLHMmDFj+Omnn5g0aRIHDx7k1Vdf5YknnmDDhg0ZzvL6669TvXp1Ll26xKVLl+jbt2/6b5KIiIgJrly5wuTJk9m2bRsAdevW5bnnntOzobKDkc+Fh4cbgBEeHm52FMMwDCMpKcm4dOmSkZSU9MD2efHiRePDDz80Ll68mO4yDz30kDFhwgTDMAwjISHBKFy4sLF+/Xrb/PXr1xuAMXv2bNu069evG+7u7sacOXMMwzCM/v37Gx06dEix3TfffNOoVq2abbx27drGRx99ZBsfPny40bhxY9t4+fLljVmzZqXYxujRo42mTZumm71Vq1bGsGHDbONlypQxxo8fbxsHjPfff982fvv2bQMwVqxYYdt+x44dU2zz3LlzBmAcPXo0zX0mJCQYx48fv+t/169fTzfzjz/+aPj6+qY5DzAWLlxoGIZh/PHHHwZgrFmzxjZ/2bJlBmDExMQYhmEYTZs2Nf7zn/+k2Ebjxo2N2rVr28bv9bqeOnXKAGznQHr69etnNGvWLM15t2/fNpydnY2ZM2fapsXHxxvFixc3Pv/8c8Mw/v95tHbt2nSPZ+/evQZgnDp1yrbMhx9+mOJ4AgICbNs0jOT3o2TJkkbPnj0NwzCM2NhYw8PDw9i6dWuKjM8++6zRr1+/DGf5937TkpHP14NkxneM5G46ZySzdM7Yh6SkJGPz5s3G6NGjjZEjRxqff/65ceTIEbNjpcmezpnM1AYqReWejh49ys6dO22/6Ds5OdG3b1+mTp1K69atUyzbtGlT23DBggWpXLkyhw8fBuDw4cP07NkzxfLNmjVjwoQJJCUl4ejoyOOPP860adMYMWIEhmHw66+/8tprrwEQFRVFaGgozz77LM8995xtG4mJiffdw0ytWrVsw56envj4+HD16lUA9u/fz/r16/Hy8kq1XmhoKJUqVUo13cnJiQoVKtxXpsz4Z/6AgAAArl69SunSpTl8+DBDhgxJsXzTpk1Zv349kLnXtUGDBrbh6tWrc+bMGQBatGjBihUr2LdvH4888kiaGUNDQ0lISKBZs2a2ac7OzjRq1Mh2jmTkeO4lPDycS5cupWgm6uTkRIMGDWxXUU+cOEF0dDQdOnRIsW58fDx169bNtiwiIiJmCQ8PJyQkhKSkJCpXrkz37t3x9PQ0O1aeokJK7mnq1KkkJiZSvHhx2zTDMHB1dWXixInZ2k1mv379ePvtt/nzzz+JiYnh3LlztuZSt2/fBmDy5Mmp7qVydHS8r/06OzunGLdYLFitVtt+u3fvzmeffZZqvTt/WP/b2bNnqVat2l33+e677/Luu++mOc/Hx4eoqCisVisODv+/Be6d7uX//Zr/M/+dHnfu5L+XzLyu//wCXr58ua3Zpru7e4r/36/7OZ6MuHPMy5Yto0SJEinmubq6PtAsIiIiOcHPz4/OnTsDyc351CNf9lMhJXeVmJjITz/9xJdffknHjh1TzOvVqxe//vpriqsd27dvt/1Sf/PmTY4dO0bVqlUBqFq1Klu2bEmxjS1btlCpUiXbH+wlS5akVatWzJw5k5iYGDp06ECRIkUAKFq0KMWLF+fkyZM8/vjjOXbM/1avXj3mz59PYGBghtsTFy9enH379t11mYIFC6Y7r3LlyiQmJrJv3z7q1atnm/7nn38CpHkVLD1Vq1Zlx44dPPnkk7Zp27dvtw1n9XUtU6ZMqmm1atVi3bp1jBo1KtW88uXL4+LiwpYtW2zrJiQksGvXrhTP+bpfvr6+BAQEsGPHDlq2bAkkn8d79uyxvZbVqlXD1dWVs2fP0qpVqyzvy8XFxXb/l4iIiJmio6NZtmwZjRo1sv07+8+/IST7qZCSu1q6dCk3b97k2WefTXUVpE+fPkydOjVFIfXRRx9RqFAhihYtynvvvUfhwoVtD0F9/fXXadiwIaNHj6Zv375s27aNiRMnpur44PHHH+fDDz8kPj6e8ePHp5g3atQoXn75ZXx9fenUqRNxcXHs3r2bmzdv2poAZrcXXniByZMn069fP9566y0KFizIiRMnmD17NlOmTEnzatj9Nu2rXr06HTt25JlnnuHLL7+kXLlyHD16lFdeeYW+ffumuopyN8OGDWPgwIE0aNCAZs2aMXPmTA4ePEi5cuVsy2TX6zp8+HBq1qzJf/7zH4YMGYKLiwvr16/nkUceoXDhwgwdOpQ333yTggULUrp0aT7//HOio6N59tlnM/X6ZOSYx44dS8WKFalSpQrjxo1L8bBob29v3njjDV599VWsVivNmzcnPDycLVu24OPjw1NPPZWh/QQGBnLq1Cn27dtHyZIl8fb2TnVFS0REJKcdP36cxYsXExUVxZUrV/jPf/6TokWL5Ay9wnJXU6dOpX379mk23+vTpw+7d+/mr7/+sk0bO3Ysw4YNo379+ly+fJklS5bg4uICJP8qMnfuXGbPnk2NGjX44IMP+Oijjxg4cGCK7T788MNcv36d6OhoWxF2x6BBg5gyZQo//vgjNWvWpFWrVkyfPp2yZctm+7HfUbx4cbZs2UJSUhIdO3akZs2avPLKKxQoUCBHv6TudDX+/PPPU716dV5++WV69uzJlClTMrWdvn37MmLECN566y3q16/PmTNnGDp0aIplsut1rVSpEqtXr2b//v00atSIpk2bsnjxYtuVvLFjx9KnTx8GDBhAvXr1OHHiBKtWrcLPzy9T+7mX119/nQEDBvDUU0/RtGlTvL29CQ4OTrHM6NGjGTFiBGPGjKFq1ap06tSJZcuWZeqY+/TpQ6dOnWjTpg3+/v78+uuv2XocIiIidxMfH8/SpUuZNWsWUVFR+Pv706dPHxVRD4jFMNLowzofiYiIwNfXl/DwcHx8fMyOg9Vq5erVqxQpUuSBfQguXbrE999/z/PPP5/uPT9inwzDICoqCk9PT7V9tlP29vky4ztGcjedM5JZOmcejPPnz7Nw4ULbI0QaN25Mu3btUt33nRvY0zmTmdpATftERERERHKRy5cv257p6ePjQ8+ePVM02c9xEWch5lr6890Lg0/e791WhZSIiIiISC5StGhRKlWqhIuLC507d862XnMzJOIsTKsMSbHpL+PoBs8czfPFlAopERERERE7ZhgGe/fupWrVqri7u2OxWHj44Ycz3Jtwtoq5dvciCpLnx1xTISUiIiIiIuaIjIxk8eLFhIaGcvr0aXr37g1gThElKegOQHlgBg4cmKoXvpw0cuRI6tSpY9r+JXOio6Pp06cPPj4+WCwWbt26RWBgIBMmTLjrehaLhUWLFj2QjCIiIg/SwYMH+e677wgNDcXJyYkSJUqQz/uJsysqpOSeBg4ciMVisf1XqFAhOnXqlKLb89zgq6++Yvr06WbHyJClS5fSqlUrvL298fDwoGHDhrkme1bNmDGDTZs2sXXrVi5duoSvry+7du1i8ODBDzxLRgo4ERGRnBIbG8vChQuZN28eMTExBAQEMHjwYBo3bqxeeu2ICinJkE6dOnHp0iUuXbrEunXrcHJyolu3bmbHyhRfX18KFChgdox7+uabb+jZsyfNmjVjx44d/PXXXzz22GMMGTKEN954I0f3HR8fn6Pbv5vQ0FCqVq1KjRo1KFasGBaLBX9/fzw8PEzLJCIi8qBdvnyZ7777jr/++guLxUKLFi149tln8ff3Nzua/IsKKckQV1dXihUrRrFixahTpw7vvPMO586dIywszLbMuXPnePTRRylQoAAFCxakZ8+enD59Ot1txsXF8fLLL1OkSBHc3Nxo3rw5u3btss1v0KABX3zxhW28V69eODs7c/v2bSD5+QkWi4UTJ05k6Bj+3bSvdevWvPzyy7z11lsULFiQYsWKMXLkyBTr3Lp1i0GDBuHv74+Pjw9t27Zl//79GdpfVpw7d47XX3+dV155hU8//ZRq1apRoUIFXn/9df773//y5ZdfsmPHDqxWKyVLluS7775Lsf7evXtxcHDgzJkzGcp/p/njlClTKFu2LG5ubmnmslqtfP7551SoUAFXV1dKly7NJ598Ypv/999/07ZtW9zd3SlUqBCDBw+2vU/w/1/7L774goCAAAoVKsQLL7xAQkICkPxefPnll2zcuBGLxULr1q2B1FeGjh8/TsuWLXFzc6NatWqsWbMmzdfwbudhRrKcOXOGV1991XYVVkRE5EHx9fXFarXi5+fH008/Tdu2bXF0dDQ7lqRBhZRk2u3bt/nll1+oUKEChQoVAiAhIYGgoCC8vb3ZtGkTW7ZswcvLi06dOqV7leOtt95i/vz5zJgxgz///JMKFSoQFBRke7Bcq1atCAkJAZJ7q9m0aRMFChRg8+bNAGzYsIESJUpQoUKFLB/LjBkz8PT0ZMeOHXz++ed89NFHKf44f+SRR7h69SorVqxgz5491KtXj3bt2tkypqV69ep4eXml+1/nzp3TXXfevHkkJCSkeeXp+eefx8vLi19//RUHBwf69evHr7/+mmKZmTNn0qxZM8qUKZPh/CdOnGD+/PksWLCAffv2pZlr+PDhjB07lhEjRnDo0CFmzZpF0aJFAYiKiiIoKAg/Pz927drFb7/9xtq1a3nxxRdTbGP9+vWEhoayfv16ZsyYwfTp023NFRcsWMBzzz1H06ZNuXTpEgsWLEiVwWq10rt3b1xcXNixYweTJk3i7bffTrFMRs/De2UpWbIkH330ke0qrIiISE66efOm7d4nd3d3nnjiCYYMGUKpUqVMTpaG+AizE9gNdfchGbJ06VK8vLyA5D+cAwICWLp0qe3p03PmzMFqtTJlyhTbL/g//vgjBQoUICQkhI4dO6bYXlRUFN999x3Tp0+3FRaTJ09mzZo1TJ06lTfffJPWrVszdepUkpKSOHDgAC4uLvTt25eQkBA6depESEgIrVq1uq/jqlWrFh9++CEAFStWZOLEiaxbt44OHTqwefNmdu7cydWrV3F1dQXgiy++YNGiRcybNy/de3eWL19uu7qRlrs96+HYsWP4+voSEBCQap6LiwvlypXj2LFjADz++ON8+eWXnDt3jipVqmC1Wpk9ezbvv/8+QIbzx8fH89NPP6XbZCAyMpKvvvqKiRMn8tRTTwFQvnx5mjdvDsCsWbOIjY3lp59+wtPTE4CJEyfSvXt3PvvsM1vB5efnx8SJE3F0dKRKlSp07dqVdevW8dxzz1GwYEE8PDxwcXGhWLFiaeZYu3YtR44cYdWqVRQvXhyATz/9NEVhmtHz8F5ZHB0d8fb2TjeLiIhIdrBarWzbto3169fTvXt3ateuDWD7t9PuGAbs+u+9l3N0S34obx6nQkoypE2bNrZmZDdv3uTbb7+lc+fO7Ny5kzJlyrB//35OnDiBt7d3ivViY2MJDQ1Ntb3Q0FASEhJo1qyZbZqzszONGjXi8OHDALRo0YLIyEj27t3L1q1badWqFa1bt2bs2LFA8hWpN998876Oq1atWinGAwICuHr1KgD79+/n9u3btqtud8TExKR5THfcuRqU0+rUqUPVqlWZO3cuH3zwARs2bODq1as88sgjQMbzlylTxlZEbdq0KUVh8v3331OxYkXi4uJo165dmjkOHz5M7dq1bUUUQLNmzbBarRw9etT2j0H16tVTNE0ICAjg77//zvDxHj58mFKlStmKKICmTZumWCaj5+H9ZhEREblft27dYtGiRbbm+CdPnrQVUnbrwDQ4tRwsTtBpOhSqmvZy7oXz/DOkQIWUZJCnp2eKJnRTpkzB19eXyZMn8/HHH3P79m3q16/PzJkzU62b1ZsjCxQoQO3atQkJCWHbtm106NCBli1b0rdvX44dO8bx48fv+4qUs7NzinGLxYLVagWSmzAGBATYmhf+O1t6qlevbvtSTEuLFi1YsWJFmvMqVapEeHg4Fy9eTFEwQPKVo9DQUNq0aWOb1r9/f3799Vc++OADZs2aRadOnWyFU0bz/7MAatCgQYrmfUWLFr3rfW6ZcbfXOrtk9Dx8EFlERETSYhgG+/fvZ8WKFcTHx+Pi4kKnTp1SPLLFLt04Cn+8nDzc4lOo9ri5eeyACinJEovFgoODAzExMQDUq1ePOXPmUKRIEXx8fO65fvny5XFxcWHLli22KzgJCQns2rWLV155xbZcq1atWL9+PTt37uSTTz6hYMGCVK1alU8++YSAgAAqVaqUI8cHycd0+fJlnJycCAwMzPB699O0r0+fPrz99tt8+eWXfPnllynmTZo0iaioKPr162eb1r9/f0aMGMGePXuYN28ekyZNuq/87u7uqe45q1ixIu7u7qxbt45BgwalWqdq1apMnz6dqKgoW1G2ZcsWHBwcqFy5cob2mxFVq1bl3LlzXLp0ydb0cfv27SmWyex5mB4XFxeSkpLuK6+IiMi/RUdHs3TpUlvrm1KlShEcHIyfn5/Jye4hMQ6W9YPEaCjdDhq8bnYiu6DOJiRD4uLiuHz5MpcvX+bw4cO89NJL3L59m+7duwPJ9+sULlyYnj17smnTJk6dOkVISAgvv/wy58+fT7U9T09Phg4dyptvvsnKlSs5dOgQzz33HNHR0Tz77LO25Vq3bs2qVatwcnKiSpUqtmkzZ86876tR99K+fXuaNm1Kr169WL16NadPn2br1q2899577N69O931ypQpQ4UKFdL9r0SJEumuW7p0aT7//HMmTJjAe++9x5EjRwgNDWXcuHG89dZbvP766zRu3Ni2fGBgII0bN2bQoEEkJSXRo0eP+87/b25ubrz99tu89dZb/PTTT4SGhrJ9+3amTp0KJL/3bm5uPPXUUxw4cID169fz0ksvMWDAgGxt492+fXsqVarEU089xf79+9m0aRPvvfdeimUyex6mJzAwkI0bN3LhwgWuXbuWbccgIiL529WrVzl8+DAODg60a9eOgQMH2n8RBbD5Pbi6F9wKQeefwKISAlRISQatXLmSgIAAAgICaNy4sa13tjvdVHt4eLBx40ZKly5N7969qVq1Ks8++yyxsbHpXhkYO3Ysffr0YcCAAdSrV48TJ06watWqFF8oLVq0wGq1piiaWrduTVJSkm3fOcVisbB8+XJatmzJ008/TaVKlXjsscc4c+ZMjt4E+sorr7Bw4UI2bdpEgwYNqFGjBrNmzeK7775L0R38HX379mX//v0EBwenuNqVnflHjBjB66+/zgcffEDVqlXp27ev7V4yDw8PVq1axY0bN2jYsCEPP/ww7dq1Y+LEiff3QvyLg4MDCxcuJCYmhkaNGjFo0KAUXbDfyZLZ8zAtH330EadPn6Z8+fJ6boeIiNyXO73xQfIPdUFBQQwaNIjmzZvbOu2ya6dXw57/ayUTNA28it99+XzEYvzz3c2HIiIi8PX1JTw8/L6aAmUXq9XK1atXKVKkyAP7cF26dInvv/+e559/Ps3e4sR+GYZha1Kn5x3ZJ3v7fJnxHSO5m84ZySydM//f+fPnWbZsGY888ggFCxY0O07mRV+FGbUg+grU/g+0/1+O7MaezpnM1Ab5++wWEREREclmSUlJrF+/nmnTpnH58mXWrl1rdqTMMwxY9UxyEVWoGrRK3Somv1NnEyIiIiIi2eTatWssXLiQixcvAlCzZk26dOlicqos2Pc/OLkMHF2h62xwTr+zrPxKhZSIiIiIyH0yDINdu3axZs0aEhMTcXNzo2vXrtSoUcPsaJkX9jdseCN5uNUX4F/T3Dx2SoWUiIiIiMh92rdvn+05keXKlaNnz552cf99piXEJHd1nhQH5bpCnRfMTmS3VEiJiIiIiNynWrVqsXfvXqpXr06jRo1ybydQG9+E6wfBsxgE/Qi59TgeABVSIiIiIiKZFBsby7Zt22jZsiWOjo44Ojry9NNP594CCuDE78n3RgF0mgEeegTI3aiQEpuwsDCzI0gmGYZBdHQ0Hh4eufuLOw/T50pEJO85deoUixYtIiIiAsMwaNu2LUDu/rf49sXkXvoA6r8OgR3NzZMLqJASPDw8cHZ2ZsGCBWZHkUwyDIPY2Fjc3Nxy95d3Hufs7IyHh4fZMURE5D4lJiaybt06tm/fDoCfnx8VK1Y0OVU2MKyw4kmIvQ5F6kLzT+69jqiQEvD19eXFF18kOjra7CiSSVarlWvXrlG4cGHTH2An6fPw8MDX19fsGCIich8uX77MggULbC0N6tWrR1BQEC4uLiYnywa7v4Sz68DJA7r+Ck6uZifKFVRICZBcTOkPvdzHarXi6OhoF08CFxERyav++usvFi9ejNVqxdPTkx49elCpUiWzY2WPy7th87vJw22+goKVzc2Ti6iQEhERERG5ixIlSuDo6EilSpXo1q0bnp6eZkfKHvG3YXl/sCZCxT5Q81mzE+UqKqRERERERP7BMAwuXbpE8eLFAShUqBDPP/88BQsWzFv3JK8fBjePg3cp6DhZXZ1nktoCiYiIiIj8n6ioKObOncvkyZM5ffq0bXqhQoXyVhF1dC4cmAZYoMsv4OZndqJcR1ekRERERESAY8eO8fvvvxMVFYWDgwPXr18nMDDQ7FjZL+IMrBmcPNzkPSjZ0tw8uZQKKRERERHJ1+Lj41m9ejV79uwBwN/fn969e1OsWDGTk+UAayIsexziwiGgCTT5wOxEuZYKKRERERHJt86fP8/ChQu5ceMGAE2aNKFdu3Y4OeXRP5N3fAoXt4CLN3SZCY7OZifKtfLoGSIiIiIicm/Xrl3jxo0b+Pj40KtXL8qWLWt2pJxzYQtsG5U83P47KFDO3Dy5nAopEREREclXkpKScHR0BKB27drExcVRu3Zt3NzcTE6Wg+LCYfnjYFih6hNQ9XGzE+V66rVPRERERPIFwzDYuXMn3377LTExMQBYLBYaN26ct4sow4A1Q5I7mfAtB+3+Z3aiPEFXpEREREQkz4uIiOD3338nNDQUgD179tC8eXOTUz0gh36Go7PB4ghdZ4Grj9mJ8gQVUiIiIiKSpx08eJClS5cSGxuLk5MTHTp0oGHDhmbHejBunoB1LyQPN/sIAhqbmycPUSElIiIiInlSbGwsy5cv5++//wagePHiBAcHU7hwYZOTPSBJCbC8PyTchpKtoOHbZifKU1RIiYiIiEie9Mcff/D3339jsVho0aIFLVu2tHUykS9s/QAu7wI3P+j8Mzjko2N/AFRIiYiIiEie1KZNG8LCwmjXrh0lS5Y0O86DdfYP2PlZ8nCHyeBTytw8eZB67RMRERGRPOHy5cusXr0awzAAcHd356mnnsp/RVTMdVgxADCg5nNQqY/ZifIkXZESERERkVzNarWydetW1q9fj9VqpWjRotSuXdvsWOYwDFg9CG5fBL/K0Ga82YnyLBVSIiIiIpJr3bx5k0WLFnH27FkAqlSpQoUKFUxOZaK/foATi8DBGbr+Cs6eZifKs1RIiYiIiEiuYxgG+/btY+XKlcTHx+Pi4kLnzp2pXbs2FovF7HjmuH4IQl5NHm4xForWNTdPHqdCSkRERERynRUrVrBr1y4ASpcuTa9evfDz8zM5lYkSY2FZf0iMgcAgqP+K2YnyPBVSIiIiIpLrVK9enb1799K6dWuaNm2Kg0M+70Nt03AI2w/u/tBpOljy+evxAKiQEhERERG7Fx8fz8WLFwkMDASgTJkyDBs2DC8vL3OD2YNTK+DPCcnDnaaDZzEz0+QbKlVFRERExK6dO3eOSZMmMWvWLK5fv26briIKiLoCKwcmD9d9Gcp1MTVOfqIrUiIiIiJil5KSktiwYQObN2/GMAx8fHyIiYkxO5b9MKzJRVT0VShcE1p+ZnaifEWFlIiIiIjYnbCwMBYuXMilS5cAqFWrFp07d8bNzc3kZHbkz6/h9Epwckvu6txJr82DpEJKREREROzKzp07WbNmDYmJibi7u9O1a1eqV69udiz7cnUfbHo7ebjVOCis1+dBUyElIiIiInYlKiqKxMREypcvT8+ePfH29jY7kn1JiIZl/SApHsr3hNpDzE6UL6mQEhERERHTxcXF4erqCkDLli0pXLgwNWrUyL8P172bkNfgxhHwKg4dp4BeI1Oo1z4RERERMU1MTAwLFixgxowZJCUlAeDo6EjNmjVVRKXl+EL463vAAp1+Ao/CZifKt3RFSkRERERMcfLkSRYvXkxERAQWi4UzZ85Qrlw5s2PZr8jzsHpQ8nDDt6BMO3Pz5HMqpERERETkgUpISGDdunXs2LEDgIIFCxIcHEzJkiVNTmbHrEmw4kmIvQFFG0Czj8xOlO+pkBIRERGRB+bSpUssWLCAa9euAVC/fn06duyIi4uLycns3K7P4dx6cPaErrPAUa+X2VRIiYiIiMgDYRgGq1at4tq1a3h6etKzZ08qVqxodiz7d2kHbBmRPNx2IvjpNbMHKqRERERE5IGwWCz06NGDkJAQOnXqhIeHh9mR7F98JCzrD0YSVO4L1Z8yO5H8H7vrte9///sfgYGBuLm50bhxY3bu3HnX5SdMmEDlypVxd3enVKlSvPrqq8TGxj6gtCIiIiKSHsMw+PPPP1m/fr1tWsGCBendu7eKqIxa9yKEnwSfMtB+kro6tyN2dUVqzpw5vPbaa0yaNInGjRszYcIEgoKCOHr0KEWKFEm1/KxZs3jnnXeYNm0aDz30EMeOHWPgwIFYLBbGjRtnwhGIiIiICCQ/VHfZsmUcPXoUgEqVKlGiRAmTU+Uyh2fBoZ/A4gBdZoJbAbMTyT/Y1RWpcePG8dxzz/H0009TrVo1Jk2ahIeHB9OmTUtz+a1bt9KsWTP69+9PYGAgHTt2pF+/fve8iiUiIiIiOef06dN8//33HD16FEdHRzp06EBAQIDZsXKX8FOwdmjycJMPoEQzc/NIKnZzRSo+Pp49e/YwfPhw2zQHBwfat2/Ptm3b0lznoYce4pdffmHnzp00atSIkydPsnz5cgYMGJDufuLi4oiLi7ONR0REAGC1WrFardl0NFlntVoxDMMusoj90/kimaVzRjJL54xkRlxcHKtXr2bfvn0AFClShF69elG0aFEAnUcZZU3EsuxxLPERGMWbYTQaDnn4tbOn75nMZLCbQuratWskJSXZPmh3FC1alCNHjqS5Tv/+/bl27RrNmzfHMAwSExMZMmQI7777brr7GTNmDKNGjUo1PSwszC7urbJarYSHh2MYBg4OdnXBUOyQzhfJLJ0zklk6ZySjDMNI0a15rVq1aNiwIRaLhatXr5qcLnfx+uu/eF3ahtXZh2sNx2O9dsPsSDnKnr5nIiMjM7ys3RRSWRESEsKnn37Kt99+S+PGjTlx4gTDhg1j9OjRjBgxIs11hg8fzmuvvWYbj4iIoFSpUvj7++Pj4/OgoqfLarVisVjw9/c3/UQS+6fzRTJL54xkls4ZyYyHHnqIkJAQWrZsSe3atXXOZMX5TVgOTEge7vAdhcvVNzXOg2BP3zNubm4ZXtZuCqnChQvj6OjIlStXUky/cuUKxYoVS3OdESNGMGDAAAYNGgRAzZo1iYqKYvDgwbz33ntpvhGurq64urqmmu7g4GD6G3eHxWKxqzxi33S+SGbpnJHM0jkj6QkLCyMmJobSpUsDUKdOHapWrcqtW7d0zmRF7E1YOQAMK1QfiEPV/mYnemDs5XsmM/u3m7PbxcWF+vXrs27dOts0q9XKunXraNq0aZrrREdHpzpYR0dHIPnysoiIiIhkP8Mw2LFjBz/88AO//fYb0dHRQPIfwy4uLiany6UMA9YMhshzUKACtP3a7ERyD3ZzRQrgtdde46mnnqJBgwY0atSICRMmEBUVxdNPPw3Ak08+SYkSJRgzZgwA3bt3Z9y4cdStW9fWtG/EiBF0797dVlCJiIiISPaJiIhg8eLFnDx5EoBixYrZRScBud6BH+HYPHBwgq6zwMXb7ERyD3ZVSPXt25ewsDA++OADLl++TJ06dVi5cqWtA4qzZ8+muAL1/vvvY7FYeP/997lw4QL+/v50796dTz75xKxDEBEREcmzDhw4wLJly4iNjcXJyYmOHTvSoEEDLHpI7P25cRT+eCl5uNnHUKyhuXkkQyxGPm8DFxERga+vL+Hh4XbT2cTVq1cpUqSI6W1Exf7pfJHM0jkjmaVzRgCSkpJYtGgRBw4cAKB48eIEBwdTuHDhVMvqnMmkpHiY1RSu/gml28LDa5IfwJuP2NM5k5nawK6uSImIiIiI/XF0dMRisWCxWGjZsiUtWrTQbRTZZfP7yUWUWyHo9FO+K6JyMxVSIiIiIpJKQkICiYmJuLu7A9ClSxcaN25MiRIlTE6Wh5xZC7v/mzwcNBW89drmJip5RURERCSFS5cuMXnyZBYvXmzrCdnNzU1FVHaKDoMVTyYP1x4KFXqam0cyTVekRERERARIvldly5YthISEYLVaiYmJITIy0i7uI89TDANWPQtRl6BQNWj1hdmJJAtUSImIiIgIN2/eZOHChZw7dw6AqlWr0q1bNzw8PExOlgft+xZOLgFHV+j6KzjrNc6NVEiJiIiI5GOGYbB3715WrlxJQkICrq6udO7cmVq1aqlb85xw7QBseD15uOXn4F/L3DySZSqkRERERPKxhIQENm3aREJCAmXKlKFXr14UKFDA7Fh5U0IMLOsHSXFQtgvUfcnsRHIfVEiJiIiI5GMuLi4EBwdz/vx5mjRpYvpzfPK0jW8lX5HyKAqdfgRd8cvVVEiJiIiI5CNxcXGsWrWKgIAAGjZsCEDp0qUpXbq0ycnyuNClsG9i8nDnGeBRxNw8ct9USImIiIjkE2fPnmXRokXcvHmTgwcPUqNGDdtzoiQH3b4Eq55OHq7/GgQGmZtHsoUKKREREZE8LikpiZCQELZs2YJhGPj6+hIcHKwi6kEwrLDyKYi5Bv51oPmnZieSbKJCSkRERCQPCwsLY8GCBVy+fBmA2rVr07lzZ1xdXU1Olk/sGQ9n1oCTe3JX50563fMKFVIiIiIieVR0dDRTpkwhPj4ed3d3unXrRrVq1cyOlX9c2QObhicPt/kKClUxN49kKxVSIiIiInmUh4cHTZo04eLFi/To0QNvb2+zI+Uf8beTuzq3JkDF3lBzkNmJJJupkBIRERHJQ/7++2+KFi1KkSLJvcK1atUKi8Wih+s+aOtfgZvHwasEdJisrs7zIBVSIiIiInlATEwMy5cv58CBAxQrVoxBgwbh6Oio50KZ4ehvcGAqYIEuv4B7QbMTSQ5QISUiIiKSy4WGhrJ48WIiIyOxWCxUrlxZV6DMEnEW1gxOHm78LpRqbWocyTkqpERERERyqYSEBNauXcvOnTsBKFSoEMHBwZQoUcLkZPmUNQmWPwFxtyCgMTT90OxEkoNUSImIiIjkQhEREfz8889cu3YNgIYNG9KhQwecnZ1NTpaP7RwDFzaBizd0mQWOei/yMhVSIiIiIrmQl5cX7u7ueHl50bNnTypUqGB2pPzt4jbYOjJ5uN23UKCcqXEk56mQEhEREcklbt68ibe3N05OTjg4ONCnTx+cnZ3x8PAwO1r+FhcOy/qDkQRVH4dqT5idSB4AdeMiIiIiYucMw2DPnj189913rFu3zjbd19dXRZTZDAPW/gciToNvWWj3P7MTyQOiK1IiIiIiduz27dssWbKEY8eOAXD58mWSkpJwdHQ0OZkAcPgXODILLI7QZSa4+pqdSB4QFVIiIiIidurIkSMsWbKE6OhoHB0dadeuHU2aNFHX5vbiVmjy1SiAh0ZC8aamxpEHS4WUiIiIiJ2Ji4tj1apV7N27F4CiRYsSHBxM0aJFTU4mNkkJyfdFJdyGki2h0XCzE8kDluVC6vbt2xw5coRr165hsVgoXLgwlSpVwtvbOzvziYiIiOQ7sbGxHDp0CICHHnqINm3a4OSk37/tyraRcHknuBaAzr+Ag5pa5jeZ+kSeOnWKGTNmsHjxYg4cOIDVak0x38HBgerVq9OrVy+efPJJypVTt48iIiIiGWG1WnFwSO4HzNfXl549e+Lh4UGZMmVMTiapnAuBHWOShztOBp9SZqYRk2SokDp06BAffPABCxcupECBArRu3ZpHHnmEcuXK4efnh2EY3Lx5k1OnTrFnzx4mTpzI6NGjCQ4OZvTo0VStWjWnj0NEREQk17p69SoLFy6kbdu2VKxYEUB/P9mrmBuw/AnAgJqDoNLDZicSk2SokKpduzZdu3Zl2bJltG/f/p6XlhMTE1m7di2TJk2idu3axMfHZ0tYERERkbzEMAy2b9/OunXrSEpKYu3atVSoUEGdSdgrw4A1z8HtC+BXCdpMMDuRmChDhdRff/2VqV9FnJyc6NSpE506deLIkSNZDiciIiKSV4WHh7N48WJOnToFQIUKFejRo4eKKHv29xQ4vgAcnKHrr+DsaXYiMVGGCqn7ubRcpUqVLK8rIiIiktcYhsGBAwdYtmwZcXFxODs707FjR+rXr68iyp5dPwzrhyUPtxgDReuZm0dMp+5fRERERB6g8+fPs2DBAgBKlChBcHAwhQoVMjmV3FViHCzrB4kxUKYD1H/V7ERiBzJUSLVt2zbTG7ZYLKxbty7T64mIiIjkZaVKlaJOnToUKFCAFi1a2HrqEzu2eTiE7Qf3wtBpBlj0nkkGCymr1ZrpS82GYWQpkIiIiEhekpCQwIYNG2jSpAleXl4AuhcqNzm1EvaMTx7uNB28AkyNI/YjQ4VUSEhIDscQERERyXsuXrzIwoULuXbtGteuXeOxxx4DUBGVW0RdgZVPJQ/XfQnKdTU3j9gV3SMlIiIiks2sViubN29mw4YNWK1WvLy8aNCggdmxJDMMA1Y9DdFXoXANaPm52YnEztx3IRUZGUl4eDhWqzXVvNKlS9/v5kVERERylRs3brBw4ULOnz8PQLVq1ejatSseHh4mJ5NM2fsNnFoBTm7JXZ07uZmdSOxMlgup7777jnHjxnHy5Ml0l0lKSsrq5kVERERyndOnTzNr1iwSEhJwdXWlS5cu1KxZU035cpuwv2Djm8nDrb5MviIl8i9Z6nJk0qRJvPDCC1SoUIGPP/4YwzB45ZVXeOeddyhWrBi1a9dm6tSp2Z1VRERExK4FBATg6elJYGAgQ4cOpVatWiqicpuEaFj6GCTFQ7nuUHuo2YnETmWpkPrmm28ICgpixYoVDB48GICuXbvyySefcOjQISIjI7l+/Xq2BhURERGxR2fOnLH1Vuzq6srAgQN58skn8fX1NTmZZMmG1+HGYfAMgKBpoEJY0pGlQio0NJTu3bsD4OzsDEB8fDwAvr6+DBo0iG+//TabIoqIiIjYn7i4OBYvXsz06dPZtWuXbbqvr6+uQuVWxxfB/kmABTr/BB6FzU4kdixL90j5+vqSmJgIgI+PDx4eHpw7d84239vbm8uXL2dPQhERERE7c+bMGRYtWsStW7cAiI6ONjeQ3L/IC7D62eThhm9Cmfbm5hG7l6VCqkaNGuzfv9823qRJE7777ju6dOmC1Wrl+++/p1KlStkWUkRERMQeJCUlsX79erZs2QJAgQIF6NWrF2XKlDE5mdwXaxKsGACxN6BofWg22uxEkgtkqZB64oknmDRpEnFxcbi6ujJq1Cjat29v6+7c2dmZ+fPnZ2tQERERETOFhYUxf/58rly5AkCdOnXo1KkTrq6uJieT+7b7Czi3Hpw9ocsscHQxO5HkAlkqpJ5++mmefvpp23izZs04ePAgS5YswdHRkY4dO+qKlIiIiOQp8fHxXL16FQ8PD7p160bVqlXNjiTZ4fIu2PJ+8nDbb6Cg/oaVjLnvB/LeUa5cOYYNG5ZdmxMRERExXUJCgq1jrRIlStC7d28CAwPx8vIyOZlki/hIWNYfrIlQ6VGoPtDsRJKLZKnXPhEREZG8zDAM/vrrL7766iuuXr1qm16jRg0VUXnJHy/DrRPgXRo6TFJX55IpWboi5eDgkKFuPZOSkrKyeRERERHTxMTEsHTpUg4dOgTAtm3b6Nmzp8mpJNsdmQ0Hp4PFAbr8Am5+ZieSXCZLhdQHH3yQqpBKSkri9OnTLFq0iMqVK9OtW7dsCSgiIiLyoJw4cYLFixdz+/ZtLBYLrVq1okWLFmbHkuwWfhrWPJ883Ph9KKn3WDIvS4XUyJEj05136dIlmjRpos4mREREJNdISEhgzZo1tgfrFipUiODgYEqUKGFyMrkvEWch5lrKadZEWP0cxEdAkfrQdIQ52STXy7bOJu4ICAhgyJAhjB49mn79+mX35kVERESy3b59+2xFVMOGDenQoYOtkwnJpSLOwrTKkBSb/jLXD8Dti+BT+sHlkjwj2wspAE9PT06dOpUTmxYRERHJdvXr1+fMmTPUqVOHChUqmB1HskPMtbsXUQBJccnLqZCSLMj2XvsOHDjA119/raZ9IiIiYreuX7/OwoULSUhIAJI70nr44YdVRIlIhmXpilTZsmXT7LXv1q1bhIeH4+HhwaJFi+43m4iIiEi2MgyDPXv2sHr1ahISEvDy8qJDhw5mxxKRXChLhVSrVq1SFVIWiwU/Pz/Kly/PY489RsGCBbMloIiIiEh2uH37Nr///jvHjx8HIDAwkEaNGpmcSkRyqywVUtOnT8/mGCIiIiI55/DhwyxZsoSYmBgcHR1p164dTZo0ydBzMUVE0pKle6SeeeYZduzYke78nTt38swzz2Q5lIiIiEh22b59O3PnziUmJoaiRYsyePBgmjZtqiJKRO5Llgqp6dOnExoamu78U6dOMWPGjCyHEhEREckuVatWxd3dnWbNmjFo0CCKFClidiR5EKyJZieQPC5Huj+/ePEi7u7uObFpERERkbtKTEzk+PHjVK1aFQBfX19eeukl/W2S3/w95d7LOLqBe+GczyJ5UoYLqcWLF7N48WLb+A8//MDatWtTLXfr1i3Wrl1Lw4YNsyehiIiISAZdvXqVBQsWcOXKFfr370/FihUBVETlN0d/g78nJw+3/C+Ubpv2cu6F9QwpybIMF1KHDh3it99+A5J76NuxYwd79uxJsYzFYsHT05OWLVsybty47E0qIiIikg7DMNi+fTvr1q0jKSkJDw8PDMMwO5aY4fohWPV08nDDt6HhG+bmkTwrw4XU8OHDGT58OJD80LqpU6fSv3//HAsmIiIikhHh4eEsWrSI06dPA1CxYkV69OiBl5eXucHkwYuLgMW9ISEq+SpU84/NTiR5WJbukbJardmdQ0RERCTTDh48yJIlS4iLi8PZ2ZmgoCDq1aunHvnyI8NIvhJ18yh4lYSuv4JDjnQHIAJksde+P//8k2+//Tbd+d9++y379u3LaiYRERGRDLFYLMTFxVGyZEmGDBlC/fr1VUTlV7v+C8cXgKML9JgPHuqdUXJWlgqp9957L82OJu74448/eP/997McSkRERCQ90dHRtuFq1arRt29fnn76aQoWLGhiKjHV2T9gc/ItKLT5GgIamZtH8oUsFVJ79uyhRYsW6c5v0aIFu3fvznIoERERkX9LSEhg+fLl/O9//+P27du26VWqVMHBIUt/0kheEHEOlj4GhhWqD4Rag81OJPlElhqORkZG4uSU/qoODg6Eh4dnOZSIiIjIP124cIGFCxdy/fp1AI4dO0a9evVMTiWmS4yDJQ9DTBgUqQvtvgU17ZQHJEs/31SsWJHVq1enO3/lypWUK1cuy6FEREREILmDqw0bNjB16lSuX7+Ot7c3TzzxhIooSRbyClzeCW5+yfdFOet5YfLgZKmQevbZZ1m2bBmvvfYat27dsk2/desWr776KitXruTZZ5/NrowiIiKSD12/fp1p06YREhKCYRhUr16doUOHUr58ebOjiT04MB32TwIs0GUW+JY1O5HkM1lq2vfyyy+zb98+JkyYwNdff03x4sUBuHjxIlarlQEDBvDqq69ma1ARERHJX7Zv386FCxdwdXWla9eu1KhRQz3ySbIre2Hd0OThh0ZB2U7m5pF8KUuFlMVi4ccff+TJJ59k/vz5nDx5EoCePXvSp08fWrdunZ0ZRUREJB/q0KEDiYmJtG7dGl9fX7PjiL2IuQG/94bEWCjXFZq8Z3Yiyafu6yllbdq0oU2bNqmmJyUlsWLFCrp163Y/mxcREZF85PDhwxw6dIjevXtjsVhwcXGhZ8+eZscSe2JYYfnjEHEafMtB55/Boh4bxRzZ+rjnrVu3MnPmTH777TeuX79OUlJSdm5eRERE8qC4uDhWrlzJvn37AKhQoQK1a9c2N5TYp20fwemV4OQOPRYkdzIhYpL7LqQOHz7MzJkzmTVrFmfOnMHT05OgoCC6d++eHflEREQkDztz5gwLFy4kPDwci8VCs2bNqFGjhtmxxB6dXAbbRiUPd/gBiqjYFnNlqZC6ePEiv/76KzNnzmT//v24u7sTExPDxx9/zOuvv46Li0t25xQREZE8JDExkfXr17N161YAChQoQHBwMKVLlzY5mdilW6Gw/Ink4TovQLUnzM0jQiYKqYiICObNm8fMmTPZuHEj7u7u9OjRg9GjR1OuXDmqV69O5cqVVUSJiIjIPS1atIiDBw8CULduXYKCgnB1dTU5ldilhGj4vQ/E3YKAptB6nNmJRIBMFFLFihUDoEuXLsyaNYvu3bvj5uYGQGhoaM6kExERkTzpoYce4uzZs3Tt2pXKlSubHUfslWHA2iEQth88ikD338BRP9qLfchwNyexsbH4+flRtmxZypUrZyuiRERERO7l1q1btitQAMWLF+fll19WESV3t/87OPQzWByh21zwLmF2IhGbDF+ROnToEL/88guzZs1i3LhxlC9fnn79+tGvXz+cnZ1zMqOIiIjkUoZh8Ndff7FixQoSExMpXLgwRYsWBcDJKVs7D5a85uI2WP9K8nDLz6FUK1PjiPxbhq9IValShY8//piTJ0+yYcMG2rVrx7fffkv16tUJCgrCYrFw/fr1nMwqIiIiuUh0dDTz5s1j0aJFxMXFERAQoB9fJWOirsCSh8GaAJUegfqvmp1IJJUsPcGsefPmfPfdd1y6dImFCxdSr149XF1dGTJkCBUrVuSNN94gJCQkm6OKiIhIbnHixAm+++47Dh06hIODA23atOHpp5+mYMGCZkcTe2dNhKV94fZFKFgVgqaCxWJ2KpFU7uuaupOTEz169KBHjx5ERkYyf/58Zs6cyYQJExg/frweyCsiIpIPrVq1iu3btwNQuHBhgoODKV68uMmpJNfYNBzObwAX7+SH7rp4m51IJE3Z1jjZ29ubgQMHMnDgQC5evMicOXOya9MiIiKSi3h5eQHQqFEj2rdvr+Z8knFHf4PdXyQPd5oOhaqYGkfkbnLkLs/ixYvz6qtqyyoiIpIfWK1WIiMj8fX1BaBp06aUKVOGkiVLmpxMcpXrh2HV08nDDd+Cir3NzSNyDxm6RyooKIiNGzdmeuPr168nKCgo0+uJiIhI7nD9+nWmTZvGL7/8QkJCAgAODg4qoiRz4iJgcTAkREGpNtD8E7MTidxThq5IlS9fng4dOlCuXDn69u1Lu3btqFu3ru3S/R2RkZHs2bOHtWvX8ttvv3HmzBmeffbZHAkuIiIi5jEMgz179rB69WoSEhJwc3Pj6tWrlCih5/xIJhkGrHoGbh4Fr5LQbTY4qGt8sX8ZOku//fZb3nzzTb766iu+/fZbRo8ejcVioWDBgvj5+WEYBjdv3uTmzZsYhkHBggV5/PHHGTZsGGXLls3pYxAREZEHKDIykt9//50TJ04AULZsWXr27Glr2ieSKbu/gOPzwcEZeswDjyJmJxLJkAyX+2XLlmXChAl88cUXbNq0iW3btnHkyBHbs6MKFSpElSpVaNq0Kc2bN8/yjaX/+9//+O9//8vly5epXbs233zzDY0aNUp3+Vu3bvHee++xYMECbty4QZkyZZgwYQJdunTJ0v5FREQkfYcPH2bJkiXExMTg6OhI+/btady4MRZ1Ty1ZcfYP2PRO8nDbryGgsbl5RDIh09dNnZycaNOmDW3atMn2MHPmzOG1115j0qRJNG7cmAkTJhAUFMTRo0cpUiT1rxPx8fF06NCBIkWKMG/ePEqUKMGZM2coUKBAtmcTERHJ7wzDYPfu3cTExFCsWDF69+6Nv7+/2bEkt4o4B0sfA8MK1QdCrefNTiSSKXbVAHXcuHE899xzPP10co8tkyZNYtmyZUybNo133nkn1fLTpk3jxo0bbN261XYFLDAw8EFGFhERyfMMwwDAYrHQs2dP/vzzT1q0aIGjo6PJySTXSoyDpY9ATBj414F23+qhu5Lr2E0hFR8fz549exg+fLhtmoODA+3bt2fbtm1prvP777/TtGlTXnjhBRYvXoy/vz/9+/fn7bffTvfLPS4ujri4ONt4REQEkNx1q9VqzcYjyhqr1YphGHaRReyfzhfJLJ0zkhmJiYmsX7+e8PBwevdO7oray8uLli1bAug8kjRl5HvGsv4VLJd2YLj6YXSfB46uoPMp37Knf5syk8FuCqlr166RlJRE0aJFU0wvWrQoR44cSXOdkydP8scff/D444+zfPlyTpw4wX/+8x8SEhL48MMP01xnzJgxjBo1KtX0sLAwYmNj7/9A7pPVaiU8PBzDMHBwyFDv9JKP6XyRzNI5Ixl1/fp1/vjjD27cuAEk3xulZnySEff6nnE7OZcCf03CwMLNpt8QH+cJV6+akFTshT392xQZGZnhZe2mkMoKq9VKkSJF+OGHH3B0dKR+/fpcuHCB//73v+kWUsOHD+e1116zjUdERFCqVCn8/f3x8fF5UNHTZbVasVgs+Pv7m34iif3T+SKZpXNG7sVqtbJ9+3ZCQkJISkrCw8OD5s2bU7VqVZ0zkiF3/Z65uhfLzrcBMJp+SIG6/UxIKPbGnv5tcnNzy/CydlNIFS5cGEdHR65cuZJi+pUrVyhWrFia6wQEBODs7JyiGV/VqlW5fPky8fHxuLi4pFrH1dUVV1fXVNMdHBxMf+PusFgsdpVH7JvOF8ksnTOSnlu3brFo0SLOnDkDQKVKlejatSvR0dE6ZyRT0vyeibmRfF9UUiyU64pD0xFg0Tklyezl36bM7N9uzl4XFxfq16/PunXrbNOsVivr1q2jadOmaa7TrFkzTpw4kaIt47FjxwgICEiziBIREZG0Wa1WfvrpJ86cOYOzszPdu3fnsccew8vLy+xokhcYVljxBISfAt9y0PlnFVGS62XpDD579iybN29OMW3//v08+eST9O3bl0WLFmUpzGuvvcbkyZOZMWMGhw8fZujQoURFRdl68XvyySdTdEYxdOhQbty4wbBhwzh27BjLli3j008/5YUXXsjS/kVERPIrBwcHOnToQKlSpRgyZAj16tXTs6Ek+2z7CE6tACc36LEA3PzMTiRy37LUtO/ll1/m9u3brF27FkhuftemTRvi4+Px9vZm3rx5/Pbbb7YefjKqb9++hIWF8cEHH3D58mXq1KnDypUrbR1QnD17NsXltlKlSrFq1SpeffVVatWqRYkSJRg2bBhvv/12Vg5LREQkXzl+/DiGYVCpUiUguXl8lSpVVEBJ9jq5DLb9X0dfHX6AIrXNzSOSTbJUSO3cuZNhw4bZxn/66SdiYmI4cOAAZcuWpVOnTnzxxReZLqQAXnzxRV588cU054WEhKSa1rRpU7Zv357p/YiIiORX8fHxrFmzht27d+Pu7s7QoUPx9vYGUBEl2etWKCx/Inm4zgtQbYC5eUSyUZaa9t24cYMiRYrYxpcuXUqrVq0oX748Dg4O9O7dO90uy0VERMQ858+f5/vvv2f37t0A1KpVK1O9VIlkWEI0/N4H4m5BQBNoPc7sRCLZKktXpPz9/W09+ty6dYvt27czduxY2/zExEQSExOzJ6GIiIjct6SkJDZu3MimTZswDANvb2969epFuXLlzI4meZFhYFk3FML2g0cR6D4PHNURmOQtWSqk2rdvz9dff42Pjw8hISFYrVZ69eplm3/o0CFKlSqVXRlFRETkPiQkJDB9+nQuXrwIQI0aNejSpQvu7u4mJ5O8yv34DCyHfwGLI3SbA94lzI4kku2yVEiNHTuWY8eO8cYbb+Di4sIXX3xB2bJlAYiLi2Pu3Ln0798/W4OKiIhI1jg7OxMQEMCNGzfo0qULNWvWNDuS5GUXt+Gz54Pk4ZafQanWpsYRySlZKqSKFi3Kli1bCA8Px93dPcUzm+48+0lXpERERMwTGRkJYOtEomPHjrRs2RIfHx8zY0leF3UFy9JHsVgTMCo+jKX+a2YnEskxWSqk7vD19U01zd3dndq11a2liIiIWQ4dOsTSpUspVqwYAwYMwGKx4OLioofVS86yJsKyx7BEXSTRpyIOHaeoF0jJ07L8SOmzZ88yZMgQKleujJ+fHxs3bgTg2rVrvPzyy+zduzfbQoqIiMi9xcbGsmjRIn777TdiYmKIjY0lOjra7FiSX2waDudCMJy9uNlyKrh4m51IJEdl6YrUoUOHaNGiBVarlcaNG3PixAlbL32FCxdm8+bNREVFMXXq1GwNKyIiImk7ffo0ixYtIjw8HIvFQvPmzWnVqhWOjo5mR5P84Ng82P0FAEbQNJJ8K5ocSCTnZamQeuuttyhQoADbt2/HYrGkeKYUQNeuXZkzZ062BBQREZH0JSYm8scff7Bt2zYA/Pz8CA4O1r3K8uBcPwwrn04ebvAmVOwDV6+am0nkAchSIbVx40Y++OAD/P39uX79eqr5pUuX5sKFC/cdTkRERO7OMAyOHz8OQL169QgKCtK9UPLgxEXA4mBIuA2l2kCLT81OJPLAZKmQslqteHh4pDs/LCwMV1fXLIcSERGR9FmtVgAcHBxwdnamd+/eREREULlyZZOTSb5iGLDqGbh5FLxKQLfZ4OAE/3d+iuR1Wepsol69eixbtizNeYmJicyePZsmTZrcVzARERFJ7datW/z0009s3brVNi0gIEBFlDx4u7+E4/PBwRm6zwOPIvdeRyQPyVIhNXz4cFauXMnQoUM5cOAAAFeuXGHt2rV07NiRw4cP884772RrUBERkfzMMAz27dvHd999x5kzZ9i6dStxcXFmx5L86ux62PR28nDbr6G4fkCX/CdLTfs6d+7M9OnTGTZsGD/88AMATzzxBIZh4OPjw08//UTLli2zNaiIiEh+FR0dzdKlSzl8+DAApUqVIjg4WM3oxRyR52FpXzCsUP0pqPW82YlETJHlB/IOGDCA3r17s2bNGo4fP47VaqV8+fIEBQXZnqIuIiIi9+f48eMsXryYqKgoHBwcaN26Nc2aNcPBIcuPghTJusQ4WPIwxISBfx1o9x3oobuST2WpkDIMA4vFgqenJ7169crmSCIiIgIQGRnJnDlzSEpKwt/fn+DgYAICAsyOJflZyKtwaQe4+UGP+eDsbnYiEdNkqZAqUaIEjzzyCI8++ijNmjXL7kwiIiICeHt7065dO8LDw2nXrh3Ozs5mR5L87OAM2P8dYIEuM6FAObMTiZgqS+0CWrVqxbRp02jZsiWlS5fmjTfeYOfOndmdTUREJF9JSkoiJCQkxbMYmzZtSqdOnVREibmu7oO1Q5KHm34IZTubGkfEHmSpkPr111+5evUqs2fPplGjRnz33Xc0bdqU8uXL8+6777Jv375sjikiIpK3Xbt2jWnTprFhwwYWLlxIYmKi2ZFEksXcgN97Q2IslO0CTUeYnUjELmT5TlV3d3ceeeQR5s2bx9WrV/nll1+oWbMm48ePp379+lSpUiU7c4qIiORJhmGwa9cuvv/+ey5evIibmxutW7fGySnL/UGJZB/DCiuegPBT4FsOuvwCFnV0IgL30WvfP3l6etKvXz+6d+/O9OnTee+99zh+/Hh2bFpERCTPioyMZPHixYSGhgJQrlw5evbsiY+Pj8nJRP7PttFwagU4uSV3LuHmZ3YiEbtx34VUdHQ0v//+O3PnzmXlypXExcVRvnx5Xn755ezIJyIikifduHGDKVOmEBMTg5OTE+3bt6dRo0ZY1JW02IuTy2HbqOTh9t9DkTqmxhGxN1kqpGJjY1m2bBlz5sxh+fLlREdHExgYyMsvv0zfvn2pW7duducUERHJU/z8/ChevDjR0dEEBwfj7+9vdiSR/+/WSVj+OGBA7f9A9SfNTiRid7JUSPn7+xMdHU3x4sUZPHgwffv2pXHjxtmdTUREJE85c+YMxYoVw9XVFYvFQp8+fXBxccHR0dHsaCL/X0J0cucScbcgoAm0GW92IhG7lKVCauDAgfTt25fmzZtndx4REZE8JzExkXXr1rF9+3bq1q1Ljx49gOSOm0TsimHA2qEQth88ikD338DRxexUInYpS4XUN998k905RERE8qTLly+zYMECwsLCALBYLFitVhwc1POZ2KH9k+DQT2BxhG5zwLuk2YlE7FaGCqmNGzcC0LJlyxTj93JneRERkfzGarWydetW1q9fj9VqxdPTkx49elCpUiWzo4mk7eJ2WD8sebjFWCjV2tQ4IvYuQ4VU69atsVgsxMTE4OLiYhtPj2EYWCwWkpKSsi2oiIhIbhEeHs6CBQs4e/YsAFWqVKFbt254enqanEwkHVFXYMnDYE2ASg9Dg9fNTiRi9zJUSK1fvx4AFxeXFOMiIiKSmqOjI9euXcPFxYVOnTpRp04ddWsu9suaCMseg9sXoGAVCJoGOl9F7ilDhVSrVq3uOi4iIpLfxcXF4erqCoCXlxePPPIIvr6++PnpAaZi5za9C+dCwNkLeiwAF2+zE4nkClm607Vt27asW7cu3fnr16+nbdu2WQ4lIiKSmxw7doxvvvmGgwcP2qYFBgaqiBL7d2we7P5v8nCnH6FQVXPziOQiWSqkQkJCuHLlSrrzr169yoYNG7IcSkREJDeIj49nyZIl/Prrr0RFRbFr1y4MwzA7lkjGXD8MK59OHm7wZvK9USKSYVnq/hy4a1vvEydO4O2ty8IiIpJ3nTt3joULF3Lz5k0AmjRpQrt27XQvlOQO8ZHJD91NuJ3cO1+LT81OJJLrZLiQmjFjBjNmzLCNf/zxx0yePDnVcrdu3eKvv/6iS5cu2ZNQRETEjiQlJbFhwwY2b96MYRj4+PjQq1cvypYta3Y0kYwxjOQrUTeOgFeJ5OdFOWT5t3WRfCvDn5ro6GjbwwQBIiMjUz1M0GKx4OnpyZAhQ/jggw+yL6WIiIidOHfuHJs2bQKgZs2adOnSBTc3N5NTiWTC7i/h+HxwcIbu88CjiNmJRHKlDBdSQ4cOZejQoQCULVuWr776ih49euRYMBEREXsUGBhIs2bNKFasGDVq1DA7jkjmnF0Pm95OHm7zFRRvYm4ekVwsS9dxT506ld05RERE7FJkZCQrV66kY8eO+Pr6AtC+fXuTU4lkQeR5WNoXDCtUexJqDzE7kUiulqFC6s6T2UuXLp1i/F7uLC8iIpIbHTx4kKVLlxIbG0tCQgL9+/c3O5JI1iTGwZKHISYM/OtA+0l66K7IfcpQIRUYGIjFYiEmJgYXFxfb+L0kJSXdd0AREZEHLTY2lhUrVvDXX38BULx4cTp27GhyKpH7EPIaXNoBrgWgx3xwdjc7kUiul6FCatq0aVgsFpydnVOMi4iI5DWnTp1i0aJFREREYLFYaNGiBS1btsTR0dHsaCJZc/An2P8tYIEuM6FAObMTieQJGSqkBg4ceNdxERGRvODo0aPMnj0bgIIFCxIcHEzJkiVNTiVyH67ug7XPJw83/RDK6fE0ItklWx8aEB8fT0JCAp6entm5WRERkQeifPny+Pv7U7p0aTp27IiLi4vZkUSyLvZm8kN3E2OhbBdoOsLsRCJ5isO9F0lt9uzZvPrqqymmjRo1Ci8vLwoUKEBwcDC3b9/OloAiIiI5xWq1sm/fPqxWKwBOTk4MGjSIbt26qYiS3M2wwvInIPwU+JaFzj+DJUt/9olIOrL0ifryyy+JioqyjW/dupVRo0YRFBTEq6++ysqVK/nkk0+yLaSIiEh2u3nzJjNmzGDx4sVs2bLFNl0FlOQJ20bDqeXg5AY9FoB7QbMTieQ5WWraFxoaylNPPWUbnzVrFsWKFWPhwoU4OTlhtVqZP38+Y8aMybagIiIi2cEwDPbt28fKlSuJj4/HxcUFHx8fs2OJZJ+Ty2HbqOTh9t9DkTqmxhHJq7JUSMXFxeHm5mYbX716NZ07d8bJKXlz1apV49tvv82ehCIiItkkKiqKpUuXcuTIESD5eYe9evXCz8/P5GQiWRBxFmKupZwWeR6WPw4YUPUJqP6kKdFE8oMsFVJly5Zl7dq1DBo0iN27d3PixIkUTfmuXLmCl5dXtoUUERG5X6dPn2bevHlERUXh4OBAmzZteOihh3Bw0H0jkgtFnIVplSEpNv1ljv0GzT8Bn9IPLpdIPpKlQur5559n2LBhHDp0iPPnz1OyZEm6detmm79lyxaqV6+ebSFFRETul4eHB7Gxsfj7+9O7d2+KFStmdiSRrIu5dvciCiApLnk5FVIiOSJLhdRLL72Em5sby5cvp379+rz99tu4uyc/IfvGjRtcvnyZIUOGZGtQERGRzIqMjMTb2xuAIkWK8MQTT1CyZElbU3QREZGsyvK/JM899xzPPfdcqukFCxZk9+7d9xVKRETkfiQlJbFhwwa2bt3KwIEDbQ/VDQwMNDeYiIjkGff9k9yhQ4c4c+YMAGXKlKFatWr3HUpERCSrrl27xoIFC7h06RIAR48etRVSIiIi2SXLhdTixYt57bXXOH36dIrpZcuWZdy4cfTo0eN+s4mIiGSYYRjs2rWLNWvWkJiYiLu7O127dtU9u5I33b5gdgKRfC9LhdTy5cvp06cPZcqU4dNPP6Vq1aoAHD58mB9++IHevXuzdOlSOnXqlK1hRURE0hIREcHvv/9OaGgoAOXLl6dnz562+6NE8pSTy/+vi3MRMVOWCqnRo0dTq1YtNm3ahKenp216jx49ePHFF2nevDmjRo1SISUiIg/EsWPHCA0NxcnJiQ4dOtCwYUMsFovZsUSylzUp+UG720ebnUREyGIh9ddff/Hpp5+mKKLu8PT0ZODAgbz77rv3HU5ERCQj6tevz40bN6hXrx6FCxc2O45I9ou+lnwV6szq5PGqA+DY3OQuztPj6Abu+jyI5JQsFVJubm7cuHEj3fk3btzAzc0ty6FERETu5tSpU2zYsIF+/frh6uqKxWKhY8eOZscSyRmXd8HvD0PkWXByhw4/QLUnoPnHyc+JSo97YT1DSiQHZamQatu2LV999RWdOnWiadOmKebt2LGDr7/+Wv+giYhItktMTGTdunVs374dgE2bNtG+fXuTU4nkEMOAvyfDHy9BUjwUqAA9FoB/zeT5PqVVKImYKEuF1Oeff07Tpk1p3rw5jRo1onLlykByF7M7d+6kSJEifPbZZ9kaVERE8rdLly6xcOFCwsLCgOTmfC1btjQ5lUgOSYiGdf+BgzOSxyv0gk7TwdXXzFQi8g9ZKqTKli3LX3/9xZgxY1ixYgVz5swBkp8jNWzYMN555x2KFCmSrUFFRCR/slqtbN26lfXr12O1WvH09KRHjx5UqlTJ7GgiOeNWKPzeB8L2g8UBmo+Bhm+COlARsSuZLqSSkpIICwujQIECjB8/nvHjx+dELhEREQBCQkLYtGkTAFWqVKFbt25pdnYkkieELoEVAyAuHDyKQNfZULqN2alEJA0OGV3QMAzeffdd/Pz8KFGiBD4+PgQHB9+10wkREZH71bhxY/z8/OjZsyePPvqoiijJm6xJsPk9WNQjuYgKaApP/KkiSsSOZfiK1PTp0xk7diwlS5akU6dOhIaGsnjxYqxWK4sXL87JjCIiko9ERUVx4MABGjduDCQ/VuPFF1/EwSHDv/2J5C7RYbCsP5xdmzxe9yVo9QU4upibS0TuKsOF1HfffUfdunXZvHkz7u7uAAwbNoz//e9/XLt2Tc/tEBGR+3b06FGWLFlCVFQUnp6e1KhRA0BFlORdl3Ykd21++zw4eUDHKVC1n9mpRCQDMvwvU2hoKE8++aStiAL4z3/+g9Vq5fjx4zkSTkRE8of4+HiWLFnC7NmziYqKokiRIvqBTvI2w4B938LsFslFlF8leHyniiiRXCTDV6Ru3ryJv79/iml3/pGLjY3N3lQiIpJvnDt3joULF3Lz5k0AmjZtStu2bXFyylLHsiL2LyEa1jwPh39JHq/YG4J+BFcfc3OJSKZk6l8pi7rdFBGRbLRt2zbWrFmDYRj4+PjQq1cvypYta3YskZxz83hy1+bX/gaLI7T8DOq/pq7NRXKhTBVS77zzDmPGjLGNJyUlATBo0KBUvShZLBb279+fDRFFRCSv8vf3xzAMatWqRefOnXFzczM7kkjOObEYVjwJ8RHgURS6zYFSrcxOJSJZlOFCqmXLlmlekdKDd0VEJKMMw+D69eu2puEVKlRg8ODBBAQEmJxMJAdZE2Hz+7Drs+Tx4s2g+1zwKm5uLhG5LxkupEJCQnIwhoiI5HUREREsXryYCxcuMGTIEAoUKACgIkrytqgrsKwfnFufPF7vFWj5OTg6mxpLRO6f7uQVEZEcd+DAAZYtW0ZsbCxOTk5cvnzZVkiJ5FkXt8GSh+H2RXD2hKBpUPlRs1OJSDZRISUiIjkmJiaGFStW8PfffwNQvHhxgoOD1bW55G2GAXsnwobXkpv1FawCPRZAoapmJxORbKRCSkREcsTJkydZvHgxERERWCwWWrZsSYsWLXB0dDQ7mkjOSYiC1c/BkV+Txys9AkFTwcXb3Fwiku1USImISI44fvw4ERERFCxYkODgYEqWLGl2JJGcdeMY/N4brh8EBydo+V+oN0xdm4vkUSqkREQk2xiGYevhtV27dri5udG0aVNcXFxMTiaSw44vgJUDIT4SPItBt9+gZHOzU4lIDnLIyEK9e/dm06ZNtvGNGzcSFhaWY6FERCR3sVqtbNq0iRkzZmC1WgFwcnKiVatWKqIkb7MmwoY3kx+yGx8JJVvCgL0qokTygQwVUosXL+bs2bO28TZt2rBmzZocCyUiIrnHzZs3mT59On/88Qdnzpzh8OHDZkcSeTCiLsNv7WH3F8njDd6Ah9cmX5ESkTwvQ037SpQowd69e3n88ceBlE03REQkfzIMg71797Jq1Sri4+NxcXGhc+fOVKtWzexoIjnvwhZY8ghEXUruSCLoR6jUx+xUIvIAZaiQeuyxx/jiiy+YO3eu7bkf77zzDmPGjEl3HYvFwv79+7MlpIiI2JeoqCiWLFnC0aNHAShdujTBwcF6NpTkfYYBf34FG99MbtZXqFpy1+YFK5udTEQesAwVUmPGjKFChQqsX7+eq1evYrFY8PT0pFChQjmdT0RE7NDChQsJDQ3FwcGBtm3b0rRpUxwcMtRaXCT3io+EVYPg2Nzk8cqPQcfJ4OJlbi4RMUWGCilHR0cGDx7M4MGDAXBwcOD999+nf//+ORpORETsU1BQEIsWLaJ79+4UK6b7QSQfuH44uUOJG4eTuzZvNQ7qvqiuzUXysSx1f37q1Cn8/f2zO4uIiNipc+fOcenSJRo1agSAv78/gwYN0v2ykj8c/Q1WPQMJt8GreHLX5iUeMjuViJgsS4VUmTJlgOSCasWKFZw5c8Y2vXPnzpQtWzb7EoqIiGmSkpIICQlhy5YtABQvXtz2YF0VUZLnJSXAprdhz/jk8VKtoets8CxqaiwRsQ9ZfiDv66+/zldffWV7XsgdDg4OvPLKK3zxxRf3HU5ERMwTFhbGggULuHz5MgC1a9emcOHCJqcSeUBuX4Klj8KFzcnjDd+G5h8nN+sTESGLhdSXX37J+PHjefjhh3n99depWrUqAIcPH2b8+PGMHz+eEiVK8Oqrr2ZrWBERyXmGYbBz507Wrl1LYmIi7u7udOvWTd2aS/5xfiMs7Zv8nCgXH+g0Ayr2MjuViNiZLBVSkydPpkePHsydOzfF9MaNGzN79mxiY2P5/vvvVUiJiOQyhmEwZ84cW7fmFSpUoEePHnh7e5ucTOQBMAzYMw42vg1GEhSuAd3nQ8FKZicTETuUpb5qT58+TVBQULrzg4KCOH36dFYziYiISSwWCxUqVMDJyYkuXbrQv39/FVGSP8RFJD9gd8MbyUVU1ceh/3YVUSKSrixdkSpSpMhdH7a7f/9+9eonIpJLxMTEEBERQdGiyTfQ169fnwoVKujhupJ/XDuY3LX5zaPg4AxtJkDtoeraXETuKktXpB555BGmTJnC2LFjiYqKsk2Pioris88+Y8qUKfTt2zfLof73v/8RGBiIm5sbjRs3ZufOnRlab/bs2VgsFnr16pXlfYuI5CcnT57ku+++Y/bs2cTFxQHJV6VUREm+cWQ2zGqcXER5lYS+G6HOf1REicg9ZemK1OjRo9m3bx/vvvsuH3zwAcWLFwfg4sWLJCYm0qZNGz766KMsBZozZw6vvfYakyZNonHjxkyYMIGgoCCOHj1KkSJF0l3v9OnTvPHGG7Ro0SJL+xURyU8SEhJYv349O3bsAKBgwYJERkbi6upqcjKRByQpHja8CXu/Th4v3Q66/goealEjIhmTpULKw8ODdevWsXjx4hTPkerUqRNdunShe/fuWX6+yLhx43juued4+umnAZg0aRLLli1j2rRpvPPOO2muk5SUxOOPP86oUaPYtGkTt27dytK+RUTyg7CwMObPn8+1a9cAaNCgAR06dMDFxcXkZCIPSOSF5K7NL25NHm/8Ljz0ETg4mptLRHKV+3oYQs+ePenZs2d2ZSE+Pp49e/YwfPhw2zQHBwfat2/Ptm3b0l3vo48+okiRIjz77LNs2rTprvuIi4uzNV8BiIiIAMBqtaZ6JpYZrFYrhmHYRRaxfzpfJDMMw2Dz5s1s2LABwzDw8vKie/fuVKhQAUDnkaQpz33PnFuPZXl/LNFXMVx9MYKmQ/keyfPyyjGaLM+dM5Lj7OmcyUwGu3qq3LVr10hKSrLd8HxH0aJFOXLkSJrrbN68malTp7Jv374M7WPMmDGMGjUq1fSwsDBiY2MznTm7Wa1WwsPDMQwDB4cs3cIm+YjOF8kMwzA4deoUhmFQtmxZWrZsiZubG1evXjU7mtixPPM9Yxh4HvoWr/2fYjGsJBSoxq2WU0jyLgv6DGSrPHPOyANjT+dMZGRkhpe1q0IqsyIjIxkwYACTJ0+mcOHCGVpn+PDhvPba/2vvvsOjKPc2jn83m04KJY2EQOi9CVJFunQJRQQ8NEURxaMHPXYBK+oLih4booIoiII06R0FKUqvkY6UJISWQPruvH+sWQhJIAlJNiT357pywcw8M/vbzZPN3nlmnhltX46NjSU0NBR/f398fHzyq9Rss1qtmEwm/P39Hd6RpPBTf5FbMQwDi8WCs7Pt7b5Xr17s3r2bZs2aYTbrNCa5tSLxPpN0GdPyYZiOLADAqDkIc/vPKOPi6eDCiqYi0WekQBWmPuPu7p7ttoUqSPn5+WE2m4mKikq3PioqiqCgoAztjxw5wvHjx+nRo4d9XdpwnLOzMxEREVSuXDndPm5ubpleTO3k5OTwb1wak8lUqOqRwk39RbJy5coVfvnlF1xdXenTpw8A3t7eVK1aFbPZrD4j2XZHv8+c2wO/9IGLh8DsCm0/xlTvsVxfyy3Zc0f3GXGIwtJncvL4hSpIubq60qhRI1avXm2fwtxqtbJ69WpGjRqVoX2NGjXYs2dPunWvvvoqcXFxfPTRR4SGhhZE2SIihU5ERAQLFy4kPj4es9lMTExMtkfuRYqMAzNgxWOQGg/eodBjDpRt4uiqRKSIKFRBCmD06NEMGTKExo0b06RJEyZNmsTVq1fts/gNHjyYkJAQxo8fj7u7O3Xq1Em3f9q9T25cLyJSHCQlJbF8+XJ27NgB2G6g3rt3b4UoKV4sybBuNOz81LZcoSN0nQme+jkQkbxT6ILUgw8+yLlz5xgzZgyRkZE0aNCAZcuW2SegOHnypMOH/ERECqOTJ08yf/58Ll68CECLFi1o27at/fookWIh7hT80hfO2u6RRrPXoPlYTW0uInku179dDxw4wNSpUzl69CgXL17EMIx0200mE6tXr87VsUeNGpXpqXwA69atu+m+06ZNy9VjiojcySwWCz///DOxsbH4+voSHh5OWFiYo8sSKVgnVsPi/pAQA24loev3UKmbo6sSkSIqV0Hqu+++Y9iwYbi4uFC9enVKlSqVoc2NwUpERPKP2Wzm/vvvZ8+ePXTu3DlHsw6J3PEMK2x9Dza+avu/fwO4/2coWcnRlYlIEZarIDVu3DgaNmzI0qVLdd69iIgDGIbBli1b8PT0pF69egBUrlw5w0ylIkVC7EnbKFNmkuNg81twcpVtufYwaP8puHgUXH0iUizlKkidOXOG5557TiFKRMQBYmNjmT9/PseOHcPV1ZWKFSvi7e3t6LJE8kfsSfimOlgSb97OydUWoOoNL5i6RKTYy1WQqlevHmfOnMnrWkRE5Bb27NnDkiVLSExMxMXFhY4dO+Ll5eXoskTyT0LMrUMUQKevoda/8r8eEZF/5CpIffDBBzzwwAN06dKFFi1a5HVNIiJyg4SEBJYsWcLevXsBCAkJoVevXpQpU8bBlYkUEmVqOboCESlmchWk3nvvPXx9fWnVqhW1atWifPnymM3ppxU1mUwsWLAgT4oUESnOkpOT+eKLL4iNjcVkMtG6dWtatWqlW0GIiIg4UK6C1O7duzGZTJQvX54rV66wf//+DG1MJtNtFyciIuDq6kqdOnWIiIigV69ehISEOLokERGRYi9XQer48eN5XIaIiFzvzJkzuLu7U7p0aQDatm1LmzZtcHFxcXBlIgUsOc7RFYiIZEq3uxcRKUSsVisbNmxg/fr1BAcHM2zYMJycnHB21tu1FENnNsOSgY6uQkQkU7f1m3n9+vUsXryYEydOAFChQgW6detG69at86Q4EZHi5MKFC8ybN49Tp04B4O3tTUpKCm5ubg6uTKSAWS3wx/uw8TUwLI6uRkQkU7kKUsnJyQwYMID58+djGAYlS5YE4NKlS0ycOJFevXrxww8/6BQUEZFsMAyD7du3s3z5cntw6tKlC/Xq1dP1plL8xJ2GpYPg77W25Urd4cRKsCRlvY/ZHTx0b0sRKVi5ClKvv/468+bN47nnnuPZZ58lMDAQgOjoaCZOnMj//d//8cYbb/Dmm2/mabEiIkVNYmIi8+bN46+//gJsI/vh4eH2P1CJFCtHfoFlwyDxPDh7QvtPoPZQiPvbdj+prHj4gU/5AitTRARyGaRmzpzJkCFDeP/999OtDwgI4L333iMqKorvvvtOQUpE5BZcXFyIi4vDbDbTrl07mjdvrlEoKX5SE2H9f2HnJ7blgIbQ7QcoXd227FNeQUlECp1cBamzZ8/StGnTLLc3bdqUWbNm5booEZGiLCkpCWdnZ8xmM2azmd69e2OxWOyj+yLFyvn9sKg/xOyxLTcaDfe8A866NlBECrdc3c2xXLlyrFu3Lsvt69evp1y5crmtSUSkyDp58iRffPEF69evt6/z8/NTiJLixzBg12T4vrEtRHkGQO+l0GaiQpSI3BFyFaSGDBnCTz/9xOOPP05ERAQWiwWr1UpERAQjR45k9uzZDB06NI9LFRG5c1ksFlatWsXUqVO5dOkSe/fuJSUlxdFliThGwgX4pS+sehxSE6DCfTB4F1Ts7OjKRESyLVen9r388sscOXKEL7/8kilTpuDkZMtjVqsVwzAYMmQIL7/8cp4WKiJyp4qOjmbevHlERkYC0KBBAzp37qyZTaV4OvUrLH4IrpwCJxfbaXyNR4MpV3/bFRFxmFwFKbPZzLRp0xg9ejRLlixJdx+prl27Uq9evTwtUkTkTmQYBps3b2b16tVYLBY8PDzo0aMHNWvWdHRpIgXPmgqb34LNb4JhhZJVoPssCGzk6MpERHLltm7IW69ePYUmEZEsxMbGsnbtWiwWC1WrVuX+++/Hy8vL0WWJFLzYE7ZRqDMbbcu1h0C7/4Grt2PrEhG5DbcVpEREJGu+vr507dqV1NRUGjVqpGnNpXiKmA0rH4Wky7bg1OELqDnQ0VWJiNy2bAUpJycnnJyciI+Px9XVFScnp1t+IDCZTKSmpuZJkSIid4KEhASWLFlCo0aNCAsLA2zXQ4kUSylXYe1/YM8U23JQE9u9oUpWcmxdIiJ5JFtBasyYMZhMJpydndMti4iIzZEjR1iwYAFxcXGcPn2aUaNG2SfiESl2onfB4v5w4SBggiYvQovXwawJVkSk6MhWkBo3btxNl0VEiquUlBRWrVrF1q1bAShTpgy9evVSiJLiyTBgx//g1/+CJRlKlIWu30P5do6uTEQkz+XqN/0bb7zB3r17s9y+b98+3njjjVwXJSJyJzhz5gxffvmlPUTdfffdjBgxgpCQEAdXJuIA8edgfg9Y+7QtRFXqDoN3K0SJSJGVqyA1btw4du/eneX2vXv38vrrr+e6KBGRwi4qKoqvv/6amJgYvLy8eOihh+jatavuDSXF04nVML0+HF0MZjfbjHzhC8HTz9GViYjkm3yZte/ChQu4urrmx6FFRAqFgIAAqlevjslkolu3bnh6ejq6JJGCZ0mBja/BH+8DBpSuabs3lL9ujSIiRV+2g9Svv/7KunXr7Mtz587l8OHDGdpdunSJH3/8kbp16+ZJgSIihYFhGOzcuZMaNWrg4eGByWSid+/emM1mTb4jxdOlI7B4AET+YVuu9xi0+RBc9EcFESkesh2k1q5daz9dz2QyMXfuXObOnZtp21q1avG///0vbyoUEXGwK1eusHDhQg4dOsSRI0fo06dPuplMRYqdAzNg1UhIjgO3knDfV1Ctj6OrEhEpUNn+FPD8888zatQoDMMgICCAL774gj590r9pmkwmPD09cXd3z/NCRUQc4cCBAyxatIj4+HjMZrMmkpDiLTkOVj8J+7+zLYe0ss3K51PesXWJiDhAtoOUh4cHHh4eABw7doyAgAD7sohIUZOUlMSyZcvYuXMnAIGBgfTu3ZuAgADHFibiKJF/wNJ/waXDYHKCZmOg2SvgpJFZESmecvXuZ7VaWbVqFT169Mh0+y+//ELdunUJCwu7ndpERBwiOjqaH374gUuXLgHQsmVL2rRpo1P5pHgyrHju/wzTrvFgTQXvUOg6E8rd4+jKREQcKlefCp577jliY2OzDFKffvopJUuWZNasWbdVnIiII3h7e2O1WilZsiTh4eFUqFDB0SWJOMaVs5iWDsbn5CrbctU+cN8UcC/l2LpERAqBXAWpTZs28cwzz2S5vX379kyaNCmXJYmIFLxLly7h6+uLyWTCw8ODgQMHUrJkSdzc3BxdmohjHF0Cy4ZiSjiHYXbHaDsJp3qPgWapFBEBcnlD3osXL+Lt7Z3ldi8vL86fP5/rokRECophGGzatIlPPvnEfj0U2K6JUoiSYik1Cdb+B+Z1g4RzGH71iOmyHOo+qhAlInKdXAWp8uXLs3Hjxiy3//bbb5QrVy7XRYmIFITLly8zffp0VqxYgcVi4ejRo44uScSxzh+Emc1g+yTbcsN/YwzYhMW3mkPLEhEpjHIVpAYMGMAPP/zAxx9/jNVqta+3WCx89NFH/PjjjwwcODDPihQRyUuGYbB7924+//xzjh8/jouLC927d6d3796OLk3EMQwD9nwN3zeCczvBvQyEL4R2H4GzbmkiIpKZXF0j9dJLL7FhwwaeeeYZ3n77bapXrw5AREQE586do02bNrzyyit5WqiISF5ISEhg0aJF7N+/H4By5crRq1cvSpcu7eDKRBwk8RKsHAF//WRbLt8eukwHr2CHliUiUtjlKki5ubmxYsUKvv32W+bOncuRI0cAaNKkCX369GHw4ME4OeVqsEtEJF+dO3eO/fv34+TkROvWrbnnnnv0fiXF1+nfYclAiD1hux9Uy7fg7v/a7hMlIiI3leubojg5OTFs2DCGDRuWl/WIiOQ5wzAw/XORfPny5encuTOhoaEEB+sv7lJMWS2w5R3Y9DoYFvCtCN1+gLJNHV2ZiMgdQ3eXFJEi7fTp0yxevJg+ffpQpkwZAJo21YdFKcbiTsGSf8Gp9bblGgOhw+fg5uPYukRE7jC5DlKRkZF8/fXXbN++ncuXL6ebdALAZDKxevXq2y5QRCQ3rFYrv/32G+vXr8cwDFauXEn//v0dXZaIYx2aBysegcSL4OIF7T+FWoM0rbmISC7kKkjt3r2bNm3akJCQQPXq1dmzZw+1atXi0qVLnD59msqVKxMaGprXtYqIZMv58+eZN28ep0+fBqBWrVp0797dwVWJOFBKAqwfDbu+sC0HNrKdyleqqmPrEhG5g+UqSL344ot4eXmxc+dOPD09CQgI4KOPPqJdu3bMnj2bkSNHMmPGjLyuVUTkpgzDYNu2baxYsYKUlBTc3Nzo2rUrdevWtV8jJVLsxOyFRf3h/D7bcuP/wj1vgdnVsXWJiNzhchWkNm7cyPPPP0/58uW5cOECgP3UvgceeIANGzbw3//+l/Xr1+ddpSIit7B7924WL14MQFhYGOHh4fj6+jq4KhEHMQzY9TmsfxZSE8Ez0Datedh9jq5MRKRIyFWQslqtBAYGAlCyZEnMZrM9UAHUrVuXr7/+Om8qFBHJpjp16rB9+3Zq1KhBs2bNNAolxVfCeVj+CBxZYFuu2AU6TwPPAIeWJSJSlOTqRhEVK1bk2LFjtgM4OVGxYkVWrVpl3/77779TsmTJPClQRCQriYmJrF27FovFAoDZbGbo0KE0b95cIUqKr7/XwfT6thDl5AJtPoReixSiRETyWK5GpO677z5mz57N22+/DcDIkSN59tlnOXr0KIZhsG7dOp599tk8LVRE5HonTpxg3rx59llD27dvD6AAJcWXJcV2X6gt7wAGlKoG3WZBYENHVyYiUiTlKki98sorDBgwgJSUFFxcXHjmmWe4evUqP//8M2azmddee42XX345r2sVESE1NZW1a9fy+++/A7bTi6tW1cxjUsxdPg6LB8LZTbblOg9D24/A1cuhZYmIFGU5DlKGYWA2m6lduzYuLi6A7S/Ar776Kq+++mqeFygikiYqKop58+YRFRUFQIMGDejcuTNubm4OrkzEgQ7OgpUjIDkWXH2g45dQ40FHVyUiUuTl+Bqp5ORkSpcuzccff5wf9YiIZGrv3r1MmTKFqKgoPD09efDBB+nZs6dClBRfyVdg2cOweIAtRJVtDoN3KUSJiBSQHI9Iubm5ERQUpA8vIlKggoODcXJyonLlyvTo0QMvL52yJMVY1HZbgLr4F2CCZq9A87HglKsz9kVEJBdyNWvf0KFDmT59OsnJyXldj4gIYDuN+MyZM/bl0qVL89hjj9G/f3+FKCm+DCts+xBmNrOFKK8Q6LcGWr6pECUiUsBy9a5bt25d5s+fT+3atRk6dChhYWF4eHhkaNe7d+/bLlBEip/4+HgWL17M/v37GTx4MBUrVgTAz8/PwZWJ5KPYk5AQk/V2w4CNr8LxZbblyj2h09fgUaZg6hMRkXRyFaQGDBhg//9rr72WaRuTyWS/t4uISHYdPnyYBQsWcOXKFZycnIiJibEHKZEiK/YkfFMdLIm3buvsDq0/gPqPg6b7FxFxmFwFqbVr1+Z1HSJSzKWkpLBixQr+/PNPwDb61KtXL4KDgx1cmUgBSIjJXojyrQThC8CvTv7XJCIiN5XtIPXyyy/Tv39/6tWrR+vWrfOzJhEpZk6fPs28efM4f/48AE2aNKFDhw72WyyIyD+6fKcQJSJSSGR7sol3332XvXv32pfPnz+P2WxmzZo1+VKYiBQfMTExnD9/Hm9vb/71r3/RpUsXhSiRzDi7O7oCERH5x21N8WMYRl7VISLFjMViwWw2A1CvXj0SExOpV69ephPXiBR5yXGOrkBERHIoV9Ofi4jklmEY/Pnnn3z22WfEx8cDtslpmjZtqhAlxYthhROrYPFD8PN9jq5GRERySDedEJECExcXx8KFCzl8+DAAf/75J/fee6+DqxIpYJeOwr5psO9biDvp6GpERCSXchSkjh8/zvbt2wG4fPkyAIcOHaJkyZKZtr/rrrturzoRKTIOHDjAL7/8QkJCAmazmQ4dOtC0aVNHlyVSMFKuwl9zYO9UOLX+2nq3klBjAAQ3h6WDHVaeiIjkXI6C1GuvvZbhvlFPPPFEhnaGYeg+UiICQGJiIsuWLWPXrl0ABAUF0atXLwICAhxcmUg+Mww4vRH2TYWInyDlyj8bTFChI9QZBlXCbRNIxJ4Es/vNp0A3u4OHbkotIlJYZDtITZ06NT/rEJEiat26dezatQuTyUTLli1p06aNfZIJkSIp7hTsn247fe/ioWvrS1aB2kOh1mDwCU2/j095eDjCdj+prHj42dqJiEihkO0gNWTIkPysQ0SKqDZt2hAZGUm7du0oX14fAqWISk2Ewwtso08nVtomkgBwKQHV+tlGn0LuAZMp62P4lFdQEhG5g2iyCRHJU1FRUezatYuOHTtiMplwd3dn6NChji5LJO8ZBkRts133FPEDJF68tq3cvVB7GFTrC65ejqtRRETyjYKUiOQJwzDYtGkTa9aswWKxEBAQQIMGDRxdlkjei4+G/d/bRp9irt2oHu9QqD3EdvpeycoOK09ERAqGgpSI3LZLly6xYMECjh8/DkC1atWoUqWKY4sSyUuWFDi2xDb6dGwxWFNt681uULW3bfSpfDtw0vV/IiLFhYKUiOSaYRjs3r2bpUuXkpSUhIuLC507d6Zhw4aYbnYtiMidImavLTwd+N42EpUmqIntuqfq/cG9pMPKExERx1GQEpFcW7FiBZs3bwagXLly9OrVi9KlSzu4KpHblHgRDv5gC1BRf15b7xkItQbZTt3zq+2w8kREpHBQkBKRXKtZsyZ//PEHrVu3pmXLljg5OTm6JJHcsVrg5CpbeDo8HyxJtvVOzlCph230KawzmF0cWqaIiBQeClIikm3JycmcOXOGsLAwAMqXL88zzzyDl5dmJZM71MVDtvs97ZsOV05dW+9fz3bdU82HwNPfYeWJiEjhpSAlItly+vRp5s2bx+XLlxkxYgR+fn4AClFy50mOg4jZtln3Tm+4tt69FNR4yDb6FNDw5vd8EhGRYk9BSkRuymKx8Ntvv/Hrr79iGAbe3t7Ex8c7uiyRnDEMOPWrLTz9NQdSrtrWm5wgrJNt9Kny/eDs5tg6RUTkjqEgJSJZOn/+PPPmzeP06dMA1KlTh65du+Lh4eHgykSyKfYk7PvWdvre5aPX1peqZgtPtQaBd4jDyhMRkTuXgpSIZGrbtm0sW7aM1NRU3N3d6dq1K3Xr1nV0WSK3lpIAh+fZJo44uRowbOtdvaH6g7YAFdxcp+6JiMhtUZASkUzFxcWRmppKxYoV6dmzJ76+vo4uSSRrhgGRW23hKWIWJF2+ti20re26p6q9waWE42oUEZEiRUFKROySkpJwc7NdI9KqVStKly5N3bp1dXNdKbyuRsL+72yn7p3ff229TwXb/Z5qDwHfio6qTkREijAFKREhMTGRZcuWERkZyfDhw3F2dsZsNlOvXj1HlyaSkSUZji6yjT4dWwqGxbbe2QOq9rGNPoW2sU0kISIikk8UpESKuePHjzN//nwuX76MyWTi+PHjVKlSxdFliWQUvcs2696BGZAQc2192ea28FS9H7jpFFQRESkYClIixVRqaipr1qxh06ZNAJQqVYrw8HDKly/v4MqkWIg9mT4M3cjDD3zKQ8J5ODDTdupe9PZr20uUhVqDbafvlamR39WKiIhkoCAlUgxFRUUxd+5coqOjAWjYsCGdOnWyXx8lkq9iT8I31cGSmHUbJxeocB+cXGk7lS9tXZWetln3wu4DJ/0KExERx9FvIZFiaPny5URHR+Pp6cn9999P9erVHV2SFCcJMTcPUQDWFDi22Pb/gIa28FRzIHiUyf/6REREskFBSqQY6tGjB2vWrKFTp054eXk5uhyRzNUYAHe/AAH1HV2JiIhIBprSSKSIMwyDXbt2sWbNGvu6UqVK0adPH4UocYykS9lr1/g5hSgRESm0NCIlUoTFx8ezaNEiDhw4AEDVqlUJDQ11cFVSLCVcgMPz4a/ZcGKlo6sRERG5bQpSIkXUoUOHWLhwIVeuXMHJyYk2bdoQEhLi6LKkOEm8CIcXwF8/2cKTNdXRFYmIiOQZBSmRIiY5OZmVK1fy559/AuDn50fv3r0pW7asgyuTYiHxEhxZaAtPx1fYJo1I418PqvWD0rXgl94OK1FERCQvFMprpD799FPCwsJwd3enadOmbN26Ncu2U6ZMoVWrVpQqVYpSpUrRoUOHm7YXKcoMw2D69On2ENW0aVMee+wxhSjJX0mXYf93MK8HfB4Ay4bA0cW2EOVXF1q+CcMOwuBd0OwV8K3g6IpFRERuW6Ebkfrxxx8ZPXo0X3zxBU2bNmXSpEl06tSJiIgIAgICMrRft24dAwYMoEWLFri7u/Pee+9x3333sW/fPp3GJMWOyWSiSZMmrFq1ivDwcCpVquTokqSoSoqFo79AxE9wfNm1ez0BlKkN1ftBtQegTM2M+3r4gdn95lOgm91t7URERAopk2EYhqOLuF7Tpk25++67+eSTTwCwWq2Ehoby1FNP8eKLL95yf4vFQqlSpfjkk08YPHjwLdvHxsbi6+vL5cuX8fHxue36b5fVaiU6OpqAgACcnArlgKEUIlarlb/++gt3d3fCwsIA26hUSkoKrq6uji1OCqXbeo9JjoMji2yn7R1bCpaka9tK17wWnvxq3/pYsSdt95PKiocf+JTPWX2SL/R7SXJKfUZyqjD1mZxkg0I1IpWcnMy2bdt46aWX7OucnJzo0KEDmzZtytYx4uPjSUlJoXTp0pluT0pKIinp2i//2NhYwPYNtFqtt1F93rBarRiGUShqkcLNMAz++OMPVq1ahbu7O48//jienp4AODs7qw9JpnL8HpN8BY4twvTXbDi2DNN1o0hGqepQrR9Gtb62USiTKe1Bbn1cr3K2r5sXm70aJV/p95LklPqM5FRh6jM5qaFQBamYmBgsFguBgYHp1gcGBnLw4MFsHeOFF14gODiYDh06ZLp9/PjxvP766xnWnzt3jsTEm5xmUkCsViuXL1/GMAyHJ3IpvK5evcq6des4deoUAL6+vkRFRVGiRAkHVyaFXXbeY0yp8bidXoX7iYW4nVmdLjylelcmsUIPEsvfT2rJGrbwZAXOnSugZyAFTb+XJKfUZySnClOfiYuLy3bbQhWkbte7777LrFmzWLduHe7u7pm2eemllxg9erR9OTY2ltDQUPz9/QvNqX0mkwl/f3+HdyQpnPbv38+SJUtISEjA2dmZJk2a0KZNG8xms6NLkztAlu8xKVfh2JJ/Rp6WYEpNsG8ySlaxjzw5+dXD02TC0wG1i2Po95LklPqM5FRh6jNZZYjMFKog5efnh9lsJioqKt36qKgogoKCbrrvhAkTePfdd1m1ahX16tXLsp2bmxtubm4Z1js5OTn8G5fGZDIVqnqkcLBarSxYsIDdu3cDEBQURHh4OIZhYDab1V8k2+zvMZZE27VOET/B0UWQGn+tUcnKtqnKq/fD5F8fTCZMjitZHEy/lySn1GckpwpLn8nJ4xeqIOXq6kqjRo1YvXo14eHhgO3D4+rVqxk1alSW+73//vu8/fbbLF++nMaNGxdQtSIFK+0H22Qycc8999C6dWtMJhPR0dEOrkzuKKkJuJ1cjOnPlbbwlHL12jbfivbwREDDa9c8iYiISAaFKkgBjB49miFDhtC4cWOaNGnCpEmTuHr1KsOGDQNg8ODBhISEMH78eADee+89xowZw8yZMwkLCyMyMhIALy8vvLy8HPY8RPJCamoqKSkpeHh4ANClSxcaN25MaGgokLMLIqUYS02EY8vgr58wHfmFUilXrm3zCbs2215gI4UnERGRbCp0QerBBx/k3LlzjBkzhsjISBo0aMCyZcvsE1CcPHky3ZDb559/TnJyMn379k13nLFjxzJu3LiCLF0kT0VFRTF37lxKlixJ//79MZlMuLu720OUyE2lJsLxFbapyo8stE1dDpgAi2cITjX7Y6rxIAQ2VngSERHJhUIXpABGjRqV5al869atS7d8/Pjx/C9IpABZrVY2bdrE2rVrsVgsXL16lcuXL1OyZElHlyaFXWoSnFhhu+bpyAJ7eALAOxSq9cNatQ/nnMIICAzEpGsXREREcq1QBimR4urSpUvMnz+fEydOAFC9enV69Oihac0la5ZkOLHSFp4Oz4fk2GvbvMpB9Qds1z2VbQImJ9u9mXRdnYiIyG1TkBIpBAzDYNeuXSxdupTk5GRcXV3p3LkzDRo0wKTTruRGlmQ4scp22t7h+ZB0+do2r2Db9U7V+kFwM1t4EhERkTynICVSCKSmpvLrr7+SnJxMaGgovXr1olSpUo4uSwoTSwqcXP3PyNM8SLp0bVuJslCtry08hbRQeBIRESkAClIihYCLiwu9evXi+PHjtGzZ0uH3UJBCwpICf6+BiNm28JR44dq2EkFQta9txr2QlgpPIiIiBUxBSsQBkpOTWbFiBQEBATRp0gSA0NBQzcgnYE2Fk2ttp+0dmps+PHkGXBee7gEns+PqFBERKeYUpEQK2KlTp5g3bx4XLlzAxcWF2rVrazKJoib2JCTEZL3dww98yl9btqbC3+uvhafr9/Xwt522V70fhLRSeBIRESkkFKRECojFYuHXX3/lt99+wzAMfHx86Nmzp0JUURN7Er6pDpbErNuY3WHofog9Zrvm6dBcSDh3bbuHH1TtYwtP5e4FJ71Vi4iIFDb67SxSAGJiYpg3bx5nzpwBoG7dunTp0gUPDw8HVyZ5LiHm5iEKbNtnNILEi9fWuZeBqr1t4Sm0jcKTiIhIIaff1CL5LCEhga+++oqkpCTc3d3p1q0bderUcXRZ4miJF8G9tC08VfsnPJldHF2ViIiIZJOClEg+8/DwoHnz5pw8eZKePXvi4+Pj6JKkMGj3P6g3QuFJRETkDqUgJZIP9u3bh5+fH4GBgQC0atUKk8mkm+sWdSlX4cgv2Wsb3EIhSkRE5A6mICWShxITE1m6dCm7d+8mMDCQ4cOH4+zsrPtCFWWGAWc3w95v4OAsSLni6IpERESkAChIieSR48ePM2/ePGJjYzGZTFSrVk0jUEXZ1SjYP90WoC4cvLbeqxxcOeW4ukRERKRAKEiJ3KbU1FTWrFnDpk2bAChVqhS9evXSzXWLIksKHFtqC09HF4Fhsa139oTqD0Cdh8G5BMxo7Ng6RUREJN8pSInchri4OL7//nuio6MBuOuuu+jUqROurq4Orkzy1PmDsG+qbQTqauS19WWb2cJT9QfB7Z9JRGJP2u4Tdav7SHn45W/NIiIikq8UpERuQ4kSJXB3d6dEiRLcf//9VKtWzdElSV5JjoOI2bD3azjz+7X1Hv5QazDUfRjK1Mq4n095eDjCdj+prHj42dqJiIjIHUtBSiSHLl26RIkSJXBxccHJyYk+ffpgNpspUaKEo0uT22UYcHqj7dS9v36yzcIHYDJDxa620adK3W49255PeQUlERGRIk5BSiSbDMNg165dLF26lAYNGtClSxcA3ReqKLhy9trEERf/ura+VDVbeKo1CLyCHVefiIiIFDoKUiLZcPXqVRYtWsTBg7bZ2SIjI7FYLJjNZgdXJrlmSYGji22n7h1bem3iCJcSUK2fLUCFtATNvCgiIiKZUJASuYW//vqLhQsXcvXqVZycnGjbti0tWrTQvaHuVOf3w55v4MB3EB99bX1wy38mjngAXL0dV5+IiIjcERSkRLKQnJzMihUr2LZtGwD+/v707t2boKAgB1cmOZYUCxGzbKfund1ybb1nINQeYgtQpas7rj4RERG54yhIiWQhMTGRffv2AdCsWTPat2+Ps7N+ZO4YhgGnfv1n4ojZkJpgW+/kDJW628JTWOdbTxwhIiIikgl9KhS5jmEYmP65JsbHx4eePXvi5uZGxYoVHVyZZFvcadj/rS1AXTpybX3pGlDnEdvEESUCHVefiIiIFAkKUiL/iImJYd68ebRu3dp+P6gaNWo4uCrJFksyHFloC0/Hl4Nhta138YIa/W0BqmxTTRwhIiIieUZBSoo9wzD4448/WLlyJampqaxcuZKqVavaR6akEDu3xxaeDnyf/ga45e61nbpXra9tFj4RERGRPKYgJcVaXFwcCxYs4MgR2ylglSpVomfPngpRhVniJdvEEXu+hqg/r60vURZqD4U6w6BUVUdVJyIiIsWEgpQUW/v27WPRokUkJibi7OxMx44dufvuuxWiCiPDCn+vt93z6dDPkJpoW+/kApXv/2fiiPtsE0mIiIiIFAB96pBi6fTp08yZMweAsmXL0rt3b/z8/BxclWQQexL2fQv7psLlY9fWl6kNdR+Bmv8CT3/H1SciIiLFloKUFEshISE0bNgQb29v7r33Xsxms6NLkjSpSXBkge3UvRMrAcO23tUHagywBajAxpo4QkRERBxKQUqKhdTUVNavX0/Tpk3x8vICoEePHjqNrzCJ3vnPxBEzIPHCtfWhbW2n7lXtDS6eDitPRERE5HoKUlLkRUZGMnfuXM6dO0dUVBQDBgzAZDIpRBUGiRfhwEzbtU/RO66t9yoHdYbaJo8oWdlR1YmIiIhkSUFKiiyr1crvv//O2rVrsVqtlChRgsaNGytAOZphhROrbaNPh+eBJcm23skFqoTbTt0r3wGcdLqliIiIFF4KUlIkXbx4kfnz53Py5EnAdmPd7t27U6KE7imUZ2JPpr930408/MCn/LXly8dh3zTYOxXiTl5b71/PdsPcGgPBUxN+iIiIyJ1BQUqKnJMnTzJjxgySk5NxdXWlS5cu1K9fXyNReSn2JHxTHSyJWbcxu8PgXRC1zXbq3snV17a5lbQFp7oPQ8BdmjhCRERE7jgKUlLkBAUF4eXlhZeXF+Hh4ZQqVcrRJRU9CTE3D1Fg2/59I0i5cm1d+Q62iSOqhIOLR76WKCIiIpKfFKSkSDh58iShoaGYTCZcXV0ZPHgw3t7eODk5Obq04i3lCniXhzrDbBNH+IY5uiIRERGRPKEgJXe05ORkli9fzvbt2+ncuTNNmzYFwNfX18GVCQDtP4P6I8CkQCsiIiJFi4KU3LH+/vtv5s2bx8WLFwG4cuXKLfaQXEuOs93nKWqb7ev0huztV7apQpSIiIgUSQpScsexWCysX7+eDRs2YBgGPj4+hIeHU7FiRUeXVjQkxdru6ZQWmqK2wcW/AMPRlYmIiIgUGgpSckeJiYlh7ty5nD17FoB69erRpUsX3N3dHVzZHSrpchahKRNeIRDYyPbl5gtrnynQUkVEREQKEwUpuaMkJycTFRWFh4cH3bp1o3bt2o4u6c6ReAmit6cPTZcOZ97WO/RaaApsZJuivETgte1R2wukZBEREZHCSkFKCr3U1FScnW1dNTg4mF69elGhQgW8vb0dXFkhlnjRFnbSAlP0Nrh0JPO2PhWuhaW04OTpf/Pje/jZ7hN1q/tIeegGuyIiIlI0KUhJobZ3716WL1/OQw89RFBQEAB16tRxcFWFiynpIpzYZTtFL/qf4HT5WOaNfcIyjjR55iLs+JSHhyNs95PKioefrZ2IiIhIEaQgJYVSQkICS5cuZc+ePQBs3ryZ8PBwxxZVGCScT3dqnilqG4GxxzNv61vphpGmu8CjTN7V4lNeQUlERESKLQUpKXSOHj3KggULiI2NxWQyce+999KqVStHl1Xw4s+lv54pahvEnUzXxPTPv4ZvZUzXjzQF3gXupQq+ZhEREZFiQkFKCo2UlBRWr17Nli1bAChdujS9evWiXLlyDq6sAMRHZxKa/s68bamqEGALS1b/hpxzCsW/XFVMTrpfk4iIiEhBUZCSQmP37t32ENWoUSPuu+8+XF1dHVxVPrgamTE0XTmdedtS1W64pqmhberxNFYrRnR0wdQtIiIiInYKUlJoNGzYkGPHjlG/fn2qVq3q6HIg9uTtT6Zw5Wz6wBS9Da6cyaShCUpXT39NU0BDcPO5racgIiIiIvlDQUoc5uLFi6xfv55u3brh4uKCk5MTffv2dXRZNrEn4Zvqt57e++EIW5gyDFtAuj4wRW2zjT5lYILSNW4YaWoArprOXUREROROoSAlBc4wDHbu3MmyZctITk7Gw8ODTp06Obqs9BJibh6iwLZ985vXAlR8VMY2JicoXTP9dOMBDcDVK1/KFhEREZGCoSAlBerq1av88ssvREREAFChQgWaNm3q4KoyYbVkr92er6793+QEZWr9E5jSglN9cCmRPzWKiIiIiMMoSEmBiYiI4JdffuHq1auYzWbatWtHs2bNcMrv2easFki6ZBtlsn+dT7+ceOPyhewdu1IPCOtkm27cvz64eObrUxERERGRwkFBSjLKi0kWbrB161aWLl0KQEBAAL179yYwMDDntRlWSLx46yB0/fbEC4CR88fKjhbjbCFKRERERIoVBSlJL6eTLGRTjRo1WLduHQ0aNKBdu3Y4Ozv/E4ou3ToIXb898YJtv9xw87WFwLQv9zLplz2uW75yBuZ0zN3jiIiIiEiRpyAl6WV3koWEmPRByrBC0mV7CLJcjSL6+H7KlnSBxPP4JMTwbJUozBd+ge+euRaObjcU3SwMXb/dvTSYXbJ//NRbvAYiIiIiUqwpSEnurHsOjNTrRovOg3FtggYzUPaGXcxZHcvV59ZB6Prt7qXBXARv1CsiIiIidwwFKcmdU2szXW0xexKb4ka84UGikzd+5arjG1jpJkGpkIYiDz/bKYy3OsXRw6/gahIRERGRQkNBSnKnyUu2SRb+GTGKs7ixcMVGDh87CUDlypXp2bMn3t536E1mfcrbrgPL40k3RERERKRoUJCS3KnW1z5b3YEDB1i4cB6JiYk4Oztz33330bhxY0wmk4OLvE0+5RWURERERCRTClJy20wmE4mJiQQHB9OrVy/8/HS6m4iIiIgUbQpSkiuJSYm4//P/GjVq8OCDD1K1alXM5iynlBARERERKTKcHF2AFDJpkyzchMXkwtRZvxAXF2dfV6NGDYUoERERESk2NCIl6d1kkoVzMedYs2YtZy4mEWu4ExERQePGjR1QpIiISOFksVhISUlxdBkOZbVaSUlJITExEScn/c1ebq2g+4yrq2uePI6ClGR0wyQLVquVjRs3sm7dVqxWD7y8/Bl4//1UrVrVgUWKiIgUHoZhEBkZyaVLlxxdisMZhoHVaiUuLu7On3hKCkRB9xknJycqVqyIq+vt3YJHQUpu6sKFC8yfP5+///4bgJo1a9K9e3c8PT0dXJmIiEjhkRaiAgIC8PT0LNYBwjAMUlNTcXZ2Ltavg2RfQfYZq9XKmTNnOHv2LOXLl7+tx1OQkpvasmULf//9N66urnTt2pV69erpTVFEROQ6FovFHqLKlCnj6HIcTkFKcqqg+4y/vz9nzpwhNTUVFxeXXB9HQUpuqn379iQlJdGmTRtKlizp6HJEREQKnbRronS2hsidIe2UPovFcltBSlcASjoRERHMmTMHwzAAW0cLDw9XiBIREbkFjb6I3Bny6mdVI1ICQFJSEsuXL2fHjh0AVK5cmYYNGzq4KhERERGRwkkjUsLJkyeZPHmyPUS1aNGCunXrOrgqERERuZONGzeOwMBATCYT8+fPz7fHye/jZ9fx48cxmUzs3LkTgHXr1mEymewzOU6bNu2OOMPnxuchWVOQKsYsFgurV69m2rRpXLx4EV9fX4YMGULHjh1xdtZgpYiISFE3dOhQTCYTJpMJV1dXqlSpwhtvvEFqauptHffAgQO88cYbTJ48mbNnz9KlS5fbrnXcuHE0aNDgto+TmeTkZN5//33q16+Pp6cnfn5+tGzZkqlTp+b6vmAtWrTg7Nmz+Pr65nG1GSn8OIY+LRdjCxYsYM+ePQDUr1+fzp074+7u7uCqREREpCB17tyZqVOnkpSUxJIlS3jyySdxcXHhpZdeyvGxLBYLAEePHgWgZ8+ehf7aseTkZDp16sSuXbt48803admyJT4+PmzevJkJEybQsGHDXAU4V1dXgoKCbru2273XkeQfjUgVY82bN8fLy4sHHniA8PBwhSgREZE8lpycnOXXjaM+N2t746hIVu1yw83NjaCgICpUqMDIkSPp0KEDCxcuBGzXUD/33HOEhIRQokQJmjZtyrp16+z7pp2utnDhQmrVqoWbmxuPPPIIvXr1Amw3Pr0+SH311VfUrFkTd3d3atSowWeffZaullOnTjFgwABKly5NiRIlaNy4MVu2bGHatGm8/vrr7Nq1yz6CNm3atAzPpV27dowaNSrdunPnzuHq6srq1aszff6TJk3i119/ZfXq1Tz55JM0aNCASpUqMXDgQLZs2ULVqlUBWLZsGffccw8lS5akTJkydO/enSNHjmT5ut54al+a+fPnU7VqVdzd3enUqZP9Xp1wbdTtq6++omLFivbPZrd67IoVKwLQsGFDTCYTbdq0yfZrvnXrVho2bIi7uzuNGze2X+oht6YRqWLk8uXLnDp1itq1awNQtmxZnn76aZ3GJyIikk/Gjx+f5baqVasycOBA+/KECROyPI2sQoUKDB061L780UcfER8fn6Hd2LFjc1/sPzw8PDh//jwAo0aNYv/+/cyaNYvg4GDmzZtH586d2bNnjz1gxMfH89577/HVV19RpkwZgoKCaNWqFcOHD+fs2bP2486YMYMxY8bwySef0LBhQ3bs2MGjjz5KiRIlGDJkCFeuXKF169aEhISwcOFCgoKC2L59O1arlQcffJC9e/eybNkyVq1aBZDpKXPDhw9n1KhRTJw4ETc3NwC+//57QkJCaNeuXabPd8aMGXTo0CHTSbZcXFzs02NfvXqV0aNHU69ePa5cucKYMWPo1asXO3fuxMkpe2MT8fHxvP3220yfPh1XV1eeeOIJ+vfvz8aNG+1tDh8+zM8//8zcuXMxm83ZeuytW7fSpEkTVq1aRe3ate2jWNl5zbt3707Hjh35/vvvOXbsGE8//XS2nosoSBUbe/bsYfHixaSkpFC6dGnKli0LoBAlIiIigO2mqKtXr2b58uU89dRTnDx5kqlTp3Ly5EmCg4MBeO6551i2bBlTp07lnXfeAWz30frss8+oX7++/ThpIef6U9vGjh3LxIkT6d27N2AbRdm/fz+TJ09myJAhzJw5k3PnzvHHH39QunRpAKpUqWLf38vLC2dn55ueLte7d29GjRrFggUL6NevH2AbNUu7Fiwzhw4dSjeCk5U+ffqkW/7mm2/w9/dn//791KlT55b7g+21+uSTT2jatCkA3377LTVr1rQHIbCNNk6fPh1/f/9sP3Za27QgmyY7r7nVauXrr7/G3d2d2rVrc+rUKUaOHJmt51Pc6VN0EZeQkMCSJUvYu3cvAMHBwTrXVkREpIDc7DqjG0cxnnvuuSzb3hgC8nLUYNGiRXh5eZGSkoLVamXgwIGMGzeOdevWYbFYqFatWrr2SUlJlClTxr7s6upKvXr1bvoYV69e5ciRIzzyyCM8+uij9vWpqan20LVz504aNmxoD1G54e7uzqBBg/jmm2/o168f27dvZ+/evfZTFTOTdu/MWzl06BBjxoxhy5YtxMTEYLVaAdvsx9kNUs7Oztx999325Ro1alCyZEkOHDhgD1IVKlRIF6Jy+9jZec0PHDhAvXr10l3e0bx582w9F1GQKtKOHDnCggULiIuLw2Qyce+999KqVSv7MLGIiIjkr5z88TK/2t5K27Zt+fzzz3F1dSU4ONh+tsqVK1cwm81s27Ytw2cHLy8v+/89PDxuOaHElStXAJgyZYp9NCZN2rE9PDxu+7mA7fS+Bg0acOrUKaZOnUq7du2oUKFClu2rVavGwYMHb3ncHj16UKFCBaZMmUJwcDBWq5U6derk+tq0rJQoUSJPHjs7r7ncHgWpImrlypX8/vvvAJQuXZrevXsTEhLi4KpERESksClRokS6U+jSNGzYEIvFQnR0NK1atbqtxwgMDCQ4OJijR4/y0EMPZdqmXr16fPXVV1y4cCHTUSlXV1f7rIA3U7duXRo3bsyUKVOYOXMmn3zyyU3bDxw4kJdffpkdO3ZkuE4qJSWF5ORkEhMTiYiIYMqUKfbXYsOGDbes5Uapqan8+eef9tGniIgILl26RM2aNbPc5/z587d87LRgff3rk53XvGbNmnz33XckJibaR6U2b96c4+dVXGnWviIq7S9FjRs3ZsSIEQpRIiIikiPVqlXjoYceYvDgwcydO5djx46xdetWxo8fz+LFi3N8vNdff53x48fz8ccf89dff7Fnzx6mTp3KBx98AMCAAQMICgoiPDycjRs3cvToUX7++Wc2bdoEQFhYGMeOHWPnzp3ExMSQlJSU5WMNHz6cd999F8Mw7DMIZuWZZ56hZcuWtG/fnk8//ZRdu3Zx9OhRfvrpJ5o1a8ahQ4coVaoUZcqU4csvv+Tw4cOsWbOG0aNH5/g1cHFx4amnnmLLli1s27aNoUOH0qxZM3uwykx2HjsgIAAPDw+WLVtGVFQUly9fBm79mg8cOBCTycSjjz7K/v37WbJkCRMmTMjx8yquFKSKCKvVav+hAWjWrBnDhg2jW7duuiZKREREcmXq1KkMHjyYZ599lurVqxMeHs4ff/xB+fLlc3ys4cOH89VXXzF16lTq1q1L69atmTZtmn3qbldXV1asWEFAQABdu3albt26vPvuu/bT0Pr06UPnzp1p27Yt/v7+/PDDD1k+1oABA3B2dmbAgAG3vL2Lm5sbK1eu5Pnnn2fy5Mk0a9aMu+++m48//ph///vf1KlTBycnJ2bNmsW2bduoU6cO//nPf/i///u/HL8Gnp6evPDCCwwcOJCWLVvi5eXFjz/+eNN9svPYzs7OfPzxx0yePJng4GB69uwJ3Po19/Ly4pdffmHPnj00bNiQV155hffeey/Hz6u4MhnZvcKuiIqNjcXX15fLly/j4+Pj6HKwWq1ER0cTEBCQ7ak0L1y4wLx584iPj2fEiBEKTsVIbvqLFG/qM5JT6jO3lpiYyLFjx9Ld96c4MwyD1NRUnJ2dHXYz3uPHj1O5cmX++OMP7rrrLofUINlX0H3mZj+zOckGukbqDmYYBtu3b2f58uWkpKTg5uZGdHQ05cqVc3RpIiIiIgUuJSWF8+fP8+qrr9KsWTOFKMlXClJ3qCtXrvDLL7/w119/AbapMsPDwylZsqRjCxMRERFxkI0bN9K2bVuqVavGnDlzHF2OFHEKUneggwcP8ssvvxAfH4/ZbKZdu3Y0b97cYcPnIiIiIoVBmzZtsn1fKJHbpSB1hzEMgz///JP4+HgCAwPp1asXgYGBji5LRERERKRYUZC6QxiGgclkwmQy0bNnT/744w/uvfde+03zRERERESk4Gj6nULOYrGwevVqFi1aZF/n7e1Nu3btFKJERERERBykUAapTz/9lLCwMNzd3WnatClbt269afvZs2dTo0YN3N3dqVu3LkuWLCmgSvNXdHQ0X331FRs2bGD79u2cOXPG0SWJiIiIiAiFMEj9+OOPjB49mrFjx7J9+3bq169Pp06diI6OzrT977//zoABA3jkkUfYsWMH4eHhhIeHs3fv3gKuPO8YhsGWLVv48ssviYyMxMPDgwceeIDg4GBHlyYiIiIiIhTCIPXBBx/w6KOPMmzYMGrVqsUXX3yBp6cn33zzTabtP/roIzp37sx///tfatasyZtvvsldd93FJ598UsCV543Lly+zaNEiVqxYgcVioUqVKowcOZJatWo5ujQREREREflHobrIJjk5mW3btvHSSy/Z1zk5OdGhQwc2bdqU6T6bNm1i9OjR6dZ16tSJ+fPnZ9o+KSmJpKQk+3JsbCxgu3O71Wq9zWdwe6xWK99//z0XLlzAxcWFjh07ctddd2EymRxemxROVqsVwzDUPyTb1Gckp9Rnbi3tNUr7ypXYk5AQk/V2Dz/wKZ+7YztA1apVefrpp3nmmWccXYrDDB48mBo1avDyyy/f9rHGjRvHggUL2LFjRx5UVjil/ezc7vT1+/fvp1OnThw8eJASJUpk+Vhp72s3vrfl5L2uUAWpmJgYLBZLhum8AwMDOXjwYKb7REZGZto+MjIy0/bjx4/n9ddfz7D+3LlzJCYm5rLyvNO4cWP+/PNP2rdvT+nSpTl37pyjS5JCzGq1cvnyZQzDwMmp0A0wSyGkPiM5pT5zaykpKVitVlJTU0lNTc35AeJO4jy9DiZL1p9DDLM7qYP3gnfehqlHHnmE7777DgAXFxfKly/PQw89xIsvvpitSa2mT5/Os88+m+7zStqHVIvFkrvX4x9vvPEGb731FgBms5ly5crRs2dPxo0bh5eX120dd+HChfz555+5Psat7Nq1iyVLlvDxxx/bX4MOHTrw66+/2tsEBARwzz338N5771GhQoWbHu+ZZ55h5MiRt/V65sScOXMYN24cJ06coEqVKrzzzjt06dIly/Znz57l+eefZ9u2bRw5coRRo0YxceLEdG2+/vprvv/+e/bt2wfAXXfdxZtvvsndd99t7y9RUVG88sorrFq1ikuXLtGqVSs+/PBDqlatCsDx48epVq1apjXMnDmTvn37Uq1aNZo0acKECRN45ZVXMm2bmpqK1Wrl/PnzuLi4pNsWFxeX7depUAWpgvDSSy+lG8GKjY0lNDQUf39/fHx8HFiZjZ+fHxUqVCAgIEC/sOSWrFYrJpMJf39/9RfJFvUZySn1mVtLTEwkLi4OZ2fn3M2om3zppiEKwGRJxDn5EjhXyl2RWXBycqJz58588803JCUlsWTJEkaNGoWbm1u6M4Rutj+Q4XmbTCbMZvNtzTDs5ORE7dq1WblyJampqWzcuJFHHnmEhIQEJk+enKF9cnIyrq6u2TquyWS67dmPb/Z4n3/+OX379qVkyZL2dSaTieHDh/PGG29gGAYnTpzgP//5D8OGDUsXsK6XFjCuP05upaSkZAgNmfn9998ZNGgQ77zzDt27d7cHlG3btlGnTp1M97FYLAQEBPDqq68yadKkTF/f3377jQEDBtCiRQvc3d15//336dq1K3v37iUkJATDMOjXrx8uLi7Mnz8fHx8fPvjgA7p06cK+ffsoUaIEFStWzDD52pdffsmECRPo3r27/TEffvhhHnvsMV555ZVMv8/Ozs44OTlRpkwZ3N3d0227cfmmjEIkKSnJMJvNxrx589KtHzx4sHH//fdnuk9oaKjx4Ycfpls3ZswYo169etl6zMuXLxuAcfny5dyUnOcsFotx9uxZw2KxOLoUuQOov0hOqc9ITqnP3FpCQoKxf/9+IyEh4dpKq9Uwkq9k7+vUBsOYwK2/Tm3I3vGs1mzXPmTIEKNnz57p1nXs2NFo1qyZYRiGMXHiRKNOnTqGp6enUa5cOWPkyJFGXFycYRiGsXbtWgNI9zV27FjDarUaFSpUMN566y1j2LBhhpeXlxEaGmpMnjw53ePs3r3baNu2reHu7m6ULl3aePTRR+3HNgzDGDt2rFG/fv10+zz66KNGUFBQuu1TpkwxwsLCDJPJZBiGYVy8eNF45JFHDD8/P8Pb29to27atsXPnTsMwDGPq1KkZap46deot97vZ490oNTXV8PX1NRYtWpRufevWrY2nn3463brvvvvO8PT0tC+nvaZLliwx7rrrLsPFxcVYu3Ztpq/FlClTjBo1ahhubm5G9erVjU8//dS+7dixYwZgzJo1y7j33nsNNzc3+/O8lX79+hndunVLt65p06bGiBEjsrV/Zs8zM6mpqYa3t7fx7bffGlar1di7d68BGHv37rW3sVgshr+/vzFlypQsj9OgQQPj4YcfTrcuKSnJcHNzM1atWpXpPpn+zP4jJ9mgUI1Iubq60qhRI1avXk14eDhg+0vY6tWrGTVqVKb7NG/enNWrV6c7B3flypU0b968ACoWERERyURqPHyc+9PPMjXrnuy1+/cVcMn82pDs8PDw4Pz584Bt9Objjz+mYsWKHD16lCeeeILnn3+ezz77jBYtWjBp0iTGjBlDREQEQLpT7j744APefPNNXn75ZebMmcPIkSNp3bo11atX5+rVq3Tq1InmzZvzxx9/EB0dzfDhwxk1ahTTpk27aW3Jycn25cOHD/Pzzz8zd+5czGYzAA888AAeHh4sXboUX19fJk+eTPv27fnrr7948MEH2bt3L8uWLWPVqlUA+Pr63nK/0qVLZ/l4N9q9ezeXL1+mcePGN32dL1y4wE8//UTTpk0zbHvxxReZMGEClSpVolSpUqxbty7d9hkzZjBmzBg++eQTGjZsyI4dO3j00UcpUaIEQ4YMSXeciRMn0rBhQ9zd3Vm3bh1t27bl2LFjhIWFZVpXTucfyK34+HhSUlLsr23aHAbXjwg5OTnh5ubGhg0bGD58eIZjbNu2jZ07d/Lpp5+mW+/q6kqDBg347bffaN++fZ7Wfb1CFaQARo8ezZAhQ2jcuDFNmjRh0qRJXL16lWHDhgG2C/dCQkIYP348AE8//TStW7dm4sSJdOvWjVmzZvHnn3/y5ZdfOvJpiIiIiNxRDMNg9erVLF++nKeeegog3R+qw8LCeOutt3j88cf57LPPcHV1xdfXF5PJRFBQULrjAHTt2pUnnngCgBdeeIEPP/yQtWvXUr16dWbOnEliYiLTp0+3TwjwySef0KNHD957770M17+D7UPzzJkzadeunX1dcnIy06dPx9/fH4ANGzawdetWoqOjcXNzA2DChAnMnz+fOXPm8Nhjj+Hl5YWzs3O6mrOzX2aPl5kTJ05gNpsJCAjIsO2zzz7jq6++wjAM4uPjqVatGsuXL8/Q7o033qBjx45ZPsbYsWOZOHEivXv3BqBixYrs37+fyZMnpwtSzzzzjL0NwN9//0316tVveopfTucfyK0XXniB4OBgOnToAECNGjUoX748L730EpMnT6ZEiRJ8+OGHnDp1irNnz2Z6jK+//pqaNWvSokWLDNuCg4M5ceJEntZ8o0IXpB588EHOnTvHmDFjiIyMpEGDBixbtsz+DT158mS6c7RbtGjBzJkzefXVV3n55ZepWrUq8+fPz/IcThEREZF85+xpGxnKjuid2Rtt6r8BAhpk77FzYNGiRXh5edknzRg4cCDjxo0DYNWqVYwfP56DBw8SGxtLamoqiYmJxMfH4+l588epW7eu/f9pYSvtvqAHDhygfv366WZVa9myJVarlYiICPvnvj179uDl5YXFYiE5OZlu3bqlu8VNhQoV0oWaXbt2ceXKFcqUKZOuloSEBI4cOZJlrdnd78bHy0xCQgJubm6YTKYM2x566CH7BAhRUVG888473HfffWzbtg1vb297u5uNZl29epUjR47wyCOP8Oijj9rXp6am2kfXsjpOkyZNspzArSC9++67zJo1i3Xr1uHu7o5hGLi4uPDzzz8zfPhwSpcujdlspkOHDnTp0iXTmfwSEhKYOXMmr732WqaP4eHhQXx8fL4+j0IXpABGjRqV5al8Nw5tgm0o9oEHHsjnqkRERESyyWTK/ul1zh7Zb3cbp+xlpW3btnz++ee4uroSHBxsvzj/+PHjdO/enZEjR/L2229TunRpNmzYwCOPPEJycvItg9SNox65uZ1L9erVWbhwIc7OzgQHB2eY3OHG6a2vXLlC2bJlM/28eLMJG7K7X1bTaV/Pz8+P+Pj4TCej8PX1pUqVKgBUqVKFr7/+mrJly/Ljjz+mO3XtZo9z5YotoE+ZMiXDaYE3nm6YnXpvFBQURFRUVLp1UVFR6UbwbseECRN49913WbVqFfXq1Uu3rVGjRuzcuZPLly+TnJyMv78/TZs2zTRYzpkzh/j4eAYPHpzp41y4cIHKlSvnSc1ZKZRBSkREREQKRokSJewf7q+3bds2rFYrEydOtJ8N9NNPP6Vr4+rqisViyfFj1qxZk2nTpnH16lX7h/2NGzfi5ORE9erV0x0/s9qyctdddxEZGYmzs3OW1wBlVnN29suuBg0aALb7GaX9PytpwSchISHbxw8MDCQ4OJijR4/y0EMP5bbMLOXn/APvv/8+b7/9NsuXL7/pqFvayNqhQ4f4888/efPNNzO0+frrr7n//vuzHCHcu3cvffv2ve2ab0bzmIqIiIg4kocfmG8x5bLZ3dauAFWpUoWUlBT+97//cfToUb777ju++OKLdG3CwsK4cuUKq1evJiYmJtunUj300EO4u7szZMgQ9u7dy9q1a3nqqacYNGhQptdHZVeHDh1o3rw54eHhrFixguPHj/P777/zyiuv2O8bFRYWxrFjx9i5cycxMTEkJSVla7/s8vf356677mLDhg0ZtsXHxxMZGUlkZCS7du1i5MiRuLu7c9999+XoMV5//XXGjx/Pxx9/zF9//cWePXuYOnUqH3zwwU3327p1KzVq1OD06dNZtnn66adZtmwZEydO5ODBg4wbN44///wz3dliL730UoaRoJ07d7Jz506uXLnCuXPn2LlzJ/v377dvf++993jttdf45ptvCAsLs78OaSNsALNnz2bdunUcPXqUBQsW0LFjR8LDwzO8PocPH+bXX3/NdAIKsI2mnj592n79VX5RkBIRERFxJJ/y8HAE/Gtb1l8PR9jaFaD69evzwQcf8N5771GnTh1mzJhhn+wrTYsWLXj88cd58MEH8ff35/3338/WsT09PVm+fDkXLlzg7rvvpm/fvrRv3z7d9U+5YTKZWLJkCffeey/Dhg2jWrVq9O/fnxMnTtgDWp8+fejcuTNt27bF39+fH374IVv75cTw4cOZMWNGhvVTpkyhbNmylC1blrZt2xITE8OSJUvSjcJl9/hfffUVU6dOpW7durRu3Zpp06ZRsWLFm+4XHx9PREQEKSkpWbZJm3/gyy+/pH79+syZMyfD/ANnz57l5MmT6fZr2LAhDRs2tE8K0rBhQ7p27Wrf/vnnn5OcnEzfvn3tr0HZsmWZMGFCuuMOGjSIGjVq8O9//5tBgwbxww8/ZKjxm2++oVy5clkG0B9++IH77rvvljc6vl0mI7Ort4qR2NhYfH19uXz5cqG4Ia/VaiU6Olo35JVsUX+RnFKfkZxSn7m1xMREjh07RsWKFXN2M88iyjAMUlNTcXZ2znTCheIgISGB6tWr8+OPP+qWPNmQl30mOTmZqlWrMnPmTFq2bJlpm5v9zOYkG+gdUUREREQkD3l4eDB9+nRiYmIcXUqxc/LkSV5++eUsQ1Re0mQTIiIiIiJ5rE2bNo4uoViqUqVKjiYouR0akRIREREREckhBSkREREREZEcUpASERERyQPFfP4ukTtGXv2sKkiJiIiI3AYXFxeAbN9DSUQcKzk5Gbh2Q+Tc0mQTIiIiIrfBbDZTsmRJoqOjAds9korrtN+g6c8l5wqyz1itVs6dO4enpyfOzrcXhRSkRERERG5TUFAQgD1MFWeGYWC1WnFyclKQkmwp6D7j5ORE+fLlb/uxFKREREREbpPJZKJs2bIEBASQkpLi6HIcymq1cv78ecqUKaObOEu2FHSfcXV1zZPHUZASERERySNms/m2r7u401mtVlxcXHB3d1eQkmy5U/vMnVOpiIiIiIhIIaEgJSIiIiIikkMKUiIiIiIiIjlU7K+RSrshV2xsrIMrsbFarcTFxd1x54iKY6i/SE6pz0hOqc9ITqnPSE4Vpj6Tlgmyc9PeYh+k4uLiAAgNDXVwJSIiIiIiUhjExcXh6+t70zYmIztxqwizWq2cOXMGb2/vQnGvg9jYWEJDQ/n777/x8fFxdDlSyKm/SE6pz0hOqc9ITqnPSE4Vpj5jGAZxcXEEBwffcnSs2I9IOTk5Ua5cOUeXkYGPj4/DO5LcOdRfJKfUZySn1Gckp9RnJKcKS5+51UhUGp24KiIiIiIikkMKUiIiIiIiIjmkIFXIuLm5MXbsWNzc3BxditwB1F8kp9RnJKfUZySn1Gckp+7UPlPsJ5sQERERERHJKY1IiYiIiIiI5JCClIiIiIiISA4pSImIiIiIiOSQgpSIiIiIiEgOKUgVsE8//ZSwsDDc3d1p2rQpW7duvWn72bNnU6NGDdzd3albty5LliwpoEqlsMhJn5kyZQqtWrWiVKlSlCpVig4dOtyyj0nRk9P3mTSzZs3CZDIRHh6evwVKoZPTPnPp0iWefPJJypYti5ubG9WqVdPvp2Imp31m0qRJVK9eHQ8PD0JDQ/nPf/5DYmJiAVUrjvbrr7/So0cPgoODMZlMzJ8//5b7rFu3jrvuugs3NzeqVKnCtGnT8r3OnFKQKkA//vgjo0ePZuzYsWzfvp369evTqVMnoqOjM23/+++/M2DAAB555BF27NhBeHg44eHh7N27t4ArF0fJaZ9Zt24dAwYMYO3atWzatInQ0FDuu+8+Tp8+XcCVi6PktM+kOX78OM899xytWrUqoEqlsMhpn0lOTqZjx44cP36cOXPmEBERwZQpUwgJCSngysVRctpnZs6cyYsvvsjYsWM5cOAAX3/9NT/++CMvv/xyAVcujnL16lXq16/Pp59+mq32x44do1u3brRt25adO3fyzDPPMHz4cJYvX57PleaQIQWmSZMmxpNPPmlftlgsRnBwsDF+/PhM2/fr18/o1q1bunVNmzY1RowYka91SuGR0z5zo9TUVMPb29v49ttv86tEKWRy02dSU1ONFi1aGF999ZUxZMgQo2fPngVQqRQWOe0zn3/+uVGpUiUjOTm5oEqUQianfebJJ5802rVrl27d6NGjjZYtW+ZrnVI4Aca8efNu2ub55583ateunW7dgw8+aHTq1CkfK8s5jUgVkOTkZLZt20aHDh3s65ycnOjQoQObNm3KdJ9Nmzalaw/QqVOnLNtL0ZKbPnOj+Ph4UlJSKF26dH6VKYVIbvvMG2+8QUBAAI888khBlCmFSG76zMKFC2nevDlPPvkkgYGB1KlTh3feeQeLxVJQZYsD5abPtGjRgm3bttlP/zt69ChLliyha9euBVKz3HnulM/Azo4uoLiIiYnBYrEQGBiYbn1gYCAHDx7MdJ/IyMhM20dGRuZbnVJ45KbP3OiFF14gODg4w5uRFE256TMbNmzg66+/ZufOnQVQoRQ2uekzR48eZc2aNTz00EMsWbKEw4cP88QTT5CSksLYsWMLomxxoNz0mYEDBxITE8M999yDYRikpqby+OOP69Q+yVJWn4FjY2NJSEjAw8PDQZWlpxEpkSLq3XffZdasWcybNw93d3dHlyOFUFxcHIMGDWLKlCn4+fk5uhy5Q1itVgICAvjyyy9p1KgRDz74IK+88gpffPGFo0uTQmrdunW88847fPbZZ2zfvp25c+eyePFi3nzzTUeXJnJbNCJVQPz8/DCbzURFRaVbHxUVRVBQUKb7BAUF5ai9FC256TNpJkyYwLvvvsuqVauoV69efpYphUhO+8yRI0c4fvw4PXr0sK+zWq0AODs7ExERQeXKlfO3aHGo3LzPlC1bFhcXF8xms31dzZo1iYyMJDk5GVdX13ytWRwrN33mtddeY9CgQQwfPhyAunXrcvXqVR577DFeeeUVnJz0d31JL6vPwD4+PoVmNAo0IlVgXF1dadSoEatXr7avs1qtrF69mubNm2e6T/PmzdO1B1i5cmWW7aVoyU2fAXj//fd58803WbZsGY0bNy6IUqWQyGmfqVGjBnv27GHnzp32r/vvv98+S1JoaGhBli8OkJv3mZYtW3L48GF76Ab466+/KFu2rEJUMZCbPhMfH58hLKUFccMw8q9YuWPdMZ+BHT3bRXEya9Ysw83NzZg2bZqxf/9+47HHHjNKlixpREZGGoZhGIMGDTJefPFFe/uNGzcazs7OxoQJE4wDBw4YY8eONVxcXIw9e/Y46ilIActpn3n33XcNV1dXY86cOcbZs2ftX3FxcY56ClLActpnbqRZ+4qfnPaZkydPGt7e3saoUaOMiIgIY9GiRUZAQIDx1ltvOeopSAHLaZ8ZO3as4e3tbfzwww/G0aNHjRUrVhiVK1c2+vXr56inIAUsLi7O2LFjh7Fjxw4DMD744ANjx44dxokTJwzDMIwXX3zRGDRokL390aNHDU9PT+O///2vceDAAePTTz81zGazsWzZMkc9hUwpSBWw//3vf0b58uUNV1dXo0mTJsbmzZvt21q3bm0MGTIkXfuffvrJqFatmuHq6mrUrl3bWLx4cQFXLI6Wkz5ToUIFA8jwNXbs2IIvXBwmp+8z11OQKp5y2md+//13o2nTpoabm5tRqVIl4+233zZSU1MLuGpxpJz0mZSUFGPcuHFG5cqVDXd3dyM0NNR44oknjIsXLxZ84eIQa9euzfTzSVo/GTJkiNG6desM+zRo0MBwdXU1KlWqZEydOrXA674Vk2FoTFVERERERCQndI2UiIiIiIhIDilIiYiIiIiI5JCClIiIiIiISA4pSImIiIiIiOSQgpSIiIiIiEgOKUiJiIiIiIjkkIKUiIiIiIhIDilIiYiIiIiI5JCClIiIFGrTpk3DZDJx/Phx+7o2bdrQpk0bh9V0o8xqdKTjx49jMpmYMGFCnh1z3LhxmEwmYmJibtk2LCyMoUOH2pfXrVuHyWRi3bp19nVDhw4lLCwsz+oTESloClIiIvkg7YO1yWRiw4YNGbYbhkFoaCgmk4nu3bs7oMLsCwsLsz8Xk8lEQEAArVq1Yt68eY4uLUfi4+MZN25cug/zBS0tjKR9eXp6UqtWLV599VViY2MdVldhUBi+PyIiOeHs6AJERIoyd3d3Zs6cyT333JNu/fr16zl16hRubm4OqixnGjRowLPPPgvAmTNnmDx5Mr179+bzzz/n8ccfL/B6VqxYkeN94uPjef311wEcPpr1+eef4+XlxZUrV1ixYgVvv/02a9asYePGjZhMJofWlhciIiJwcrr532qnTJmC1Wq1Lxem74+ISHYoSImI5KOuXbsye/ZsPv74Y5ydr73lzpw5k0aNGmXrNKnCICQkhH/961/25cGDB1OlShU+/PDDLINUamoqVqsVV1fXPK8nP45ZkPr27Yufnx8Ajz/+OH369GHu3Lls3ryZ5s2bZ7pPfHw8np6eBVlmrmXnDwQuLi4FUImISP7RqX0iIvlowIABnD9/npUrV9rXJScnM2fOHAYOHJjpPlarlUmTJlG7dm3c3d0JDAxkxIgRXLx4MV27BQsW0K1bN4KDg3Fzc6Ny5cq8+eabWCyWdO3atGlDnTp12L9/P23btsXT05OQkBDef//9XD+voKAgatasybFjx4D01+RMmjSJypUr4+bmxv79+wE4ePAgffv2pXTp0ri7u9O4cWMWLlyY4bj79u2jXbt2eHh4UK5cOd566610oxbXP6cbRy0SExMZN24c1apVw93dnbJly9K7d2+OHDnC8ePH8ff3B+D111+3n1o3btw4+/55XWNOtGvXDsD+eqZ9z7Zt28a9996Lp6cnL7/8MgDR0dE88sgjBAYG4u7uTv369fn222+zPPaHH35IhQoV8PDwoHXr1uzduzfd9t27dzN06FAqVaqEu7s7QUFBPPzww5w/fz7T48XExNCvXz98fHwoU6YMTz/9NImJiena3HiNVGauv0bqZt+fqVOnYjKZ2LFjR4ZjvPPOO5jNZk6fPn3TxxIRyQ8akRIRyUdhYWE0b96cH374gS5dugCwdOlSLl++TP/+/fn4448z7DNixAimTZvGsGHD+Pe//82xY8f45JNP2LFjBxs3brT/JX/atGl4eXkxevRovLy8WLNmDWPGjCE2Npb/+7//S3fMixcv0rlzZ3r37k2/fv2YM2cOL7zwAnXr1rXXlRMpKSn8/ffflClTJt36qVOnkpiYyGOPPYabmxulS5dm3759tGzZkpCQEF588UVKlCjBTz/9RHh4OD///DO9evUCIDIykrZt25Kammpv9+WXX+Lh4XHLeiwWC927d2f16tX079+fp59+mri4OFauXMnevXvp0KEDn3/+OSNHjqRXr1707t0bgHr16gEUSI03c+TIEYB0r+f58+fp0qUL/fv351//+heBgYEkJCTQpk0bDh8+zKhRo6hYsSKzZ89m6NChXLp0iaeffjrdcadPn05cXBxPPvkkiYmJfPTRR7Rr1449e/YQGBgIwMqVKzl69CjDhg0jKCiIffv28eWXX7Jv3z42b96c4VTDfv36ERYWxvjx49m8eTMff/wxFy9eZPr06bl+/v7+/ll+fypWrMiTTz7JjBkzaNiwYbr9ZsyYQZs2bQgJCcn1Y4uI5JohIiJ5burUqQZg/PHHH8Ynn3xieHt7G/Hx8YZhGMYDDzxgtG3b1jAMw6hQoYLRrVs3+36//fabARgzZsxId7xly5ZlWJ92vOuNGDHC8PT0NBITE+3rWrdubQDG9OnT7euSkpKMoKAgo0+fPrd8LhUqVDDuu+8+49y5c8a5c+eMXbt2Gf379zcA46mnnjIMwzCOHTtmAIaPj48RHR2dbv/27dsbdevWTVeT1Wo1WrRoYVStWtW+7plnnjEAY8uWLfZ10dHRhq+vrwEYx44dS/ecWrdubV/+5ptvDMD44IMPMtRvtVoNwzCMc+fOGYAxduzYDG3yo8bMjB071gCMiIgI49y5c8axY8eMyZMnG25ubkZgYKBx9epV+/MDjC+++CLd/pMmTTIA4/vvv7evS05ONpo3b254eXkZsbGxhmFc+354eHgYp06dsrfdsmWLARj/+c9/7Osy60c//PCDARi//vprhtrvv//+dG2feOIJAzB27dplX1ehQgVjyJAh9uW1a9cagLF27Vr7uiFDhhgVKlSwL9/s+zNgwAAjODjYsFgs9nXbt283AGPq1KkZ2ouIFASd2iciks/69etHQkICixYtIi4ujkWLFmV5Wt/s2bPx9fWlY8eOxMTE2L8aNWqEl5cXa9eutbe9fhQkLi6OmJgYWrVqRXx8PAcPHkx3XC8vr3TXOLm6utKkSROOHj2areewYsUK/P398ff3p379+syePZtBgwbx3nvvpWvXp08f+ylaABcuXGDNmjX069fPXmNMTAznz5+nU6dOHDp0yH5a1pIlS2jWrBlNmjSx7+/v789DDz10y/p+/vln/Pz8eOqppzJsu9XkDQVV4/WqV6+Ov78/FStWZMSIEVSpUoXFixenuwbKzc2NYcOGpdtvyZIlBAUFMWDAAPs6FxcX/v3vf3PlyhXWr1+frn14eHi60ZomTZrQtGlTlixZYl93fT9KTEwkJiaGZs2aAbB9+/YMtT/55JPpltNe8+uPmdcGDx7MmTNn0vX/GTNm4OHhQZ8+ffLtcUVEbkan9omI5DN/f386dOjAzJkziY+Px2Kx0Ldv30zbHjp0iMuXLxMQEJDp9ujoaPv/9+3bx6uvvsqaNWsyTJ19+fLldMvlypXLEChKlSrF7t27s/UcmjZtyltvvWWfsrtmzZqULFkyQ7uKFSumWz58+DCGYfDaa6/x2muvZfmcQkJCOHHiBE2bNs2wvXr16res78iRI1SvXj3dhB7ZVVA1Xu/nn3/Gx8cHFxcXypUrR+XKlTO0CQkJyTCpxokTJ6hatWqGGfFq1qxp3369qlWrZjhutWrV+Omnn+zLFy5c4PXXX2fWrFnp+hdk7EeZHbNy5co4OTnl6z20OnbsSNmyZZkxYwbt27fHarXyww8/0LNnT7y9vfPtcUVEbkZBSkSkAAwcOJBHH32UyMhIunTpkmkIAdtEEwEBAcyYMSPT7WmjPZcuXaJ169b4+PjwxhtvULlyZdzd3dm+fTsvvPBChskPzGZzpsczDCNb9fv5+dGhQ4dbtrvxWqG0Op577jk6deqU6T5VqlTJVg35xRE13nvvvfZZ+7Jyu9ddZVe/fv34/fff+e9//0uDBg3w8vLCarXSuXPnbE2iURDTtZvNZgYOHMiUKVP47LPP2LhxI2fOnEk3yioiUtAUpERECkCvXr0YMWIEmzdv5scff8yyXeXKlVm1ahUtW7a86QfpdevWcf78eebOncu9995rX58261thUalSJcB2+tmtgliFChU4dOhQhvURERG3fJzKlSuzZcsWUlJSspxWO6sP/AVVY16oUKECu3fvxmq1phuVSjuVs0KFCunaZ1brX3/9ZZ8t7+LFi6xevZrXX3+dMWPG3HS/67ddP/J4+PBhrFar/Zi5datANnjwYCZOnMgvv/zC0qVL8ff3zzL4iogUBF0jJSJSALy8vPj8888ZN24cPXr0yLJdv379sFgsvPnmmxm2paamcunSJeDaCNP1I0rJycl89tlneVv4bQoICKBNmzZMnjyZs2fPZth+7tw5+/+7du3K5s2b2bp1a7rtWY3OXa9Pnz7ExMTwySefZNiW9hqlXX+U9hoWdI15oWvXrkRGRqYL46mpqfzvf//Dy8uL1q1bp2s/f/78dFODb926lS1btthnasysHwFMmjQpyxo+/fTTdMv/+9//AHI1++P1svr+pKlXrx716tXjq6++4ueff6Z///65OpVTRCSv6B1IRKSADBky5JZtWrduzYgRIxg/fjw7d+7kvvvuw8XFhUOHDjF79mw++ugj+vbtS4sWLShVqhRDhgzh3//+NyaTie+++y7bp+oVpE8//ZR77rmHunXr8uijj1KpUiWioqLYtGkTp06dYteuXQA8//zzfPfdd3Tu3Jmnn37aPrV42ijMzQwePJjp06czevRotm7dSqtWrbh69SqrVq3iiSeeoGfPnnh4eFCrVi1+/PFHqlWrRunSpalTpw516tQpkBrzwmOPPcbkyZMZOnQo27ZtIywsjDlz5rBx40YmTZqU4XqhKlWqcM899zBy5EiSkpKYNGkSZcqU4fnnnwfAx8eHe++9l/fff5+UlBRCQkJYsWLFTUc2jx07xv3330/nzp3ZtGkT33//PQMHDqR+/fq39dxu9v1JM3jwYJ577jkAndYnIo7nwBkDRUSKrOunP7+ZG6c/T/Pll18ajRo1Mjw8PAxvb2+jbt26xvPPP2+cOXPG3mbjxo1Gs2bNDA8PDyM4ONh4/vnnjeXLl2eYZrp169ZG7dq1MzzGjdNP57TG66VNt/1///d/mW4/cuSIMXjwYCMoKMhwcXExQkJCjO7duxtz5sxJ12737t1G69atDXd3dyMkJMR48803ja+//vqW058bhm0a71deecWoWLGi4eLiYgQFBRl9+/Y1jhw5Ym/z+++/G40aNTJcXV0zTLWd1zVmJm0K8XPnzt20XVbfM8MwjKioKGPYsGGGn5+f4erqatStWzfDFODXfz8mTpxohIaGGm5ubkarVq3STVNuGIZx6tQpo1evXkbJkiUNX19f44EHHjDOnDmT4fVJq33//v1G3759DW9vb6NUqVLGqFGjjISEhHTHzM3054Zx8++PYRjG2bNnDbPZbFSrVu2mr5+ISEEwGUYh/POliIiIyA1iYmIoW7YsY8aMyXKGRRGRgqJrpEREROSOMG3aNCwWC4MGDXJ0KSIiukZKRERECrc1a9awf/9+3n77bcLDw297hkARkbygU/tERESkUGvTpg2///47LVu25PvvvyckJMTRJYmIKEiJiIiIiIjklK6REhERERERySEFKRERERERkRxSkBIREREREckhBSkREREREZEcUpASERERERHJIQUpERERERGRHFKQEhERERERySEFKRERERERkRz6fxhefAPeVc2+AAAAAElFTkSuQmCC", + "text/plain": [ + "

" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "📊 Brier Score: 0.1297\n", + "✅ GOOD: Reliable enough for research.\n" + ] + } + ], + "source": [ + "# ============================================================\n", + "# CELL: Model Calibration (Fixed Version)\n", + "# ============================================================\n", + "\n", + "from sklearn.calibration import calibration_curve\n", + "from sklearn.metrics import brier_score_loss\n", + "import matplotlib.pyplot as plt\n", + "import torch\n", + "import numpy as np\n", + "\n", + "# 1. Generate Predictions first! (This fixes the NameError)\n", + "print(\"🚀 Running final evaluation to get probabilities...\")\n", + "predictions_output = trainer.predict(test_fast)\n", + "\n", + "# Extract Labels (y_true) and Probabilities (y_probs)\n", + "y_true = predictions_output.label_ids\n", + "logits = predictions_output.predictions\n", + "\n", + "# Convert Logits -> Probabilities (Softmax)\n", + "y_probs = torch.softmax(torch.tensor(logits), dim=-1)[:, 1].numpy()\n", + "\n", + "print(f\"✅ Extracted {len(y_probs)} predictions.\")\n", + "\n", + "# 2. Calibration Plot Function\n", + "def plot_calibration_report(y_true, y_probs, model_name=\"PathoPreter\"):\n", + " # Calculate Brier Score (Lower is better)\n", + " brier = brier_score_loss(y_true, y_probs)\n", + " \n", + " # Generate Calibration Curve\n", + " prob_true, prob_pred = calibration_curve(y_true, y_probs, n_bins=10)\n", + " \n", + " # Plotting\n", + " plt.figure(figsize=(10, 7))\n", + " plt.plot([0, 1], [0, 1], linestyle='--', color='gray', label='Perfectly Calibrated')\n", + " plt.plot(prob_pred, prob_true, marker='s', label=f'{model_name} (Brier: {brier:.4f})', color='darkorange')\n", + " \n", + " plt.xlabel('Mean Predicted Probability', fontsize=12)\n", + " plt.ylabel('Fraction of Positives (Actual)', fontsize=12)\n", + " plt.title(f'🧬 Clinical Reliability Diagram: {model_name}', fontsize=14)\n", + " plt.legend(loc='lower right')\n", + " plt.grid(alpha=0.3)\n", + " \n", + " plt.text(0.05, 0.9, \"Above line = Under-confident\\nBelow line = Over-confident\", \n", + " fontsize=10, bbox=dict(facecolor='white', alpha=0.5))\n", + " \n", + " plt.show()\n", + " \n", + " print(f\"📊 Brier Score: {brier:.4f}\")\n", + " if brier < 0.1:\n", + " print(\"✅ EXCELLENT: Model is highly reliable for clinical use.\")\n", + " elif brier < 0.15:\n", + " print(\"✅ GOOD: Reliable enough for research.\")\n", + " elif brier < 0.2:\n", + " print(\"⚠️ MODERATE: Probabilities slightly skewed.\")\n", + " else:\n", + " print(\"❌ POOR: Use rankings only.\")\n", + "\n", + "# 3. Run it\n", + "plot_calibration_report(y_true, y_probs)" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Aligned evaluation rows: 14221\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "======================================================================\n", + "🏆 PATHOGENICITY SOTA LEADERBOARD — ClinVar-Unseen Test\n", + "======================================================================\n", + "Model | ROC-AUC | PR-AUC \n", + "----------------------------------------------------------------------\n", + "PathoPreter | 0.9186 | 0.9284\n", + "BayesDel | 0.5949 | 0.7097\n", + "CADD | 0.5921 | 0.7079\n", + "ClinPred | 0.5886 | 0.6095\n", + "AlphaMissense | 0.5879 | 0.6050\n", + "REVEL | 0.5847 | 0.6026\n", + "ESM1b | 0.5763 | 0.5938\n", + "PrimateAI | 0.5661 | 0.5850\n", + "MPC | 0.5596 | 0.5821\n", + "EVE | 0.5223 | 0.5504\n", + "======================================================================\n", + "\n", + "🔥 PathoPreter gain over CADD: +0.3265 ROC-AUC\n", + "🔥 PathoPreter gain over REVEL: +0.3339 ROC-AUC\n", + "🔥 PathoPreter gain over AlphaMissense: +0.3307 ROC-AUC\n" + ] + } + ], + "source": [ + "# ============================================================\n", + "# CELL FINAL — PathoPreter vs DBNSFP SOTA Leaderboard\n", + "# ============================================================\n", + "\n", + "import numpy as np\n", + "import pandas as pd\n", + "import torch\n", + "from sklearn.metrics import roc_auc_score, average_precision_score, confusion_matrix, ConfusionMatrixDisplay\n", + "import matplotlib.pyplot as plt\n", + "\n", + "# ---------------------------------------------\n", + "# 1. Run model on HF test dataset\n", + "# ---------------------------------------------\n", + "preds = trainer.predict(test_fast) # use test_fast if you used cell-3 pipeline\n", + "logits = preds.predictions\n", + "labels = preds.label_ids\n", + "\n", + "# Convert logits -> probability of Pathogenic\n", + "probs = torch.softmax(torch.tensor(logits), dim=-1)[:,1].cpu().numpy()\n", + "\n", + "# ---------------------------------------------\n", + "# 2. Recover original variant IDs\n", + "# ---------------------------------------------\n", + "# test_fast was created from test_df in cell-3, in same order\n", + "test_df = pd.read_parquet(\"test_enriched_SEQUENCES.parquet\").reset_index(drop=True)\n", + "\n", + "assert len(test_df) == len(probs), \"❌ Row mismatch between model and dataframe\"\n", + "\n", + "pred_df = pd.DataFrame({\n", + " \"variant_id\": test_df[\"variant_id\"].values,\n", + " \"y_true\": labels,\n", + " \"PathoPreter\": probs\n", + "})\n", + "\n", + "# ---------------------------------------------\n", + "# 3. Collect DBNSFP benchmark tools\n", + "# ---------------------------------------------\n", + "bench_cols = [\n", + " \"CADD_raw_rankscore\",\n", + " \"REVEL_rankscore\",\n", + " \"AlphaMissense_rankscore\",\n", + " \"PrimateAI_rankscore\",\n", + " \"MPC_rankscore\",\n", + " \"ClinPred_rankscore\",\n", + " \"BayesDel_addAF_rankscore\",\n", + " \"EVE_rankscore\",\n", + " \"ESM1b_rankscore\"\n", + "]\n", + "\n", + "bench_df = test_df[[\"variant_id\"] + bench_cols].copy()\n", + "\n", + "# ---------------------------------------------\n", + "# 4. Merge safely (alignment lock)\n", + "# ---------------------------------------------\n", + "merged = pred_df.merge(bench_df, on=\"variant_id\", how=\"inner\")\n", + "\n", + "print(\"Aligned evaluation rows:\", len(merged))\n", + "\n", + "y = merged[\"y_true\"].values\n", + "\n", + "# ---------------------------------------------\n", + "# 5. Confusion Matrix (PathoPreter)\n", + "# ---------------------------------------------\n", + "y_pred = (merged[\"PathoPreter\"] > 0.5).astype(int)\n", + "\n", + "cm = confusion_matrix(y, y_pred)\n", + "disp = ConfusionMatrixDisplay(cm, display_labels=[\"Benign\", \"Pathogenic\"])\n", + "disp.plot(cmap=\"Blues\")\n", + "plt.title(\"PathoPreter Confusion Matrix (Unseen Test)\")\n", + "plt.show()\n", + "\n", + "# ---------------------------------------------\n", + "# 6. Full SOTA Leaderboard\n", + "# ---------------------------------------------\n", + "models = {\n", + " \"PathoPreter\": merged[\"PathoPreter\"],\n", + " \"CADD\": merged[\"CADD_raw_rankscore\"],\n", + " \"REVEL\": merged[\"REVEL_rankscore\"],\n", + " \"AlphaMissense\": merged[\"AlphaMissense_rankscore\"],\n", + " \"PrimateAI\": merged[\"PrimateAI_rankscore\"],\n", + " \"MPC\": merged[\"MPC_rankscore\"],\n", + " \"ClinPred\": merged[\"ClinPred_rankscore\"],\n", + " \"BayesDel\": merged[\"BayesDel_addAF_rankscore\"],\n", + " \"EVE\": merged[\"EVE_rankscore\"],\n", + " \"ESM1b\": merged[\"ESM1b_rankscore\"],\n", + "}\n", + "\n", + "results = []\n", + "for name, scores in models.items():\n", + " auc = roc_auc_score(y, scores)\n", + " pr = average_precision_score(y, scores)\n", + " results.append((name, auc, pr))\n", + "\n", + "results.sort(key=lambda x: x[1], reverse=True)\n", + "\n", + "print(\"\\n\" + \"=\"*70)\n", + "print(\"🏆 PATHOGENICITY SOTA LEADERBOARD — ClinVar-Unseen Test\")\n", + "print(\"=\"*70)\n", + "print(f\"{'Model':20s} | {'ROC-AUC':8s} | {'PR-AUC':8s}\")\n", + "print(\"-\"*70)\n", + "\n", + "for name, auc, pr in results:\n", + " print(f\"{name:20s} | {auc:8.4f} | {pr:8.4f}\")\n", + "\n", + "print(\"=\"*70)\n", + "\n", + "# ---------------------------------------------\n", + "# 7. Gains vs CADD (industry baseline)\n", + "# ---------------------------------------------\n", + "scores = dict((n,a) for n,a,_ in results)\n", + "\n", + "print(f\"\\n🔥 PathoPreter gain over CADD: {scores['PathoPreter'] - scores['CADD']:+.4f} ROC-AUC\")\n", + "print(f\"🔥 PathoPreter gain over REVEL: {scores['PathoPreter'] - scores['REVEL']:+.4f} ROC-AUC\")\n", + "print(f\"🔥 PathoPreter gain over AlphaMissense: {scores['PathoPreter'] - scores['AlphaMissense']:+.4f} ROC-AUC\")\n" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Collecting seaborn\n", + " Downloading seaborn-0.13.2-py3-none-any.whl.metadata (5.4 kB)\n", + "Requirement already satisfied: numpy!=1.24.0,>=1.20 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from seaborn) (2.2.6)\n", + "Requirement already satisfied: pandas>=1.2 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from seaborn) (2.3.3)\n", + "Requirement already satisfied: matplotlib!=3.6.1,>=3.4 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from seaborn) (3.10.8)\n", + "Requirement already satisfied: contourpy>=1.0.1 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (1.3.2)\n", + "Requirement already satisfied: cycler>=0.10 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (0.12.1)\n", + "Requirement already satisfied: fonttools>=4.22.0 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (4.61.1)\n", + "Requirement already satisfied: kiwisolver>=1.3.1 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (1.4.9)\n", + "Requirement already satisfied: packaging>=20.0 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (25.0)\n", + "Requirement already satisfied: pillow>=8 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (12.1.0)\n", + "Requirement already satisfied: pyparsing>=3 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (3.3.1)\n", + "Requirement already satisfied: python-dateutil>=2.7 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (2.9.0.post0)\n", + "Requirement already satisfied: pytz>=2020.1 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from pandas>=1.2->seaborn) (2025.2)\n", + "Requirement already satisfied: tzdata>=2022.7 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from pandas>=1.2->seaborn) (2025.3)\n", + "Requirement already satisfied: six>=1.5 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from python-dateutil>=2.7->matplotlib!=3.6.1,>=3.4->seaborn) (1.17.0)\n", + "Downloading seaborn-0.13.2-py3-none-any.whl (294 kB)\n", + "Installing collected packages: seaborn\n", + "Successfully installed seaborn-0.13.2\n" + ] + } + ], + "source": [ + "!pip install seaborn" + ] + }, + { + "cell_type": "code", + "execution_count": 13, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/tmp/ipykernel_10337/1626609002.py:31: UserWarning: Glyph 127942 (\\N{TROPHY}) missing from font(s) DejaVu Sans.\n", + " plt.tight_layout()\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "import pandas as pd\n", + "import matplotlib.pyplot as plt\n", + "\n", + "# 1. Create the dataframe from the 'results' list you already have\n", + "# (This fixes the NameError)\n", + "leaderboard_df = pd.DataFrame(results, columns=[\"Model\", \"ROC-AUC\", \"PR-AUC\"])\n", + "\n", + "# 2. Define the plotting function\n", + "def plot_sota_gains(df):\n", + " # Sort for plotting\n", + " df_plot = df.sort_values(\"ROC-AUC\", ascending=True)\n", + " \n", + " plt.figure(figsize=(10, 6))\n", + " \n", + " # Highlight PathoPreter in Red/Orange, others in Blue\n", + " colors = ['#A1C9F4' if x != 'PathoPreter' else '#FF9F9B' for x in df_plot['Model']]\n", + " \n", + " bars = plt.barh(df_plot['Model'], df_plot['ROC-AUC'], color=colors, edgecolor='black')\n", + " \n", + " # Add values on the bars\n", + " for bar in bars:\n", + " width = bar.get_width()\n", + " plt.text(width + 0.01, bar.get_y() + bar.get_height()/2, \n", + " f'{width:.4f}', va='center', fontweight='bold')\n", + "\n", + " plt.axvline(x=0.5, color='red', linestyle='--', label='Random Guess (0.5)')\n", + " plt.title(\"🏆 PathoPreter v1 vs. Global Benchmarks (ROC-AUC)\", fontsize=14)\n", + " plt.xlabel(\"ROC-AUC Score\", fontsize=12)\n", + " plt.xlim(0.5, 1.0) # Zoom in to show the contrast clearly\n", + " plt.grid(axis='x', linestyle=':', alpha=0.7)\n", + " plt.tight_layout()\n", + " plt.show()\n", + "\n", + "# 3. Plot it\n", + "plot_sota_gains(leaderboard_df)" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + " error: subprocess-exited-with-error\n", + " \n", + " × Getting requirements to build wheel did not run successfully.\n", + " │ exit code: 1\n", + " ╰─> [34 lines of output]\n", + " # pysam: Cython 3.2.4 is available - using cythonize if necessary\n", + " # pysam: htslib mode is shared\n", + " # pysam: HTSLIB_CONFIGURE_OPTIONS=None\n", + " '.' is not recognized as an internal or external command,\n", + " operable program or batch file.\n", + " '.' is not recognized as an internal or external command,\n", + " operable program or batch file.\n", + " # pysam: htslib configure options: None\n", + " Traceback (most recent call last):\n", + " File \"c:\\Users\\yrohi\\Documents\\CODING\\PROJECTS\\Pathopreter_xgboost\\.venv\\lib\\site-packages\\pip\\_vendor\\pyproject_hooks\\_in_process\\_in_process.py\", line 389, in \n", + " main()\n", + " File \"c:\\Users\\yrohi\\Documents\\CODING\\PROJECTS\\Pathopreter_xgboost\\.venv\\lib\\site-packages\\pip\\_vendor\\pyproject_hooks\\_in_process\\_in_process.py\", line 373, in main\n", + " json_out[\"return_val\"] = hook(**hook_input[\"kwargs\"])\n", + " File \"c:\\Users\\yrohi\\Documents\\CODING\\PROJECTS\\Pathopreter_xgboost\\.venv\\lib\\site-packages\\pip\\_vendor\\pyproject_hooks\\_in_process\\_in_process.py\", line 143, in get_requires_for_build_wheel\n", + " return hook(config_settings)\n", + " File \"C:\\Users\\yrohi\\AppData\\Local\\Temp\\pip-build-env-9qb6yr4x\\overlay\\Lib\\site-packages\\setuptools\\build_meta.py\", line 331, in get_requires_for_build_wheel\n", + " return self._get_build_requires(config_settings, requirements=[])\n", + " File \"C:\\Users\\yrohi\\AppData\\Local\\Temp\\pip-build-env-9qb6yr4x\\overlay\\Lib\\site-packages\\setuptools\\build_meta.py\", line 301, in _get_build_requires\n", + " self.run_setup()\n", + " File \"C:\\Users\\yrohi\\AppData\\Local\\Temp\\pip-build-env-9qb6yr4x\\overlay\\Lib\\site-packages\\setuptools\\build_meta.py\", line 512, in run_setup\n", + " super().run_setup(setup_script=setup_script)\n", + " File \"C:\\Users\\yrohi\\AppData\\Local\\Temp\\pip-build-env-9qb6yr4x\\overlay\\Lib\\site-packages\\setuptools\\build_meta.py\", line 317, in run_setup\n", + " exec(code, locals())\n", + " File \"\", line 468, in \n", + " File \"\", line 81, in run_make_print_config\n", + " File \"C:\\Users\\yrohi\\AppData\\Local\\Programs\\Python\\Python310\\lib\\subprocess.py\", line 421, in check_output\n", + " return run(*popenargs, stdout=PIPE, timeout=timeout, check=True,\n", + " File \"C:\\Users\\yrohi\\AppData\\Local\\Programs\\Python\\Python310\\lib\\subprocess.py\", line 503, in run\n", + " with Popen(*popenargs, **kwargs) as process:\n", + " File \"C:\\Users\\yrohi\\AppData\\Local\\Programs\\Python\\Python310\\lib\\subprocess.py\", line 971, in __init__\n", + " self._execute_child(args, executable, preexec_fn, close_fds,\n", + " File \"C:\\Users\\yrohi\\AppData\\Local\\Programs\\Python\\Python310\\lib\\subprocess.py\", line 1456, in _execute_child\n", + " hp, ht, pid, tid = _winapi.CreateProcess(executable, args,\n", + " FileNotFoundError: [WinError 2] The system cannot find the file specified\n", + " [end of output]\n", + " \n", + " note: This error originates from a subprocess, and is likely not a problem with pip.\n", + "ERROR: Failed to build 'pysam' when getting requirements to build wheel\n", + "'apt-get' is not recognized as an internal or external command,\n", + "operable program or batch file.\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Requirement already satisfied: tqdm in c:\\users\\yrohi\\documents\\coding\\projects\\pathopreter_xgboost\\.venv\\lib\\site-packages (4.67.1)\n", + "Requirement already satisfied: colorama in c:\\users\\yrohi\\documents\\coding\\projects\\pathopreter_xgboost\\.venv\\lib\\site-packages (from tqdm) (0.4.6)\n" + ] + } + ], + "source": [ + "!pip install -q transformers datasets accelerate bitsandbytes pysam pandas pyarrow fastparquet\n", + "!apt-get install -q samtools\n", + "!pip install tqdm\n" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Collecting seaborn\n", + " Using cached seaborn-0.13.2-py3-none-any.whl.metadata (5.4 kB)\n", + "Requirement already satisfied: numpy!=1.24.0,>=1.20 in c:\\users\\yrohi\\documents\\coding\\projects\\pathopreter_xgboost\\.venv\\lib\\site-packages (from seaborn) (1.26.4)\n", + "Requirement already satisfied: pandas>=1.2 in c:\\users\\yrohi\\documents\\coding\\projects\\pathopreter_xgboost\\.venv\\lib\\site-packages (from seaborn) (2.1.4)\n", + "Requirement already satisfied: matplotlib!=3.6.1,>=3.4 in c:\\users\\yrohi\\documents\\coding\\projects\\pathopreter_xgboost\\.venv\\lib\\site-packages (from seaborn) (3.10.8)\n", + "Requirement already satisfied: contourpy>=1.0.1 in c:\\users\\yrohi\\documents\\coding\\projects\\pathopreter_xgboost\\.venv\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (1.3.2)\n", + "Requirement already satisfied: cycler>=0.10 in c:\\users\\yrohi\\documents\\coding\\projects\\pathopreter_xgboost\\.venv\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (0.12.1)\n", + "Requirement already satisfied: fonttools>=4.22.0 in c:\\users\\yrohi\\documents\\coding\\projects\\pathopreter_xgboost\\.venv\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (4.61.1)\n", + "Requirement already satisfied: kiwisolver>=1.3.1 in c:\\users\\yrohi\\documents\\coding\\projects\\pathopreter_xgboost\\.venv\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (1.4.9)\n", + "Requirement already satisfied: packaging>=20.0 in c:\\users\\yrohi\\documents\\coding\\projects\\pathopreter_xgboost\\.venv\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (25.0)\n", + "Requirement already satisfied: pillow>=8 in c:\\users\\yrohi\\documents\\coding\\projects\\pathopreter_xgboost\\.venv\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (12.0.0)\n", + "Requirement already satisfied: pyparsing>=3 in c:\\users\\yrohi\\documents\\coding\\projects\\pathopreter_xgboost\\.venv\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (3.3.1)\n", + "Requirement already satisfied: python-dateutil>=2.7 in c:\\users\\yrohi\\documents\\coding\\projects\\pathopreter_xgboost\\.venv\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (2.9.0.post0)\n", + "Requirement already satisfied: pytz>=2020.1 in c:\\users\\yrohi\\documents\\coding\\projects\\pathopreter_xgboost\\.venv\\lib\\site-packages (from pandas>=1.2->seaborn) (2025.2)\n", + "Requirement already satisfied: tzdata>=2022.1 in c:\\users\\yrohi\\documents\\coding\\projects\\pathopreter_xgboost\\.venv\\lib\\site-packages (from pandas>=1.2->seaborn) (2025.3)\n", + "Requirement already satisfied: six>=1.5 in c:\\users\\yrohi\\documents\\coding\\projects\\pathopreter_xgboost\\.venv\\lib\\site-packages (from python-dateutil>=2.7->matplotlib!=3.6.1,>=3.4->seaborn) (1.17.0)\n", + "Using cached seaborn-0.13.2-py3-none-any.whl (294 kB)\n", + "Installing collected packages: seaborn\n", + "Successfully installed seaborn-0.13.2\n" + ] + } + ], + "source": [ + "!pip install seaborn\n" + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Loaded test rows: 14073\n", + "Final eval rows: 14073\n", + " CADD_raw_rankscore: Acc=0.5280 F1=0.6790 AP=0.7048 thresh=0.2114\n", + " REVEL_rankscore: Acc=0.5326 F1=0.6799 AP=0.6004 thresh=0.5393\n", + " AlphaMissense_rankscore: Acc=0.5302 F1=0.6792 AP=0.6035 thresh=0.2655\n", + " PrimateAI_rankscore: Acc=0.5205 F1=0.6749 AP=0.5813 thresh=0.2336\n", + " MPC_rankscore: Acc=0.5136 F1=0.6707 AP=0.5786 thresh=0.2331\n", + " ClinPred_rankscore: Acc=0.5393 F1=0.6837 AP=0.6066 thresh=0.2707\n", + " BayesDel_addAF_rankscore: Acc=0.5383 F1=0.6838 AP=0.7063 thresh=0.2284\n", + " EVE_rankscore: Acc=0.5039 F1=0.6689 AP=0.5466 thresh=0.0639\n", + " ESM1b_rankscore: Acc=0.5241 F1=0.6742 AP=0.5911 thresh=0.4397\n", + "\n", + "============================================================\n", + "🏆 BENCHMARK LEADERBOARD (thresholds optimized for F1)\n", + "============================================================\n", + " Model Accuracy Best_Threshold F1_Score AP\n", + " ClinPred 53.93% 0.2707 0.6837 0.6066\n", + "BayesDel_addAF 53.83% 0.2284 0.6838 0.7063\n", + " REVEL 53.26% 0.5393 0.6799 0.6004\n", + " AlphaMissense 53.02% 0.2655 0.6792 0.6035\n", + " CADD 52.80% 0.2114 0.6790 0.7048\n", + " ESM1b 52.41% 0.4397 0.6742 0.5911\n", + " PrimateAI 52.05% 0.2336 0.6749 0.5813\n", + " MPC 51.36% 0.2331 0.6707 0.5786\n", + " EVE 50.39% 0.0639 0.6689 0.5466\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "# ============================================================\n", + "# CELL: BENCHMARK ACCURACY & CONFUSION MATRIX (robust)\n", + "# ============================================================\n", + "\n", + "import pandas as pd\n", + "import numpy as np\n", + "import matplotlib.pyplot as plt\n", + "import seaborn as sns\n", + "from sklearn.metrics import (\n", + " confusion_matrix,\n", + " accuracy_score,\n", + " precision_recall_curve,\n", + " f1_score,\n", + " average_precision_score,\n", + ")\n", + "\n", + "# -------------- config --------------\n", + "TEST_FILE = \"test_enriched_SEQUENCES.parquet\"\n", + "\n", + "benchmarks = [\n", + " \"CADD_raw_rankscore\", \"REVEL_rankscore\", \"AlphaMissense_rankscore\",\n", + " \"PrimateAI_rankscore\", \"MPC_rankscore\", \"ClinPred_rankscore\",\n", + " \"BayesDel_addAF_rankscore\", \"EVE_rankscore\", \"ESM1b_rankscore\"\n", + "]\n", + "\n", + "# -------------- load --------------\n", + "df = pd.read_parquet(TEST_FILE)\n", + "print(\"Loaded test rows:\", len(df))\n", + "\n", + "# -------------- clean & map labels --------------\n", + "# find label column\n", + "label_col = None\n", + "for c in (\"clean_label\", \"label\", \"labels\"):\n", + " if c in df.columns:\n", + " label_col = c\n", + " break\n", + "if label_col is None:\n", + " raise KeyError(\"Could not find 'label', 'clean_label', or 'labels' column in test file.\")\n", + "\n", + "def map_label_val(x):\n", + " if pd.isna(x):\n", + " return np.nan\n", + " s = str(x).strip().lower()\n", + " if s in (\"pathogenic\", \"path\", \"p\", \"1\", \"true\", \"t\", \"yes\", \"y\"):\n", + " return 1\n", + " if s in (\"benign\", \"benign \", \"b\", \"0\", \"false\", \"f\", \"no\", \"n\"):\n", + " return 0\n", + " # try numeric\n", + " try:\n", + " nv = float(s)\n", + " if nv == 1.0: return 1\n", + " if nv == 0.0: return 0\n", + " except:\n", + " pass\n", + " return np.nan\n", + "\n", + "raw_labels = df[label_col]\n", + "y_true = raw_labels.map(map_label_val)\n", + "\n", + "# drop unknowns\n", + "if y_true.isnull().any():\n", + " drop_count = int(y_true.isnull().sum())\n", + " print(f\"⚠️ Dropping {drop_count} rows with unmappable labels.\")\n", + " valid_mask = ~y_true.isnull()\n", + " df = df.loc[valid_mask].reset_index(drop=True)\n", + " y_true = y_true.loc[valid_mask].reset_index(drop=True)\n", + "else:\n", + " y_true = y_true.reset_index(drop=True)\n", + "\n", + "y_true = y_true.astype(int)\n", + "print(\"Final eval rows:\", len(y_true))\n", + "if len(np.unique(y_true)) < 2:\n", + " raise RuntimeError(\"Need at least two label classes in test set for evaluation.\")\n", + "\n", + "# -------------- evaluate each benchmark --------------\n", + "results = []\n", + "confusion_matrices = {}\n", + "\n", + "for tool in benchmarks:\n", + " if tool not in df.columns:\n", + " print(f\"⚠️ Skipping {tool} (not present in parquet).\")\n", + " continue\n", + "\n", + " scores = df[tool].astype(float).copy()\n", + " # fill NaN with median (stable)\n", + " if scores.isna().any():\n", + " median_val = float(scores.median())\n", + " scores = scores.fillna(median_val)\n", + "\n", + " # handle constant scores (no thresholds)\n", + " unique_vals = np.unique(scores)\n", + " if len(unique_vals) == 1:\n", + " # fallback threshold = median (or 0.5) and compute metrics\n", + " best_thresh = unique_vals[0]\n", + " y_pred = (scores >= best_thresh).astype(int)\n", + " acc = accuracy_score(y_true, y_pred)\n", + " f1_best = f1_score(y_true, y_pred, zero_division=0)\n", + " results.append({\n", + " \"Model\": tool.replace(\"_rankscore\", \"\").replace(\"_raw\", \"\"),\n", + " \"Accuracy\": acc,\n", + " \"Best_Threshold\": float(best_thresh),\n", + " \"F1_Score\": float(f1_best),\n", + " \"AP\": float(average_precision_score(y_true, scores))\n", + " })\n", + " confusion_matrices[tool.replace(\"_rankscore\", \"\").replace(\"_raw\", \"\")] = confusion_matrix(y_true, y_pred)\n", + " print(f\" {tool}: constant scores — using threshold {best_thresh:.4f}\")\n", + " continue\n", + "\n", + " # compute PR curve then choose threshold maximizing F1 (properly index thresholds)\n", + " precision, recall, thresholds = precision_recall_curve(y_true, scores)\n", + " # precision,recall length = len(thresholds)+1\n", + " f1_all = (2 * precision * recall) / (precision + recall + 1e-12)\n", + " # only consider indices that have associated thresholds => 1..end\n", + " if len(thresholds) > 0:\n", + " idx_rel = np.argmax(f1_all[1:]) # best among entries that map to thresholds\n", + " best_idx = idx_rel + 1 # absolute index into precision/recall arrays\n", + " best_thresh = float(thresholds[best_idx - 1])\n", + " else:\n", + " # very rare: fallback\n", + " best_thresh = float(np.median(scores))\n", + "\n", + " y_pred = (scores >= best_thresh).astype(int)\n", + " acc = accuracy_score(y_true, y_pred)\n", + " f1_best = f1_score(y_true, y_pred, zero_division=0)\n", + " ap = average_precision_score(y_true, scores)\n", + "\n", + " results.append({\n", + " \"Model\": tool.replace(\"_rankscore\", \"\").replace(\"_raw\", \"\"),\n", + " \"Accuracy\": acc,\n", + " \"Best_Threshold\": best_thresh,\n", + " \"F1_Score\": float(f1_best),\n", + " \"AP\": float(ap)\n", + " })\n", + " confusion_matrices[tool.replace(\"_rankscore\", \"\").replace(\"_raw\", \"\")] = confusion_matrix(y_true, y_pred)\n", + "\n", + " print(f\" {tool}: Acc={acc:.4f} F1={f1_best:.4f} AP={ap:.4f} thresh={best_thresh:.4f}\")\n", + "\n", + "# -------------- report --------------\n", + "if not results:\n", + " print(\"❌ No benchmarks found.\")\n", + "else:\n", + " res_df = pd.DataFrame(results).sort_values(\"Accuracy\", ascending=False).reset_index(drop=True)\n", + " print(\"\\n\" + \"=\" * 60)\n", + " print(\"🏆 BENCHMARK LEADERBOARD (thresholds optimized for F1)\")\n", + " print(\"=\" * 60)\n", + " display_cols = [\"Model\", \"Accuracy\", \"Best_Threshold\", \"F1_Score\", \"AP\"]\n", + " print(res_df[display_cols].to_string(index=False, formatters={\n", + " 'Accuracy': '{:.2%}'.format,\n", + " 'Best_Threshold': '{:.4f}'.format,\n", + " 'F1_Score': '{:.4f}'.format,\n", + " 'AP': '{:.4f}'.format\n", + " }))\n", + "\n", + " # plot confusion matrices in grid\n", + " n_tools = len(res_df)\n", + " cols = 3\n", + " rows = (n_tools + cols - 1) // cols\n", + " fig, axes = plt.subplots(rows, cols, figsize=(5 * cols, 4 * rows))\n", + " axes = axes.flatten()\n", + "\n", + " for i, row in res_df.iterrows():\n", + " name = row[\"Model\"]\n", + " cm = confusion_matrices.get(name)\n", + " ax = axes[i]\n", + " sns.heatmap(cm, annot=True, fmt=\"d\", cmap=\"Blues\", cbar=False, ax=ax,\n", + " xticklabels=[\"Pred Benign\", \"Pred Patho\"], yticklabels=[\"Act Benign\", \"Act Patho\"])\n", + " ax.set_title(f\"{name}\\nAcc: {row['Accuracy']:.1%} F1: {row['F1_Score']:.3f}\")\n", + "\n", + " # hide unused axes\n", + " for j in range(i + 1, len(axes)):\n", + " axes[j].axis(\"off\")\n", + "\n", + " plt.tight_layout()\n", + " plt.show()\n" + ] + }, + { + "cell_type": "code", + "execution_count": 23, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Loaded test rows: 300000\n", + "Final eval rows: 300000\n", + " CADD_raw_rankscore: Acc=0.4964 F1=0.6372 AP=0.8636 thresh=0.6457\n", + " REVEL_rankscore: Acc=0.5144 F1=0.6622 AP=0.6449 thresh=0.4644\n", + " AlphaMissense_rankscore: Acc=0.5152 F1=0.6620 AP=0.6435 thresh=0.2841\n", + " PrimateAI_rankscore: Acc=0.5042 F1=0.6573 AP=0.6171 thresh=0.2384\n", + " MPC_rankscore: Acc=0.4952 F1=0.6519 AP=0.6013 thresh=0.2343\n", + " ClinPred_rankscore: Acc=0.5199 F1=0.6661 AP=0.6644 thresh=0.1142\n", + " BayesDel_addAF_rankscore: Acc=0.5170 F1=0.6560 AP=0.8867 thresh=0.6330\n", + " EVE_rankscore: Acc=0.4831 F1=0.6496 AP=0.5674 thresh=0.0924\n", + " ESM1b_rankscore: Acc=0.5072 F1=0.6565 AP=0.6269 thresh=0.3914\n", + "\n", + "============================================================\n", + "🏆 BENCHMARK LEADERBOARD (thresholds optimized for F1)\n", + "============================================================\n", + " Model Accuracy Best_Threshold F1_Score AP\n", + " ClinPred 51.99% 0.1142 0.6661 0.6644\n", + "BayesDel_addAF 51.70% 0.6330 0.6560 0.8867\n", + " AlphaMissense 51.52% 0.2841 0.6620 0.6435\n", + " REVEL 51.44% 0.4644 0.6622 0.6449\n", + " ESM1b 50.72% 0.3914 0.6565 0.6269\n", + " PrimateAI 50.42% 0.2384 0.6573 0.6171\n", + " CADD 49.64% 0.6457 0.6372 0.8636\n", + " MPC 49.52% 0.2343 0.6519 0.6013\n", + " EVE 48.31% 0.0924 0.6496 0.5674\n" + ] + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAABdIAAASlCAYAAACspitqAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjgsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvwVt1zgAAAAlwSFlzAAAPYQAAD2EBqD+naQABAABJREFUeJzs3Xd0FFUfxvFn0wMJIfRO6L2JFOkdpCodlKYiSBWRpiJFBEF6F0RADF1EEFEQKUrnpYn03ntJgJCQ5L5/hKwsCUsCCdnA93NOznFn7ty5s7h5Mr+duWMxxhgBAAAAAAAAAIBoOSX0AAAAAAAAAAAAcGQU0gEAAAAAAAAAsINCOgAAAAAAAAAAdlBIBwAAAAAAAADADgrpAAAAAAAAAADYQSEdAAAAAAAAAAA7KKQDAAAAAAAAAGAHhXQAAAAAAAAAAOygkA4AAAAAAAAAgB0U0oFEzs/PT23btrW+XrdunSwWi9atW5dgY3oWbdu2lZ+fX0IPAwCARO/RvxEcaZ+zZs2SxWLRyZMn431MAIAXi8Vi0cCBA5962y5dusTtgJ5SQuQ0gGdDIR1wYMeOHVOHDh2UPXt2eXh4KFmyZCpbtqzGjRunoKCgON/fyZMnZbFYrD/Ozs7KkiWL3nzzTe3evTvO9wcAwNOKLMQ+/JMmTRpVrlxZK1euTOjhxdjD43dxcVGKFClUvHhxde/eXfv370/o4SWYkiVLymKxaMqUKdGuj+7fP/Knb9++z3m0AIC4NHnyZFksFpUqVSqhh/JEkReyWSwW/fDDD9G2KVu2rCwWiwoWLPicRwcgrrkk9AAARG/FihVq0qSJ3N3d1bp1axUsWFAhISH6+++/1atXL/3777+aNm1alO0qVKigoKAgubm5PfW+W7Roodq1ayssLEwHDhzQlClTtHLlSm3ZskVFixZ9hqMCACBuDR48WNmyZZMxRpcuXdKsWbNUu3ZtLV++XHXr1k3o4cVI9erV1bp1axljdOvWLe3Zs0ezZ8/W5MmTNXz4cH300UcJPcTn6siRI9q+fbv8/Pzk7++vDz744LFtI//9H0ahAgASN39/f/n5+Wnbtm06evSocubMmdBDeiIPDw/NnTtXb7/9ts3ykydPatOmTfLw8IiyzaFDh+TkxPWtQGJCIR1wQCdOnFDz5s2VNWtW/fnnn0qfPr11XefOnXX06FGtWLEi2m2dnJyiDenYeOWVV2z+AChbtqzq16+vKVOm6Jtvvol2mzt37ihp0qTPtF8AAGLr9ddf16uvvmp9/e677ypt2rSaN29eoimk586dO8qJ91dffaV69eqpZ8+eyps3r2rXrp1Ao3v+fvjhB6VJk0ajRo1S48aNdfLkycdO+/bovz8AIHE7ceKENm3apCVLlqhDhw7y9/fXgAEDEnpYT1S7dm0tW7ZMV69eVapUqazL586dq7Rp0ypXrly6ceOGzTbu7u7Pe5gAnhFffQEOaMSIEbp9+7ZmzJhhU0SPlDNnTnXv3j3abaObI71SpUoqWLCg9u/fr8qVKytJkiTKmDGjRowYEaPxVKlSRVLEHzXSf7dTr1+/Xp06dVKaNGmUKVMma/uVK1eqfPnySpo0qby9vVWnTh39+++/UfpdunSpChYsKA8PDxUsWFA//fRTjMYDAMDjJE+eXJ6ennJx+e96kZEjR6pMmTJKmTKlPD09Vbx4cS1evNhmu4oVK6pIkSLR9pknTx7VrFnT+jo8PFxjx45VgQIF5OHhobRp06pDhw5RTpB37NihmjVrKlWqVPL09FS2bNn0zjvvxOg4UqZMqfnz58vFxUVffvmlzbrg4GANGDBAOXPmlLu7uzJnzqzevXsrODg4Rn3bM3PmTFWpUkVp0qSRu7u78ufPH+30KsYYDRkyRJkyZVKSJElUuXLlaLNekv79919VqVJFnp6eypQpk4YMGaLw8PDHjmHu3Llq3Lix6tatKx8fH82dO/eZjwsAkDj4+/vL19dXderUUePGjeXv7//EbQYOHCiLxaKDBw+qadOmSpYsmVKmTKnu3bvr3r170W4TeS7q7u6uAgUK6LfffrNZf+rUKXXq1El58uSRp6enUqZMqSZNmjz22R4NGjSQu7u7Fi1aZLN87ty5atq0qZydnaNs8+gc6ffv39egQYOUK1cueXh4KGXKlCpXrpxWr15tbXPx4kW1a9dOmTJlkru7u9KnT68GDRpEGVdMzsnbtm0rLy8vnTt3Tm+88Ya8vLyUOnVqffzxxwoLC7NpO3/+fBUvXlze3t5KliyZChUqpHHjxtm0uXnzpj788ENlzpxZ7u7uypkzp4YPH24384HEhivSAQe0fPlyZc+eXWXKlImzPm/cuKFatWqpYcOGatq0qRYvXqw+ffqoUKFCev311+1ue+zYMUkRJ/UP69Spk1KnTq3PP/9cd+7ckSTNmTNHbdq0Uc2aNTV8+HDdvXtXU6ZMUbly5bRr1y7rFWWrVq1So0aNlD9/fg0bNkzXrl2z/kEAAEBM3bp1S1evXpUxRpcvX9aECRN0+/Ztmyu8x40bp/r16+utt95SSEiI5s+fryZNmuiXX35RnTp1JEmtWrVS+/bttW/fPpupQbZv367Dhw/rs88+sy7r0KGDZs2apXbt2qlbt246ceKEJk6cqF27dmnjxo1ydXXV5cuXVaNGDaVOnVp9+/ZV8uTJdfLkSS1ZsiTGx5YlSxZVrFhRa9euVUBAgJIlS6bw8HDVr19ff//9t95//33ly5dP//zzj8aMGaPDhw9r6dKlz/R+TpkyRQUKFFD9+vXl4uKi5cuXq1OnTgoPD1fnzp2t7T7//HMNGTJEtWvXVu3atbVz507VqFFDISEhNv1dvHhRlStXVmhoqPr27aukSZNq2rRp8vT0jHb/W7du1dGjRzVz5ky5ubmpYcOG8vf31yeffBJt+8h//4c9fCUgACBx8ff3V8OGDeXm5qYWLVpoypQp2r59u0qUKPHEbZs2bSo/Pz8NGzZMW7Zs0fjx43Xjxg19//33Nu3+/vtvLVmyRJ06dZK3t7fGjx+vRo0a6fTp09Zz3u3bt2vTpk1q3ry5MmXKpJMnT2rKlCmqVKmS9u/fryRJktj0mSRJEjVo0EDz5s2zTkm2Z88e/fvvv/r222+1d+/eJ45/4MCBGjZsmN577z2VLFlSAQEB2rFjh3bu3Knq1atLkho1aqR///1XXbt2lZ+fny5fvqzVq1fr9OnT1nPtmJ6TS1JYWJhq1qypUqVKaeTIkfrjjz80atQo5ciRw3ocq1evVosWLVS1alUNHz5cknTgwAFt3LjReoHf3bt3VbFiRZ07d04dOnRQlixZtGnTJvXr108XLlzQ2LFjn3j8QKJgADiUW7duGUmmQYMGMWqfNWtW06ZNG+vrtWvXGklm7dq11mUVK1Y0ksz3339vXRYcHGzSpUtnGjVqZF124sQJI8kMGjTIXLlyxVy8eNGsW7fOFCtWzEgyP/74ozHGmJkzZxpJply5ciY0NNS6fWBgoEmePLlp3769zRgvXrxofHx8bJYXLVrUpE+f3ty8edO6bNWqVUaSyZo1a4yOHQDw8orMokd/3N3dzaxZs2za3r171+Z1SEiIKViwoKlSpYp12c2bN42Hh4fp06ePTdtu3bqZpEmTmtu3bxtjjPnrr7+MJOPv72/T7rfffrNZ/tNPPxlJZvv27XaPQ5Lp3LnzY9d3797dSDJ79uwxxhgzZ84c4+TkZP766y+bdlOnTjWSzMaNG63LHv0bISYefa+MMaZmzZome/bs1teXL182bm5upk6dOiY8PNy6/JNPPjGSbPb54YcfGklm69atNtv7+PgYSebEiRM2++rSpYvJnDmztd/Ivw127dpl0+5x//6c3gBA4rVjxw4jyaxevdoYY0x4eLjJlCmT6d69u007SWbAgAHW1wMGDDCSTP369W3aderUySZDI7d1c3MzR48etS7bs2ePkWQmTJhgXRZdHm7evDnKeXXk+feiRYvML7/8YiwWizl9+rQxxphevXpZ87NixYqmQIECNv09mtNFihQxderUeez7c+PGDSPJfP31149tE5tz8jZt2hhJZvDgwTZtixUrZooXL2593b17d5MsWTKbc/9HffHFFyZp0qTm8OHDNsv79u1rnJ2dre8JkNgxtQvgYAICAiRJ3t7ecdqvl5eXzdV5bm5uKlmypI4fPx6l7YABA5Q6dWqlS5dOlSpV0rFjxzR8+HA1bNjQpl379u1tblFbvXq1bt68qRYtWujq1avWH2dnZ5UqVUpr166VJF24cEG7d+9WmzZt5OPjY92+evXqyp8/f5weNwDgxTZp0iStXr1aq1ev1g8//KDKlSvrvffes7ny++Grn2/cuKFbt26pfPny2rlzp3W5j4+P9UoyY4ykiKu0FixYoDfeeMP6HJBFixbJx8dH1atXt8m64sWLy8vLy5p1yZMnlyT98ssvun///lMfn5eXlyQpMDDQuv98+fIpb968NvuPnIYtcv9P6+H3KvJq74oVK+r48eO6deuWJOmPP/5QSEiIunbtKovFYm3/4YcfRunv119/VenSpVWyZEnrstSpU+utt96K0jY0NFQLFixQs2bNrP1GTjPzuFv7H/73j/wBACRO/v7+Sps2rSpXrixJslgsatasmebPnx9lqpHoPHznlCR17dpVUkQWPaxatWrKkSOH9XXhwoWVLFkym3Pjh/Pw/v37unbtmnLmzKnkyZPb/P3wsBo1aihFihSaP3++jDGaP3++WrRo8cRxR0qePLn+/fdfHTlyJNr1np6ecnNz07p166JMJxcppufkD+vYsaPN6/Lly9u8F8mTJ9edO3fsZuyiRYtUvnx5+fr62uy3WrVqCgsL04YNG2LyFgAOj6ldAAeTLFkySf+dMMeVTJky2ZzsSpKvr2+0t5i9//77atKkiZycnJQ8eXIVKFAg2gehZMuWzeZ1ZOBHnsw/KvLYTp06JUnKlStXlDZ58uR57B8mAAA8qmTJkjYPm2zRooWKFSumLl26qG7dunJzc9Mvv/yiIUOGaPfu3TbziD+ai61bt9aCBQv0119/qUKFCvrjjz906dIltWrVytrmyJEjunXrltKkSRPteC5fviwpYs71Ro0aadCgQRozZowqVaqkN954Qy1btozVw8Vu374t6b8v2I8cOaIDBw4oderUdvf/tDZu3KgBAwZo8+bNunv3rs26W7duycfH57E5njp1avn6+tosO3XqlEqVKhVlP3ny5ImybNWqVbpy5YpKliypo0ePWpdXrlxZ8+bN0/Dhw+XkZHsd0KP//gCAxCksLEzz589X5cqVrc/mkqRSpUpp1KhRWrNmjWrUqGG3j0dzKUeOHHJycooyf3iWLFmibOvr62tTnA4KCtKwYcM0c+ZMnTt3zvoluyTrF8uPcnV1VZMmTTR37lyVLFlSZ86cUcuWLe2O+WGDBw9WgwYNlDt3bhUsWFC1atVSq1atVLhwYUkRDycdPny4evbsqbRp06p06dKqW7euWrdurXTp0kmK+Tl5JA8Pjyh/Uzz6XnTq1EkLFy7U66+/rowZM6pGjRpq2rSpatWqZW1z5MgR7d27N97+PgEcBYV0wMEkS5ZMGTJk0L59++K03+gebiLJ5g+CSLly5VK1atWe2Oej85tGPkRkzpw51iB/2MMPfgMAID44OTmpcuXKGjdunI4cOaLr16+rfv36qlChgiZPnqz06dPL1dVVM2fOjPIQy5o1aypt2rT64YcfVKFCBf3www9Kly6dTSaGh4fbvUI68gTSYrFo8eLF2rJli5YvX67ff/9d77zzjkaNGqUtW7ZYrzR/kn379snZ2dn65XV4eLgKFSqk0aNHR9s+c+bMMeo3OseOHVPVqlWVN29ejR49WpkzZ5abm5t+/fVXjRkzJt4fFhb5njZt2jTa9evXr7depQgAeLH8+eefunDhgubPn6/58+dHWe/v7//EQvqjHv3CPFJMzo27du2qmTNn6sMPP9Rrr70mHx8fWSwWNW/e3G4etmzZUlOnTtXAgQNVpEiRWN1xXaFCBR07dkw///yzVq1apW+//VZjxozR1KlT9d5770mKuPurXr16Wrp0qX7//Xf1799fw4YN059//qlixYrF+pz8ce/Fw9KkSaPdu3fr999/18qVK7Vy5UrNnDlTrVu31uzZsyVF/H1SvXp19e7dO9o+cufOHeP3AXBkVLUAB1S3bl1NmzZNmzdv1muvvZbQw4mxyNvj0qRJY7cQnzVrVkmK9pa1Q4cOxc/gAAAvjdDQUEkRV3P/+OOP8vDw0O+//25zJfjMmTOjbOfs7KyWLVtq1qxZGj58uJYuXRplGrMcOXLojz/+UNmyZR/7wMyHlS5dWqVLl9aXX36puXPn6q233tL8+fOtJ8T2nD59WuvXr9drr71mvSI9R44c2rNnj6pWrfrYAsHTWr58uYKDg7Vs2TKbq/UevQ384RzPnj27dfmVK1ei3GqeNWvWGOX9nTt39PPPP6tZs2Zq3LhxlPbdunWTv78/hXQAeEH5+/srTZo0mjRpUpR1S5Ys0U8//aSpU6fazd4jR47Y3DV99OhRhYeH2zxcM6YWL16sNm3aaNSoUdZl9+7d082bN+1uV65cOWXJkkXr1q2zPpgzNlKkSKF27dqpXbt2un37tipUqKCBAwfa/N2QI0cO9ezZUz179tSRI0dUtGhRjRo1Sj/88EOMz8ljy83NTfXq1VO9evUUHh6uTp066ZtvvlH//v2VM2dO5ciRQ7dv347TfQKOiDnSAQfUu3dvJU2aVO+9954uXboUZf2xY8c0bty4BBiZfTVr1lSyZMk0dOjQaOeDvXLliiQpffr0Klq0qGbPnm1zW9zq1au1f//+5zZeAMCL5/79+1q1apXc3NyUL18+OTs7y2Kx2MytevLkSS1dujTa7Vu1aqUbN26oQ4cOun37ts3zRaSIq6XDwsL0xRdfRNk2NDTUeoJ948aNKHd9FS1aVJJsppd5nOvXr6tFixYKCwvTp59+arP/c+fOafr06VG2CQoK0p07d57Y9+NEfmHw6O3rj37pUK1aNbm6umrChAk2bceOHRulz9q1a2vLli3atm2bddmVK1eiXNH/008/6c6dO+rcubMaN24c5adu3br68ccfY/TeAQASl6CgIC1ZskR169aNNgO6dOmiwMBALVu2zG4/jxbhJ0yYIEl6/fXXYz0mZ2fnKDk+YcKEJ87VbrFYNH78eA0YMMBmariYuHbtms1rLy8v5cyZ05p9d+/e1b1792za5MiRQ97e3tY2MT0nf5ZxOTk5Waebidxv06ZNtXnzZv3+++9Rtr9586b1IgcgseOKdMAB5ciRQ3PnzlWzZs2UL18+tW7dWgULFlRISIg2bdqkRYsWqW3btgk9zCiSJUumKVOmqFWrVnrllVfUvHlzpU6dWqdPn9aKFStUtmxZTZw4UZI0bNgw1alTR+XKldM777yj69eva8KECSpQoIB1PlgAAJ5k5cqVOnjwoKSI+Tfnzp2rI0eOqG/fvkqWLJnq1Kmj0aNHq1atWmrZsqUuX76sSZMmKWfOnNE+J6RYsWIqWLCg9aGer7zyis36ihUrqkOHDho2bJh2796tGjVqyNXVVUeOHNGiRYs0btw4NW7cWLNnz9bkyZP15ptvKkeOHAoMDNT06dOVLFky1a5d26bPw4cP64cffpAxRgEBAdqzZ48WLVqk27dvW8ceqVWrVlq4cKE6duyotWvXqmzZsgoLC9PBgwe1cOFC/f777089Z3iNGjWsV5xFfpEwffp0pUmTRhcuXLC2S506tT7++GMNGzZMdevWVe3atbVr1y6tXLlSqVKlsumzd+/emjNnjmrVqqXu3bsradKkmjZtmrJmzWrz/vv7+ytlypQqU6ZMtGOrX7++pk+frhUrVkR5+DkAIHFbtmyZAgMDVb9+/WjXly5dWqlTp5a/v7+aNWv22H5OnDih+vXrq1atWtq8ebN++OEHtWzZUkWKFIn1mOrWras5c+bIx8dH+fPn1+bNm/XHH38oZcqUT9y2QYMGatCgQaz3mT9/flWqVEnFixdXihQptGPHDi1evFhdunSRFPH3QtWqVdW0aVPlz59fLi4u+umnn3Tp0iU1b95cUuzOyWPqvffe0/Xr11WlShVlypRJp06d0oQJE1S0aFHly5dPktSrVy8tW7ZMdevWVdu2bVW8eHHduXNH//zzjxYvXqyTJ09G+RsBSJQMAId1+PBh0759e+Pn52fc3NyMt7e3KVu2rJkwYYK5d++eMcaYrFmzmjZt2li3Wbt2rZFk1q5da11WsWJFU6BAgSj9t2nTxmTNmtX6+sSJE0aS+frrr+2Oa+bMmUaS2b59e7Tr165da2rWrGl8fHyMh4eHyZEjh2nbtq3ZsWOHTbsff/zR5MuXz7i7u5v8+fObJUuWRBkTAADRicyih388PDxM0aJFzZQpU0x4eLi17YwZM0yuXLmMu7u7yZs3r5k5c6YZMGCAedyfwiNGjDCSzNChQx+7/2nTppnixYsbT09P4+3tbQoVKmR69+5tzp8/b4wxZufOnaZFixYmS5Ysxt3d3aRJk8bUrVs3ShY+PH4nJyeTPHlyU6xYMdO9e3fz77//RrvvkJAQM3z4cFOgQAHj7u5ufH19TfHixc2gQYPMrVu3rO0e/RshJpYtW2YKFy5sPDw8jJ+fnxk+fLj57rvvjCRz4sQJa7uwsDAzaNAgkz59euPp6WkqVapk9u3bF+0+9+7daypWrGg8PDxMxowZzRdffGFmzJhh7fPSpUvGxcXFtGrV6rHjunv3rkmSJIl58803jTFP/lsEAJB41KtXz3h4eJg7d+48tk3btm2Nq6uruXr1qpFkBgwYYF0Xmen79+83jRs3Nt7e3sbX19d06dLFBAUF2fQjyXTu3DlK/4/m140bN0y7du1MqlSpjJeXl6lZs6Y5ePDgY8+/Fy1aZPcYozsnf7SvIUOGmJIlS5rkyZMbT09PkzdvXvPll1+akJAQY4wxV69eNZ07dzZ58+Y1SZMmNT4+PqZUqVJm4cKFUfYXk3PyNm3amKRJk0bZ9tG/kRYvXmxq1Khh0qRJY9zc3EyWLFlMhw4dzIULF2y2CwwMNP369TM5c+Y0bm5uJlWqVKZMmTJm5MiR1mMAEjuLMdE8aRAAAAB4SY0bN049evTQyZMnbeYKBwAAjmfgwIEaNGiQrly5wlXPAOIVc6QDAAAADxhjNGPGDFWsWJEiOgAAAAAr5kgHAADAS+/OnTtatmyZ1q5dq3/++Uc///xzQg8pTl28eNHuek9PT/n4+Dyn0QAAAACJD4V0AAAAvPSuXLmili1bKnny5Prkk08e+8CzxCp9+vR217dp00azZs16PoMBAAAAEiHmSAcAAABecH/88Yfd9RkyZFD+/Pmf02gAAACAxIdCOgAAAAAAAAAAdvCwUQAAAAAAAAAA7KCQDgAAAAAAAACAHRTSkWhNnjxZFotFpUqVSuih2PDz85PFYony07FjR5t2a9as0TvvvKPcuXMrSZIkyp49u9577z1duHAhxvs6d+6cmjZtquTJkytZsmRq0KCBjh8/btMmODhYXbt2VerUqZUpUyYNGTIkSj9nz56Vl5eXNm7c+HQH/UB0x22xWJQuXTprmwsXLqhv376qXLmyvL29ZbFYtG7dumfab6RNmzapXLlySpIkidKlS6du3brp9u3bMd7+0qVL6tChgzJmzCgPDw/5+fnp3XffjbbtggUL9Nprrylp0qRKnjy5ypQpoz///POZ+gSAF11iz+5nzbDH7cdisShXrlzWdmR33Gb3wIEDoz1GDw+Pp+4TAF4WL3t2xzRDyO6EOe8+c+aMBg0apJIlS8rX11epUqVSpUqVnvhsGOBpuST0AICn5e/vLz8/P23btk1Hjx5Vzpw5E3pIVkWLFlXPnj1tluXOndvmdZ8+fXT9+nU1adJEuXLl0vHjxzVx4kT98ssv2r17t00IRuf27duqXLmybt26pU8++USurq4aM2aMKlasqN27dytlypSSpK+//lrff/+9Pv30UwUGBmrw4MHKkSOHWrRoYe2rV69eql+/vsqWLfvMx169enW1bt3aZpmnp6f1vw8dOqThw4crV65cKlSokDZv3vzM+5Sk3bt3q2rVqsqXL59Gjx6ts2fPauTIkTpy5IhWrlz5xO3PnDljPf6OHTsqY8aMOn/+vLZt2xal7cCBAzV48GA1btxYbdu21f3797Vv3z6dO3fuqfsEgJdBYs/uZ82wsWPHRjnRPHXqlD777DPVqFHDuozsjvvslqQpU6bIy8vL+trZ2fmZ+wSAF93Lnt2RnpQhZHfCnHf//PPPGj58uN544w21adNGoaGh+v7771W9enV99913ateuXZwcN2BlgETo+PHjRpJZsmSJSZ06tRk4cGBCD8kqa9aspk6dOk9st379ehMWFhZlmSTz6aefPnH74cOHG0lm27Zt1mUHDhwwzs7Opl+/ftZlpUqVMoMGDbK+btOmjWnevLn19V9//WWSJk1qzpw588R9Pokk07lzZ7ttAgICzLVr14wxxixatMhIMmvXrn3mfb/++usmffr05tatW9Zl06dPN5LM77//HqPts2XLZq5evWq33ebNm43FYjGjR4+Osz4B4GXwImR3fGTYF198YSSZjRs3WpeR3XGb3QMGDDCSzJUrV+KsTwB4GZDdMc8Qsjthzrv37dsX5d/m3r17Jm/evCZTpkxPHA8QW0ztgkTJ399fvr6+qlOnjho3bix/f/9o2928eVM9evSQn5+f3N3dlSlTJrVu3VpXr161trl3754GDhyo3Llzy8PDQ+nTp1fDhg117Ngxa5sLFy7o4MGDun//fozHGBISojt37jx2fYUKFeTk5BRlWYoUKXTgwIEn9r948WKVKFFCJUqUsC7LmzevqlatqoULF1qXBQUFydfX1/o6RYoUunv3riQpPDxc3bt3V+/evZUpU6YYH9uz8Pb2VooUKWLU9urVqzp48KB1vI8TEBCg1atX6+2331ayZMmsy1u3bi0vLy+b9yM6Bw8e1MqVK9WrVy+lTJlS9+7de+y/9dixY5UuXTp1795dxpjH3sIWmz4B4GXwImR3bDIspubOnats2bKpTJky1mVkd9xmdyRjjAICAmSMibM+AeBFRnb/50kZQnYnzHl3gQIFlCpVKptl7u7uql27ts6ePavAwEC7YwJii0I6EiV/f381bNhQbm5uatGihY4cOaLt27fbtLl9+7bKly+vCRMmqEaNGho3bpw6duyogwcP6uzZs5KksLAw1a1bV4MGDVLx4sU1atQode/eXbdu3dK+ffusffXr10/58uWLMnXH4/z5559KkiSJvLy85Ofnp3HjxsVou9u3b+v27dtRguBR4eHh2rt3r1599dUo60qWLKljx45ZA6NEiRKaNm2a/vnnH23evFnz5s1TyZIlJUkzZszQ1atX1atXrxiNLybu3bunq1ev2vwEBwc/VV8TJ05Uvnz5nng79T///KPQ0NAo74ebm5uKFi2qXbt22d0+cv60tGnTqmrVqvL09JSnp6def/11nTx50qbtmjVrVKJECY0fP16pU6eWt7e30qdPr4kTJz51nwDwMnhRs/tZ7Nq1SwcOHFDLli1tlpPdcZvdkbJnzy4fHx95e3vr7bff1qVLl565TwB4kZHd/3lShpDdCXPe/TgXL15UkiRJlCRJkhi1B2IsIS+HB57Gjh07jCSzevVqY4wx4eHhJlOmTKZ79+427T7//HPrbWiPCg8PN8YY89133xlJ0d4uFNnGmIjbsiSZEydOPHF89erVM8OHDzdLly41M2bMMOXLlzeSTO/evZ+4beTt3WvWrLHb7sqVK0aSGTx4cJR1kyZNMpLMwYMHjTHGnDlzxhQoUMBIMpJM+fLlTWBgoLl586ZJnTq1mT9//hPHFVOR+3j0Z+bMmdG2f9ItZpG30T3pFrTIfjZs2BBlXZMmTUy6dOnsbt+tWzcjyaRMmdLUqlXLLFiwwHz99dfGy8vL5MiRw9y5c8cYY8z169et7by8vMzXX39tFixYYGrVqmUkmalTp8a6TwB4GbyI2R0Xt0n37NnTSDL79++3WU52x112G2PM2LFjTZcuXYy/v79ZvHix6d69u3FxcTG5cuWyuTWd7AaA/5DdEWKaIWR3wpx3R+fIkSPGw8PDtGrVym474GlQSEei06NHD5M2bVoTGhpqXdazZ88oywoUKGCKFClit686deqYVKlSmfv378fXcE14eLipWbOmcXFxsTsf2vr1642Li4tp2rTpE/s8ffq0kWSGDx8eZd2MGTOMJLNr1y7rspCQELNr1y7z77//Wudl79GjhylXrpwxJmK+tpIlS5pMmTKZrl27muDg4FgeZQRJpkGDBmb16tU2P+fPn4+2fVzN1fb9998bSWbr1q1R1rVq1cr4+PjY3f6dd94xkkyBAgVs5q2fN2+ekWSmT59ujPnvfZdk84dQWFiYyZ8/v80cbDHtEwBeBi9idj9rhoWFhZmMGTOaYsWKRbue7Paxu/2z5qy/v7+RZIYNGxZnfQLAi4TsfrzoMsQYsjshzrsfdefOHVO0aFHj6+trzp07F8sjBJ6MqV2QqISFhWn+/PmqXLmyTpw4oaNHj+ro0aMqVaqULl26pDVr1ljbHjt2TAULFrTb37Fjx5QnTx65uLjE25gtFot69Oih0NBQrVu3Lto2Bw8e1JtvvqmCBQvq22+/fWKfkU/jju7WrXv37tm0kSRXV1cVLVpU+fPnl5OTkw4ePKjJkydr3Lhxun79uurUqaM33nhDixYt0urVq/Xll18+xZFGyJQpk6pVq2bzkz59+qfuLyae9H48/F7Y275p06Y289Y3adJELi4u2rRpk007V1dXNW7c2NrOyclJzZo109mzZ3X69OlY9QkAL7oXNbuf1fr163Xu3Dm99dZb0a4nu+Mmux+nZcuWSpcunfU287joEwBeFGS3fdFliER2J8R598PCwsLUvHlz7d+/X4sXL1aGDBlieYTAk1FIR6Ly559/6sKFC5o/f75y5cpl/WnatKkkPfbhJwktc+bMkqTr169HWXfmzBnVqFFDPj4++vXXX+Xt7f3E/lKkSCF3d3dduHAhyrrIZfZCo0ePHnr77bf1yiuvaMWKFUqRIoX69eun0qVLq3fv3g77Pj5O5B8Mj3s/nhSgkevTpk1rs9zZ2VkpU6bUjRs3JEW87x4eHkqZMqWcnZ1t2qZJk0aSrG1j2icAvOhexOyOC/7+/nJyclKLFi1i1J7sthUXOZs5c2abf1+yGwAikN0x29eT9kN224qP8+6HtW/fXr/88otmzZqlKlWqxPCogNiJv68DgXjg7++vNGnSaNKkSVHWLVmyRD/99JOmTp0qT09P5ciRw+bBJdHJkSOHtm7dqvv378vV1TW+hq3jx49LklKnTm2z/Nq1a6pRo4aCg4O1Zs2aGH+D7OTkpEKFCmnHjh1R1m3dulXZs2d/bEH+l19+0aZNm3TkyBFJ0vnz5232myFDhhg/3MVRFCxYUC4uLtqxY4f1jzsp4gnuu3fvtlkWneLFi0tSlOMOCQnR1atXrf9uTk5OKlq0qLZv366QkBC5ublZ254/f17Sf//GMe0TAF50L1p2x4Xg4GD9+OOPqlSpUoyuliK7o3rWnDXG6OTJkypWrFic9QkALwqy277oMuRRZHdU8XHeHalXr16aOXOmxo4dG+OLFICnktBzywAxdffuXePt7W3eeeedaNdv3LjRZg6tuHzoyfnz582BAwdMSEiI3TFeu3bNZr44YyLmSStbtqxxc3MzFy5csC6/ffu2KVmypPH29jY7duyw2++pU6fMgQMHbJZ99dVXRpLZvn27ddnBgweNs7Oz6dOnT7T9BAcHm1y5cpkRI0ZYl82cOdOkTp3aOl/dmDFjjJ+fn93xPI4k07lz5xi3f9JcbVeuXDEHDhyI0cO9atWqZdKnT28CAgKsy7799lsjyaxcudK67M6dO+bAgQPmypUr1mX37t0zadKkMdmzZzdBQUHW5d98842RZBYuXGhdNmbMGCPJTJs2zbosKCjIZM+e3eTPn/+p+gSAF9WLlt0Pe1KGRZfdkZYsWWIkmRkzZtgdmzFktzHPnt2XL1+Osu/Ih7M//P8R2Q0AZPej2R3TDHkY2f38zruNMWbEiBFGkvnkk0+eOH7gWVFIR6Ixf/58I8ksXbo02vVhYWEmderUpl69esYYYwIDA03+/PmNs7Ozad++vZk6daoZOnSoKV26tNm9e7cxxpjQ0FBTqVIlI8k0b97cTJo0yYwYMcLUqFHDZj8xfXr4zJkzTY4cOUyfPn2s+ytYsKCRZIYOHWrTtkGDBkaSeeedd8ycOXNsfn766SebthUrVjSPfu8VEBBgcuTIYdKkSWNGjBhhxowZYzJnzmwyZMgQbdgbExEwuXLlsnmoyaVLl0ySJElMkyZNzJgxY0yKFCnsPuncnpgG+hdffGG++OIL07x5c+t7ELnsYTF9ergxxvzvf/8z7u7uplixYmbKlCnm008/NR4eHqZGjRo27dauXWskmQEDBtgsnz17tpFkSpQoYcaPH28+/vhj4+rqasqXL2/zR9rdu3dNgQIFjKurq/n444/N+PHjTYkSJYyzs7P59ddfn6pPAHhRvWjZbUzMMyy67I7UqFEj4+7ubm7evGl3bMaQ3cY8e3Z7enqatm3bmlGjRplJkyaZFi1aGIvFYooWLRqlaEB2A3jZkd222R2bDIlEdj+/8+7IixNy5coVpa4yZ84cc/HixSceExAbFNKRaNSrV894eHjY/Za0bdu2xtXV1Vy9etUYE/FNdZcuXUzGjBmNm5ubyZQpk2nTpo11vTERv6A//fRTky1bNuPq6mrSpUtnGjdubI4dO2ZtE9NA37Fjh6lXr551f15eXqZcuXLRXsGUNWtW65OoH/3JmjWrTdvHnYyfOXPGNG7c2CRLlsx4eXmZunXrmiNHjkQ7tosXLxpvb2+zbNmyKOtWrlxp8ubNa5InT25at24do2+ioxPTQH/ccT96jLEJdGMinoJepkwZ4+HhYVKnTm06d+5s8025MY8PdGMinhZepEgR4+7ubtKmTWu6dOkSZXtjIv4IatOmjUmRIoVxd3c3pUqVMr/99lu0Y4ppnwDwInrRstuYmGfY47L71q1bxsPDwzRs2NDuuIwhuyM9a3a/9957Jn/+/Mbb29u4urqanDlzmj59+jw2j8luAC8zstt2WWwzhOyO8LzOuyPH/rifmB4TEFMWY4wRAAAAAAAAAACIllNCDwAAAAAAAAAAAEdGIR0AAAAAAAAAADsopAMAAAAAAAAAYAeFdAAAAAAAAAAA7KCQDgAAAAAAAACAHRTSAQAAAAAAAACwwyWhBxDp4q37CT0EAA9kq9QjoYcA4IGgXRMTegiPRXYDjoPsBhyHI2f35UCyG3AUWSuQ3YCjiGl2c0U6AAAAAAAAAAB2UEgHAAAAAAAAAMAOCukAAAAAAAAAANhBIR0AAAAAAAAAADsopAMAAAAAAAAAYAeFdAAAAAAAAAAA7KCQDgAAAAAAAACAHRTSAQAAAAAAAACwg0I6AAAAAAAAAAB2UEgHAAAAAAAAAMAOCukAAAAAAAAAANhBIR0AAAAAAAAAADsopAMAAAAAAAAAYAeFdAAAAAAAAAAA7KCQDgAAAAAAAACAHRTSAQAAAAAAAACwg0I6AAAAAAAAAAB2UEgHAAAAAAAAAMAOCukAAAAAAAAAANhBIR0AAAAAAAAAADsopAMAAAAAAAAAYAeFdAAAAAAAAAAA7KCQDgAAAAAAAACAHRTSAQAAAAAAAACwg0I6AAAAAAAAAAB2UEgHAAAAAAAAAMAOCukAAAAAAAAAANhBIR0AAAAAAAAAADsopAMAAAAAAAAAYAeFdAAAAAAAAAAA7KCQDgAAAAAAAACAHRTSAQAAAAAAAACwg0I6AAAAAAAAAAB2UEgHAAAAAAAAAMAOCukAAAAAAAAAANhBIR0AAAAAAAAAADsopAMAAAAAAAAAYAeFdAAAAAAAAAAA7KCQDgAAAAAAAACAHRTSAQAAAAAAAACwg0I6AAAAAAAAAAB2UEgHAAAAAAAAAMAOCukAAAAAAAAAANgR60L6pUuX1KpVK2XIkEEuLi5ydna2+QEAAI6F7AYAIHEhuwEAcDwusd2gbdu2On36tPr376/06dPLYrHEx7gAAEAcIbsBAEhcyG4AABxPrAvpf//9t/766y8VLVo0HoYDAADiGtkNAEDiQnYDAOB4Yj21S+bMmWWMiY+xAACAeEB2AwCQuJDdAAA4nlgX0seOHau+ffvq5MmT8TAcAAAQ18huAAASF7IbAADHE+upXZo1a6a7d+8qR44cSpIkiVxdXW3WX79+Pc4GBwAAnh3ZDQBA4kJ2AwDgeGJdSB87dmw8DAMAAMQXshsAgMSF7AYAwPHEupDepk2b+BgHAACIJ2Q3AACJC9kNAIDjiXUhPSAgINrlFotF7u7ucnNze+ZBAQCAuEN2AwCQuJDdAAA4nlgX0pMnTy6LxfLY9ZkyZVLbtm01YMAAOTnF+lmmAAAgjpHdAAAkLmQ3AACOJ9aF9FmzZunTTz9V27ZtVbJkSUnStm3bNHv2bH322We6cuWKRo4cKXd3d33yySdxPmAAABA7ZDcAAIkL2Q0AgOOJdSF99uzZGjVqlJo2bWpdVq9ePRUqVEjffPON1qxZoyxZsujLL78k0AEAcABkNwAAiQvZDQCA44n1PWCbNm1SsWLFoiwvVqyYNm/eLEkqV66cTp8+/eyjAwAAz4zsBgAgcSG7AQBwPLEupGfOnFkzZsyIsnzGjBnKnDmzJOnatWvy9fV99tEBAIBnRnYDAJC4kN0AADieWE/tMnLkSDVp0kQrV65UiRIlJEk7duzQwYMHtXjxYknS9u3b1axZs7gdKQAAeCpkNwAAiQvZDQCA47EYY0xsNzpx4oSmTZumQ4cOSZLy5MmjDh06yM/P76kHcvHW/afeFkDcylapR0IPAcADQbsmxkk/ZDfwYiO7AcfhyNl9OZDsBhxF1gpkN+AoYprdT1VIjw+cjAOOg5NxwHHE1cl4fCC7AcdBdgOOw5Gzm0I64DgopAOOI6bZHaOpXfbu3auCBQvKyclJe/futdu2cOHCMdoxAACIP2Q3AACJC9kNAIBji1EhvWjRorp48aLSpEmjokWLymKxKLoL2S0Wi8LCwuJ8kAAAIHbIbgAAEheyGwAAxxajQvqJEyeUOnVq638DAADHRnYDAJC4kN0AADi2GBXSs2bNGu1/AwAAx0R2AwCQuJDdAAA4thgV0h915MgRrV27VpcvX1Z4eLjNus8//zxOBgYAAOIO2Q0AQOJCdgMA4FhiXUifPn26PvjgA6VKlUrp0qWTxWKxrrNYLAQ6AAAOhuwGACBxIbsBAHA8sS6kDxkyRF9++aX69OkTH+MBAABxjOwGACBxIbsBAHA8TrHd4MaNG2rSpEl8jAUAAMQDshsAgMSF7AYAwPHEupDepEkTrVq1Kj7GAgAA4gHZDQBA4kJ2AwDgeGI9tUvOnDnVv39/bdmyRYUKFZKrq6vN+m7dusXZ4AAAwLMjuwEASFzIbgAAHI/FGGNis0G2bNke35nFouPHjz/VQC7euv9U2wGIe9kq9UjoIQB4IGjXxGfug+wGXnxkN+A4HDm7LweS3YCjyFqB7AYcRUyzO9ZXpJ84cSLWgwEAAAmH7AYAIHEhuwEAcDyxniM9UkhIiA4dOqTQ0NC4HA8AAIgnZDcAAIkL2Q0AgOOI9RXpd+/eVdeuXTV79mxJ0uHDh5U9e3Z17dpVGTNmVN++feN8kIgfe3bu0LwfZurwwf26dvWKhowYp/KVqkqSQkPv69spE7Rl01+6cO6sknp5qXiJ0urQpYdSpU5j7aNfzy46evigbt64Li/vZCpesrQ6dvnI2iY4OFijvxqsQwf36/TJ43qtbEV9OXL8Y8f0z56d6t6xnbJlz6kZ/j/G7xsAJKCyr+RQj9bV9Er+LEqf2kdNe0zT8nV7o207/tPmat+4nHp9vVgT566zLl80toOK5M6o1Cm8dSPgrtZuPaTPxv+sC1du2Wz/YauqeqdRWWVJ76trN+/om4V/acSM363rm7/+qnq0raacmdPo1u0grdq4X5+MXarrt+5Y2/h4eWpgl3pqUKWIUvgk0ekLN9Rr5GL9/vf+uH1jEC/I7heHveyOdPLEMX0zcYz27NyhsLAwZc2WXV8MH6u06dJLks6dPa3J40bqnz27dP9+iEqWLqfuH/dTipSpouwvJCREH7RroaNHDunbHxYrV+681nXGGC3wn6XlPy3WpYvn5ZPcV280aqZW73SI3zcBSCBPyu5PO9RWk5qvKFM6X4XcD9OuA6c1cOJybd93ytqmaN5MGtL9DRUvkEVhYUZL1+xWn1E/6k5QiLVN5nS+GvdJM1V8NbduBwXLf/lW9Z+wTGFh4ZKk8sVzadW33aOMz69aP126Fhhl+cftquuLbg000X+teo3k7+vEgux+cezeuUPz5szUoQMR2f3lyHGq8NB59/TJE7Rl4186/+C8+9WSpdWxq+159/czvtHmjRt05NAhubq6auW6zVH2M/brofpnz26dOHZEWbNl18y5tp/3C+fPqWn9mlG2mzrTXwUKFYnjowYcw/M67672Wj7171hb+XKk172Q+9q485j6jFqi0xeuW9s86bz77XqlNH1wK5sx3Qu+L9/STAHkSGJ9RXq/fv20Z88erVu3Th4eHtbl1apV04IFC+J0cIhfQfeClDNXHn3Y69Mo6+7du6fDh/ar9TsdNH3OQn0xfKzOnD6pT3p2sWlXrHhJDRw6SnMW/aIvho/R+bNn9Hnf/z7k4eFhcnN3V6Nmb6l4idJ2xxMYGKChAz/RK6+WipsDBBxYUk93/XP4nD4cZv/3Zv3KhVWykJ/OX74ZZd2G7Yf1dp/vVOTNwWrZ61tlz5xKc79+16bNqN6N1fbN19RvzE8q8uYQNf7wG+146IT+tSLZ9e0XrTV76Wa90vhLvd17hl4tmFWT+7ewtnF1cdaKqV2UNUMKvdVrhgq/8YU6fTFX5y/bFuzhuMjuF4e97JYiiuRd27dWlqzZNHbqTH0390e1ebej3NzcIrYPuquPu74vi8WiMZNnaOL0OQq9f1/9enZReHh4lP6mThillA+dyD9s/KhhWvHzEnXq/rHmLFyuoSMnKG+BQnF3sICDeVJ2Hz11WT2GL9KrTYaqarvROnX+upZP7qJUvl6SpPSpfbRialcdO3NFFVqNVIPOk5Q/Rzqbk2YnJ4uWjP9Abq4uqtx2lNp/Pkdv1y+lzz+oE2V/hRoMll+1ftafy9dvR2lTPH8WvduorPYePhtH7wKeF7L7xXEvKCK7P+rzmPPug/vV5r0OmvHDQn359VidPnVSfT+yPe++H3pflarW1BuNm9ndV536b6pK9Vp224yZ/K2W/rbO+pMnX/7YHxSQSDyP8+6sGVJq0Zj3tW77YZVq/pXqd5qklMmTav6o9tY2MTnvlqRbgUE22Z6n9ufP9gYgzsX6ivSlS5dqwYIFKl26tCwWi3V5gQIFdOzYsTgdHOJX6TLlVbpM+WjXeXl5a/TEb22Wde/1iTq2baFLFy9Yr2pr2rK1dX269Bn0Vpv39GmvbgoNvS8XF1d5eiZRz74RH/x9e3bpdmDUq2Qijf5qsKrVrCMnJyf9vf7PZz08wKGt2rhfqzbav5o7Q2ofje7TRPU6TdJPEz6Isn6C/1rrf5++cEMjZ67WwtHt5eLipNDQcOXJllbtG5dX8SZf6sipy5KkU+ev2fRRqnA2nTp/TZPnrbeun/HjRvVsW83aps0br8k3WRJVajtKoaHhD/Z3XUg8yO4Xh73slqRvp4xXqbLl9UG3ntZlGTNlsf73vj27dPHCeX07Z7GSekUU9/oN/FJ1q5bRzh1b9WrJ16xtt2z6S9u3btIXX43V1k1/2ezn5Ilj+vnHhZo1/ydlyRrxQLz0GTPFyTECjupJ2b3gtx02r/uMWqJ2b5ZRwVwZtG7bYb1evqDuh4bpw2ELZYyRJHX9coF2LPpE2TOn0vEzV1XttXzKlz2d6nScoMvXA7X38DkNnrxCQ7o10JCpv+p+aJi1/yvXA3XrdtBjx5PU000zh7ZVpy/mqe979gtrcDxk94ujdNnyKl328efdYybbnnf36P2J3m9je979boeIwvqvy5c+dj8f9vpEknTzxnUdO3r4se18fJIrZaqod6EBL6Lncd79Sv7McnZy0sBJv1jzfez3a7RozPvWNjE575YkIxPt3WVwHLG+Iv3KlStKkybqlUl37tyxCXi8eO7cvi2LxSIvL+9o1wfcuqXVv/2igoWLysXFNVZ9/7r8J50/d1Zt3ov6Swt4GVksFs0Y0lpjZq/RgeMXn9jeN1kSNX/9VW3Zc8Ja7K5ToZBOnLuq2hUK6sAvA3VwxSBN/rylfJMlsW63de8JZUrnq5rlIq5ESZPCW29WK6rfHpqypU7FQtq694TG9m2mk38M1Y5Fn6jXOzXk5MTv/MSC7H45hIeHa/PGDcqcxU8fd31fDWpWUMd2LfTXujXWNiH378tiscj1wRXqkuTm5i4nJyf9s3unddn1a1c1cuhAfTpwmNwfuhIy0qa/1itDxkza/Pd6NWtQU80a1NCIIZ8r4BZ3qgBSxN1c7zYsq5uBd/XP4XOSJHc3F92/H2Y9yZakoOCIKV3KFM0hKeIL7n1Hz+vy9f9OoldvOiAfb0/lz5HeZh9bF/TV8VVf6pcpXfRakexRxjC2XzP99tc+rd16KM6PD/GP7H55Pem8+1n1/aiL6lWvoE7vttLf69c+eQPgBRYX5907959RuAlX6wal5eRkUTIvD7WsU1J/bj1kbROT825J8vJ016FfB+vIyi+0cMz7ypc9XRwfMZ5VrAvpr776qlasWGF9HRni3377rV577bXHbWYjODhYAQEBNj/BwcGxHQqeo+DgYH0zcYyq1qhtvYIt0tQJo1WzQgnVq15Wly5e1JdfT4hV32dPn9K0SWP02aCv5OIS65skgBdSz3bVFRoWrknz1tltN6RbA13dNErn149Q5vQp1KTHNOs6v0yplCV9CjWsVkzv9Z+j9p//oGL5MtvchrZ5z3G1+2S25nz1jgK2jdOpNcN06/Y9ffjVf7e+ZcuYUm9WKyZnZ4ve7DpFX03/Td1bVeXqtkSE7H453Lh+XUF372ru7Bkq+Vo5jZwwTeUrVVX/Ph9q987tkqQCBQvLw8NT30wcrXv3ghQUdFeTx41UWFiYrl27Kili7vNhgz9T/TebKm/+gtHu68K5M7p08bzWrVmlTwYOVd/Ph+jQwf0207sBL6PXyxfUlY2jdHPrGHV9u7Lqdpyoazcj5j5dt+2Q0qZMph6tq8rVxVnJvT01pFsDSVK61D6SpLQpk+nyI1eiXb4eELEuVTJJ0sWrt9RlyDy1+Phbtez1rc5evKHfp3dX0bz/3RXSpGZxFc2bWf0nLIv3Y0b8ILtfTsHBwZoyYYyq1Yx63v2sPJMkUecPe2nw8NEaMXaSChd9RZ983I1iOl5qcXHefer8NdXtNEmDutTTra1jdemvkcqYNrne7v2dtU1MzruPnLqsDoP81eTDb9Tus9lysli0dlZPZUyTPK4PG88g1oX0oUOH6pNPPtEHH3yg0NBQjRs3TjVq1NDMmTP15ZdfxqiPYcOGycfHx+ZnwujhsR48no/Q0Psa+ElPGWP0UZ/+UdY3b9VO385ZpJETpsnZ2UlDB/WzudLGnrCwMA3u31vt2ndW5qx+cTxyIHEqli+zOreopPcH/PDEtmO+/0Olmw9XnY4TFRYWrm+/eGieVYtFHu6uerf/HG3cdUx//e+IPhjkr0ol8yhX1ogrnPJmT6eRvRtr2LSVKvPWcNXrNElZ06fQhE+b/9ePk5OuXA9U5y/madeBM1q8aqdGzPhd7zUuF/cHj3hBdr8cjIm44qVshcpq2rK1cuXOq7favKfXylXUz0sWSpKS+6bQoGGjtOmvdapVsaTqVHlNt28HKHfe/NYizY8L/RV0947eavveY/cVboxCQkL0yYChKlKsuIoVL6k+nw3Wrv9t0+lTJ+L9WAFHtX77YZVqPkyV247Wqk379cOId5T6wRzpB45fVPvP56hbq6q6vnm0Tv4xVCfPXdPFqwEy0Tyj4HGOnLqsGT9u1K4DZ7Rlzwl1HOSvLXuPq+tbVSRJmdIm19e9Gqndp7MUHBIaL8eJ+Bdf2T1+FNntqEJD72tA34jz7p59o553P6vkyX3V/O02KlCwsPIVKKSOXXuoxut1NW/OzDjfF5AYxNV5d9qU3prcv6X8l29Vube/VrV3xyjkfpjmjvzvAraYnHdv3XtCc3/Zpr2Hz+nv/x1V84+n6+qN23q3cdm4PXA8k1hf/luuXDnt3r1bX331lQoVKqRVq1bplVde0ebNm1WoUMweMNWvXz999NFHNstu3It1TR/PQWjofQ3o11OXLpzXmMnfRfutePLkvkqe3FeZs/opq192NalXTf/+s0cFCxd9Yv93797RoQP/6ujhgxo3cqikiFvTjTGq8loRjRw/Ta+U4OGjeLmULZZDaVJ46fCvg63LXFyc9dVHDdXlrcrKW2eAdfm1m3d07eYdHT19WYdOXNTR34eoVOFs2rr3hC5evaX798N09PRla/uDJy5JkjKnS6Ejpy6rV7sa2rz7mMZ8HzH1w74j53U3KFhrZn6kQZN+0cWrARH9hIYpPNw81M9FpU/tI1cXZ5v5WuGYyO6Xg09yXzk7u8gvWw6b5Vn9suufPf9N21KidFnN++k33bx5Q87OzvL2TqY3a1VUhgcPJ9u1fZv+/WePqpd7xaafDm2aqVrNOvpk4FClTJVKzs4uNl+CZ/WLmFri0sUL1nnTgZfN3XshOn7mqo6fuapt/5zUPz9/rjZvltHI71ZJiphHfcFvO5QmhbfuBAXLGKnb21V04mzEM0wuXQvQqwWz2vSZJkXEleiXrgY8dr879p1SmWIRn/1i+bIobcpk2jy3j3W9i4uzyr2SQx2bVZBPqQ9tMh2OKb6y+1YI2e2IQkPv6/O+PXXx4nmNmxL9eXd8yF+wsLZv3fxc9gU4mrg67+7QrIICbgfp03E/W9u/8+lsHf19iEoW8tO2f07G6Lz7UaGh4dpz6IxyZE4dj+8CYuup5tHIkSOHpk+f/tQ7dXd3l7u7u82yu+b+U/eH+BFZRD935rTGTvlOPsmTP3GbyCvR798PidE+kib10sx5P9ksW7p4vnbt2KZBX41W+gwZYz1uILGbu2K7/nxkPtPlkztr7opt+v7nLY/dLnLOcjfXiF/tm3cfl6urs7JlSqUTZyOmbIi8Ej3yYaFJPN0U+kghPOzByXXk1ambdx9Xs9dflcVisX7Gc2VJowtXblFET0TI7hefq6ur8uYvoNOnba8IP3P6pNKmyxClffLkvpKkndu36saN6ypbobIkqdvH/fTuB12t7a5duayPu3XQgC9HKl+BiOJNocLFFBYWqnNnT1sfZnrm9ElJUrpo9gW8rJwsFrm7Rj3lipwDvXWD0roXcl9rthyUFHE1Wp93ayq1r5eu3LgtSapaOq9uBQbZnbu1cJ5Mungl4hkFa7cdUvHGtlcsTxv0tg6duKRRs1ZTRE9E4iO77wWS3Y4msoh+9vRpjfsmZufdceXI4YNKmYoiHV5OcXXencTDLUq2hj240yyybUzOu6PbT4GcGfT7Ex6WiufrmSakNsZo7dq1CgoKUpkyZeTr6xtX48JzcPfuXZ07e9r6+sL5czpy+KCSJfNRylSp9Hnfj3T44H59NXqSwsLCde1qRCEumY+PXF1dtX/fXh3cv0+Fir4ib+9kOn/2jGZ8M0EZM2VWgUJFrf2ePH5M90PvKyDglu7evaMjhyNOFHLlzisnJydlz5HLZly+vink5uYWZTnwIknq6WbzzbJfxpQqnDujbgTc1ZmLN3T91h2b9vdDw3TpaoCOnIq4urxEwawqXiCrNu06ppuBd5UtU2oN6FRHx05f0da9EUW0P7ce0s79p/XNwLfU6+sf5eRk0di+TfXH5gPWq9RXrP9Hk/u3VPsm5bR60wGlT+Wjr3s10vZ/TurCgxPy6Yv+UsdmFTSqd2NNnrdeObOkVq93a1ifOI7EhexO3Oxld9p06dX87XYa9OnHKlLsVRUrXlLbNv+tzX+v19gp/922/evyn5TVL7uS+/rq33/2aMKor9SkRWvrVeRp09k+0NDTM+IBxRkyZVaatBEPPCpe8jXlzptfw7/4XF0+6iMTHq6xX3+pV0u9xlRteGHZy+5rN++oz3s1tWL9P7p49ZZSJvdSh6YVlCFNci1Z/d8dIR2bVdCWPcd1+26IqpbOq6EfvqH+E37WrdtBkqQ/Nh/QgeMXNWNIG306bqnSpkymAZ3r6puFGxRyP2Kali4tK+nk+Wvaf+yCPNxc1e7NMqpUIrfqdpooSbp9N1j7j12wGfudoBBdv3UnynIkDmR34nb37l2dO/NQdp87pyOHDiqZT8R5d//eH+nwof0aPmaSwqM575Yi7vYKuHVLly5eUFh4mI4cijinzpg5i5Ikicjps2dOK+juXV2/dlXB94Ktbfyy55Crq6tW/vKzXF1clStvXknS+j//0K/LflLvzwY9t/cCeN6ex3n3yr/+Vde3Kqvf+7W08Lf/yTuJuwZ1qa9T569p98GzkmJ23t3v/Vratvekjp25ouTenurRppqypE+hmT9teh5vFWIoxoX0mzdvqnv37tq5c6dKly6tUaNGqXbt2tq0KeIfNE2aNFq1apUKFy4cb4NF3Dp0YJ8+/OAd6+tJY0dIkmrVaaC27Ttp44aIh468+3Zjm+3GTvlOxYqXlLuHhzas/UMzp03SvXtBSpEytUq+Vlat3+kgNzc3a/s+PT7QxQvnra/fe9Df+m374u3YAEf3Sv6sWvVtd+vrER83kiTNWbYlRnO03b13Xw2qFNFnHesoqaebLl69pVWbDmj49O+sJ9rGGDX+8BuN7tNEq2d8qDtBIVq1cb/6jl5i7eeH5VvlndRDHZtV1Fc9GurW7SCt23ZInz10W9rZSzdVv/NkjejZUNsX9tP5yzc1ae46jZq1Oq7eDsQTsvvFYy+7+w34UhUqV9NHfT+X/+xvNX7UMGXJ4qfBX41R4aL/TdNy5tRJTZ80VgEBt5QufUa93e59NW3ZOlbjcHJy0rBREzVu5FB169BGnh6eKlmmvDp37xU3Bwo4IHvZ3fXL+crjl1Zv1yullMmT6vqtu9rx7ylVe2eMzZXkrxbMqs861pFXEjcdOnlJXb6cp3krtlvXh4cbNeo+ReM+aa51s3rqzr1g+S/fpsFT/nvopJuri77q0VAZ0vjo7r372nfknGp3nKANO448h3cB8Y3sfvEc2r9P3Tr+l90TxzzI7roN9M77nfT3g/Pudi1tz7vHT/1OxV4tKUn6dupE/fbLQ9NGvNU4SpvhX3yu3Tt3RGmzcNnv1ju9Z82YqksXLsjZ2VlZ/LJp4NCRqlytRpweL+BInsd59/rth9X2k9nq0aaaPmpTXXfvhWjr3hOq33my7gVH3AUUk/NuX+8kmvx5S6VN6a0bAUHadeC0KrcdrYN27kjD82cxMXwq5HvvvacNGzaoTZs2Wr58uZycnGSM0dixY+Xk5KTevXvLy8tLy5cvf6qBXLzFLWaAo8hWqUdCDwHAA0G7Jj71tmQ38PIguwHH4cjZfZmpXQCHkbUC2Q04iphmd4wL6RkzZtTcuXNVsWJFnTt3TpkzZ9aff/6pSpUqSZK2bdum+vXr6+LFp/umhJNxwHFwMg44jmc5GSe7gZcH2Q04DkfObgrpgOOgkA44jphmd4wf2X3p0iXlzp1bUkS4e3h4KHPmzNb1WbJk0ZUrV2I5TAAAEF/IbgAAEheyGwAAxxXjQnp4eLicnZ2tr52dnW2eLPu4p8wCAICEQXYDAJC4kN0AADiuGD9sVJK+/fZbeXl5SZJCQ0M1a9YspUqVSpIUGBgY96MDAADPhOwGACBxIbsBAHBMMZ4j3c/PL0bffp84ceKpBsI8q4DjYJ5VwHE8yzyrZDfw8iC7AcfhyNnNHOmA42COdMBxxDS7Y3xF+smTJ592LAAAIAGQ3QAAJC5kNwAAjivGc6QDAAAAAAAAAPAyopAOAAAAAAAAAIAdFNIBAAAAAAAAALCDQjoAAAAAAAAAAHZQSAcAAAAAAAAAwI5YF9KdnZ11+fLlKMuvXbsmZ2fnOBkUAACIO2Q3AACJC9kNAIDjiXUh3RgT7fLg4GC5ubk984AAAEDcIrsBAEhcyG4AAByPS0wbjh8/XpJksVj07bffysvLy7ouLCxMGzZsUN68eeN+hAAA4KmQ3QAAJC5kNwAAjivGhfQxY8ZIivhmfOrUqTa3k7m5ucnPz09Tp06N+xECAICnQnYDAJC4kN0AADiuGBfST5w4IUmqXLmylixZIl9f33gbFAAAeHZkNwAAiQvZDQCA44pxIT3S2rVr42McAAAgnpDdAAAkLmQ3AACOJ9YPG23UqJGGDx8eZfmIESPUpEmTOBkUAACIO2Q3AACJC9kNAIDjiXUhfcOGDapdu3aU5a+//ro2bNgQJ4MCAABxh+wGACBxIbsBAHA8sS6k3759W25ublGWu7q6KiAgIE4GBQAA4g7ZDQBA4kJ2AwDgeGJdSC9UqJAWLFgQZfn8+fOVP3/+OBkUAACIO2Q3AACJC9kNAIDjifXDRvv376+GDRvq2LFjqlKliiRpzZo1mjdvnhYtWhTnAwQAAM+G7AYAIHEhuwEAcDyxLqTXq1dPS5cu1dChQ7V48WJ5enqqcOHC+uOPP1SxYsX4GCMAAHgGZDcAAIkL2Q0AgOOxGGNMXHW2b98+FSxY8Km2vXjrflwNA8AzylapR0IPAcADQbsmxmv/ZDfwYiC7AcfhyNl9OZDsBhxF1gpkN+AoYprdsZ4j/VGBgYGaNm2aSpYsqSJFijxrdwAAIJ6R3QAAJC5kNwAACe+pC+kbNmxQ69atlT59eo0cOVJVqlTRli1b4nJsAAAgDpHdAAAkLmQ3AACOI1ZzpF+8eFGzZs3SjBkzFBAQoKZNmyo4OFhLly7lyeEAADggshsAgMSF7AYAwDHF+Ir0evXqKU+ePNq7d6/Gjh2r8+fPa8KECfE5NgAA8AzIbgAAEheyGwAAxxXjK9JXrlypbt266YMPPlCuXLnic0wAACAOkN0AACQuZDcAAI4rxlek//333woMDFTx4sVVqlQpTZw4UVevXo3PsQEAgGdAdgMAkLiQ3QAAOK4YF9JLly6t6dOn68KFC+rQoYPmz5+vDBkyKDw8XKtXr1ZgYGB8jhMAAMQS2Q0AQOJCdgMA4LgsxhjztBsfOnRIM2bM0Jw5c3Tz5k1Vr15dy5Yte6q+Lt66/7TDABDHslXqkdBDAPBA0K6Jcdof2Q28mMhuwHE4cnZfDiS7AUeRtQLZDTiKmGZ3jK9Ij06ePHk0YsQInT17VvPmzXuWrgAAwHNAdgMAkLiQ3QAAOIZnuiI9LnFVG+A4uKoNcBxxfVVbXCK7AcdBdgOOw5GzmyvSAcfBFemA43guV6QDAAAAAAAAAPCio5AOAAAAAAAAAIAdFNIBAAAAAAAAALCDQjoAAAAAAAAAAHZQSAcAAAAAAAAAwA4K6QAAAAAAAAAA2EEhHQAAAAAAAAAAOyikAwAAAAAAAABgB4V0AAAAAAAAAADsoJAOAAAAAAAAAIAdFNIBAAAAAAAAALCDQjoAAAAAAAAAAHZQSAcAAAAAAAAAwA4K6QAAAAAAAAAA2EEhHQAAAAAAAAAAOyikAwAAAAAAAABgB4V0AAAAAAAAAADsoJAOAAAAAAAAAIAdFNIBAAAAAAAAALCDQjoAAAAAAAAAAHZQSAcAAAAAAAAAwA4K6QAAAAAAAAAA2EEhHQAAAAAAAAAAOyikAwAAAAAAAABgB4V0AAAAAAAAAADsoJAOAAAAAAAAAIAdFNIBAAAAAAAAALCDQjoAAAAAAAAAAHZQSAcAAAAAAAAAwA4K6QAAAAAAAAAA2EEhHQAAAAAAAAAAOyikAwAAAAAAAABgB4V0AAAAAAAAAADsoJAOAAAAAAAAAIAdFNIBAAAAAAAAALCDQjoAAAAAAAAAAHZQSAcAAAAAAAAAwA4K6QAAAAAAAAAA2OGS0AOIlDypa0IPAQAAxALZDTgQC9fHAHiyZJ5kN+AwnJwTegQAYom/uAEAAAAAAAAAsINCOgAAAAAAAAAAdlBIBwAAAAAAAADADgrpAAAAAAAAAADYQSEdAAAAAAAAAAA7KKQDAAAAAAAAAGAHhXQAAAAAAAAAAOygkA4AAAAAAAAAgB0U0gEAAAAAAAAAsINCOgAAAAAAAAAAdlBIBwAAAAAAAADADgrpAAAAAAAAAADYQSEdAAAAAAAAAAA7KKQDAAAAAAAAAGAHhXQAAAAAAAAAAOygkA4AAAAAAAAAgB0U0gEAAAAAAAAAsINCOgAAAAAAAAAAdlBIBwAAAAAAAADADgrpAAAAAAAAAADYQSEdAAAAAAAAAAA7KKQDAAAAAAAAAGAHhXQAAAAAAAAAAOygkA4AAAAAAAAAgB0U0gEAAAAAAAAAsINCOgAAAAAAAAAAdlBIBwAAAAAAAADADgrpAAAAAAAAAADYQSEdAAAAAAAAAAA7KKQDAAAAAAAAAGAHhXQAAAAAAAAAAOygkA4AAAAAAAAAgB0U0gEAAAAAAAAAsINCOgAAAAAAAAAAdlBIBwAAAAAAAADADgrpAAAAAAAAAADYQSEdAAAAAAAAAAA7KKQDAAAAAAAAAGAHhXQAAAAAAAAAAOygkA4AAAAAAAAAgB0U0gEAAAAAAAAAsINCOgAAAAAAAAAAdlBIBwAAAAAAAADADgrpAAAAAAAAAADYQSEdAAAAAAAAAAA7KKQDAAAAAAAAAGDHMxXSjTEyxsTVWAAAQDwjuwEASFzIbgAAHMNTFdK///57FSpUSJ6envL09FThwoU1Z86cuB4bAACII2Q3AACJC9kNAIBjcYntBqNHj1b//v3VpUsXlS1bVpL0999/q2PHjrp69ap69OgR54MEAABPj+wGACBxIbsBAHA8FhPLe8SyZcumQYMGqXXr1jbLZ8+erYEDB+rEiRNPNZB7oU+1GYB44FuiS0IPAcADQbsmPnMfZDfw4vMt2S2hhwDggaCd45+5D7IbePH5luqe0EMA8EDQ/8bFqF2sp3a5cOGCypQpE2V5mTJldOHChdh2BwAA4hnZDQBA4kJ2AwDgeGJdSM+ZM6cWLlwYZfmCBQuUK1euOBkUAACIO2Q3AACJC9kNAIDjifUc6YMGDVKzZs20YcMG61xtGzdu1Jo1a6INegAAkLDIbgAAEheyGwAAxxPrK9IbNWqkrVu3KlWqVFq6dKmWLl2qVKlSadu2bXrzzTfjY4wAAOAZkN0AACQuZDcAAI4n1g8bjS889ARwHDxsFHAccfGw0fhCdgOOg4eNAo4jLh42Gl/IbsBx8LBRwHHE9GGjsZ7aRZLCw8N19OhRXb58WeHh4TbrKlSo8DRdAgCAeER2AwCQuJDdAAA4llgX0rds2aKWLVvq1KlTevRidovForCwsDgbHAAAeHZkNwAAiQvZDQCA44l1Ib1jx4569dVXtWLFCqVPn14WiyU+xgUAAOII2Q0AQOJCdgMA4HhiXUg/cuSIFi9erJw5c8bHeAAAQBwjuwEASFzIbgAAHI9TbDcoVaqUjh49Gh9jAQAA8YDsBgAgcSG7AQBwPDG6In3v3r3W/+7atat69uypixcvqlChQnJ1dbVpW7hw4bgdIQAAiDWyGwCAxIXsBgDAsVnMo08uiYaTk5MsFkuUh5xYO3mw7lkeenIv9Kk2AxAPfEt0SeghAHggaNfEp9qO7AZeLr4luyX0EAA8ELRz/FNtR3YDLxffUt0TeggAHgj637gYtYvRFeknTpx4psEAAIDni+wGACBxIbsBAHBsMSqkZ82a1frfGzZsUJkyZeTiYrtpaGioNm3aZNMWAAAkDLIbAIDEhewGAMCxxfpho5UrV9b169ejLL9165YqV64cJ4MCAABxh+wGACBxIbsBAHA8sS6kR87J9qhr164padKkcTIoAAAQd8huAAASF7IbAADHE6OpXSSpYcOGkiIecNK2bVu5u7tb14WFhWnv3r0qU6ZM3I8QAAA8FbIbAIDEhewGAMBxxbiQ7uPjIynim3Fvb295enpa17m5ual06dJq37593I8QAAA8FbIbAIDEhewGAMBxxbiQPnPmTEmSn5+fPv74Y24nAwDAwZHdAAAkLmQ3AACOy2KMMQk9CEm6F5rQIwAQybdEl4QeAoAHgnZNTOghPBbZDTgO35LdEnoIAB4I2jk+oYfwWGQ34Dh8S3VP6CEAeCDof+Ni1C7GV6Q/bPHixVq4cKFOnz6tkJAQm3U7d+58mi4BAEA8IrsBAEhcyG4AAByLU2w3GD9+vNq1a6e0adNq165dKlmypFKmTKnjx4/r9ddfj48xAgCAZ0B2AwCQuJDdAAA4nlgX0idPnqxp06ZpwoQJcnNzU+/evbV69Wp169ZNt27dio8xIgEtnD9Xjd+spzIlX1GZkq+oVctm+vuv9db1gwd+rjq1qqnkK4VVqVxpde/ygU4cPxZtXzdv3lD1KhVUpEAeBQQEPK9DABxS2VdyaPHYDjq+6ksF7ZqoepUKP7bt+E+bK2jXRHVpWclm+aKxHXT418G6sWWMjq/6UjO+aK30qX2ibP9hq6rau/Rz3dw6Rsd+H6Le79a0We/m6qKBnevp0K+DdXPrGB1cMUitG5S2aePj5akxfZvq+KovdXPrGO1d+rlqlsv/9G8Aniuy+8U1ZdIEFSmQx+anQd1a1vVXr1zRJ317qUqFsir1alE1a/ym/lj1u00fr1evEqWPGdOnWdefO3c2yvoiBfJo757dz+swAYcQkd3v6/jvXyho53jVq1TIZv2nHV7X7h8/1dWNX+v8uq+0YkpnlSiY1aZNziyptXB0e51ZM1SXNozQmhndVeHVXDZtRvVqpI3+vXRzy2htmdfb7piyZ06ly3+N0IX1X9ksf7teSQXtHG/zc2PzqGc4ejxvZPeL7X87tqtrp46qVqmcihTIoz/X/GGzvv8nfaPk7gfvv2vT5kn5/bDTp07ptRLFVK70q/F2TIAjKlsshxaPaa/jvw1W0P/GRcnuh43v11RB/xunLi0qWpdlSZ9CU/q30IFln+v6xq/178/99VmH1+Xq4hxtH9kzpdLlDcN1Yd2wx+6nSY1iCvrfOC0c9W6UdXn80mrR6Pd0cf1Xuvr3CP39fU9lTucbiyNGfIv11C6nT59WmTJlJEmenp4KDAyUJLVq1UqlS5fWxImOO5crYi9N2nTq3uNjZcmaVcYYLf95qbp36awFP/6knDlzKX/+AqpTt57SpU+vgFu3NGXSBHVs/65+XbVGzs62v1gG9v9UuXPn0eVLlxLoaADHkdTTXf8cPqfvf96sBaPff2y7+pULq2QhP52/fDPKug3bD+vrGb/r4tVbypAmuYb1eFNzv35XlduOtrYZ1buxqpbOq35jftK+I+eVwieJfJPZPrTqhxHvKG0Kb3Uc5K9jp68ofWofOVks1vWuLs5aMbWLLl8P1Fu9Zujc5ZvKkiGFbgUGPfsbgeeC7H6x5ciZS9O+nWl97fzQH/afftJHgQEBGjdxinx9ffXriuXq1fNDzV34o/Ll++/LsE5duqlR46bW10miebjdtBmzlCNHTutrn+TJ4/hIAMeW1MPtQXZv0YJR70VZf/TUZfUYvkgnzl2Tp7urur5VWcsndVLBBl/o6s3bkqQl4zro6Okrer3jRAXdu68ub1XSknHvq0D9wbp0LdDa1/c/b1GJgllVMFeGx47HxcVJ3w9to427jql0kWxR1t8KDFKRhkOsrx3jyViIKbL7xRYUdFd58uTRGw0b6aPu0T+fqmy58ho85L9inJubW5Q2Mcnv+/fvq2+vj/RK8Ve1Z/euOBg9kHgk9XyQ3cu2asHIqIXrSBHn3VmjnHfn8UsjJyeLugxdoGNnrqpAjvSa9FlzJfV0U7+xP9u0/S+Xj6t0Eb9o95MlfQoN+/AN/b3zaJR12TKl1JoZ3TX75y0a8s1KBdy5p/zZ0+te8P1YHzfiT6wL6enSpdP169eVNWtWZcmSRVu2bFGRIkV04sQJOchzSxGHKlWuYvO6a/ceWjh/nvbu2a2cOXOpcdNm1nUZM2ZSl24fqknDBjp/7pwyZ8liXbdw/lwFBgbq/Y6d9PdfG57b+AFHtWrjfq3auN9umwypfTS6TxPV6zRJP034IMr6Cf5rrf99+sINjZy5WgtHt5eLi5NCQ8OVJ1tatW9cXsWbfKkjpy5Lkk6dv2bTR/Uy+VS+eE7lrztQNwLuPujruk2bNm+8Jt9kSVSp7SiFhoZH2waOjex+sbk4OytV6tTRrtuza5c+/XyAChWOuOvl/Y6d9MP3s3Xg339tCulJkyZ9bB+RfHySP7EN8CJbtemAVm068Nj1C377n83rPqN/Urs3X1PB3Bm0btthpUyeVLmyptEHg+dq35HzkqT+45epY9Pyyp8jvbWQ3vPrHyVJqXy97BbSB3aqq0MnL2vttkPRFtKNjE1xHokL2f1iK1e+osqVr2i3jZub2xNzNyb5PXH8WPllz65SpV6jkI6XzpOyW3pw3t2rkep1maKfxtle5LZ680Gt3nzQ+vrkuWvKPedPtW9cNkohfeAHdXTo5CWt3X442kK6k5NFs4a00hffrFTZYtmV3NvTZv2gTnX1+8b9+nT8MuuyE2evPdoNElisp3apUqWKli2L+Edt166devTooerVq6tZs2Z6880343yAcBxhYWFa+esKBQXdVZEixaKsv3v3rn7+aYkyZsqkdOnSWZcfO3pU30yZrCFDh8vJKdb/ywEvJYvFohlDWmvM7DU6cPziE9v7Jkui5q+/qi17TliL3XUqFNKJc1dVu0JBHfhloA6uGKTJn7eUb7Ik1u3qVCyknftP66O21XTs9yHau/RzDevxpjzcXW3abN17QmP7NtPJP4Zqx6JP1OudGnJyskQZBxwT2f1iO3X6lKpVKqfaNauqX++eunD+vHVdkWLF9PtvK3Xr5k2Fh4dr5a8rFBwSrFdLlLTp47tvp6tCmVJq2ugNzfruW4WGhkbZT/cuH6hS+dfU5u0WWvfnmng/LiAxc3Vx1rsNy+hm4F39c/icJOnazTs6dOKSWtYpqSQebnJ2dtJ7jcrq0rUA7TpwJlb9VyyRSw2rFdWHXy16bBsvT3cdWjFQR34dpIWj2ytf9nSPbQvHQ3Zjx/ZtqlT+NdWvU1NDBg/QzZs3orR5Un5v3bJZq1f9pk8+G/C8hg0kKhaLRTO+eFtj5vwZo/NuSUrm5aHrDy5Ci2TN5eGPz+VP2tfSlRu3NfvnLdGOo1a5/Dpy+rKWTeyoU6uHaMPsHnanokHCiPUV6dOmTVN4eESRpnPnzkqZMqU2bdqk+vXrq0OHDjHqIzg4WMHBwTbLjLO73N3dYzscPAdHDh9Sq5bNFRISrCRJkmjM+EnKkfO/W7sXzPPXmFEjFRR0V37Zsumb6TPl+uC2s5CQEPXt9ZF6fNxL6TNk0NmzsTtJAF5WPdtVV2hYuCbNW2e33ZBuDdSxeQUl9XTX1r0n1LDbVOs6v0yplCV9CjWsVkzv9Z8jJycnjfi4oeZ+/a5e7zBBkpQtYyqVKZpD94JD1eyj6Urpm1Tj+jVTCp+k6jDwhwdtUqpSidyav3K73uw6RTkyp9bYfs3k6uKsodNWxtt7gLhDdr+4ChUurC++HCY/v2y6cuWKvpkySe1av6Uff16upEm99PWoserds4cqlC0lFxcXeXh4aMy4icqS9b95m1u81Ur58ueXj4+Pdu/epfFjR+vKlSvq1aefJClJkiTq2auvir7yipwsFv2xepU+7NZZY8dPUqUqVRPq0AGH9Hr5Avp+WFsl8XDVxasBqvvBZF27ece6vs4Hk7Rg9Hu68vcIhYcbXblxWw26TNXNWEyXlsIniaYPfEvt+s9R4J170bY5cuqyOgyKuPI9mZenPmxdRWtn9lDxJsN0Lprp4uB4yO6XW5ly5VW1WnVlzJRJZ86c0YSxo9WpQ3vNmbvAOoXqk/L75s0b+vzTfho6/Gt5eXkl5OEADqtn26oPzrvXP7mxIuZA/6B5BZur0a25/NkcBd4Jjna7MkWzq22D0irVckS069Ok8JJ3Ug993LaaBk3+VZ+NX64aZfJp/tfvqGaHifp7Z/TPIsTzF6tC+pYtW7R8+XKFhISoatWqqlWrlpo3b67mzZvHaqfDhg3ToEGDbJZ92n+APvt8YKz6wfPh55dNC39cqtu3A7V61e/q/0kfzZj1g7WYXrtufZUuU1ZXr1zR7Jkz1Kvnh5r9wzy5u7tr3JhRypYjh+rWa5DARwEkHsXyZVbnFpVUpuXwJ7Yd8/0fmrV0s7KkT6FPO7yub79oZS2mO1ks8nB31bv95+jo6YipXT4Y5K/N8/oqV9Y0OnLqspycLDLGqN2nsxRwO+JkvM+oJZr79bvqPmyB7gXfl5OTk65cD1TnL+YpPNxo14EzypAmuT5sXZVCeiJAdr/YHr4tPHeevCpUuIher15Zv/+2Ug0bNdGkCeMUGBigaTNmKXlyX6398w/17vmhZn7vr1y580iSWrdtZ9OHq6urhgwaoO49esrNzU2+vils2hQsVFhXrlzWrJkzKKQDj1i//YhKtRiuVMm91O7N1/TD8Haq0HqUrtyImCN9TN8munI9UNXeHaeg4Ptq+8Zr+nHs+yrXaqQuXg2I0T4m92+hBb/9TxvtnFRv3XtSW/eetL7esve4dv/4qd5tVEaDp/z6TMeI+Ed24/Xadaz/nSt3HuXOnUd1alXTju3bVKr0a5KenN+DBvTX63XqqvirJZ77+IHEoFjeTOrcvKLKvPV1jNpnSO2jZRM7askfuzXzp83W5ZM/ax6Ry7uiz2WvJO6aMfhtdRoy3+bL9YdFPqPsl/X7NGHuOknS3sPnVKqwn9o3Kksh3YHEuJC+ePFiNWvWTJ6ennJ1ddXo0aM1fPhwffzxx7Heab9+/fTRRx/ZLDPOfCvuqFzd3KxXruUvUFD/7vtH/j98r88HDpYkeXt7y9vbW1mz+qlw4SIqV6ak/vxjtV6vU1fbt27RkSOH9cqq3yXJOp9fpXKl9d77HdWpS7eEOSjAgZUtlkNpUnjp8K+DrctcXJz11UcN1eWtyspb579bM6/dvKNrN+/o6OnLOnTioo7+PkSlCmfT1r0ndPHqLd2/H2YtokvSwRMRD/vNnC6Fjpy6rItXA3T+8i1rET2izUU5OTkpY9rkOnb6SkQ/oWEKDzc2bdKn9pGri7Puh4bF59uBZ0B2v3ySJUumrFn9dOb0aZ05fVrz5/6gH3/+RTlz5pIk5cmbVzv/t0Pz5/mr/4DB0fZRqHARhYaG6vy5s/LLlj36NoWKaMumTfF2HEBidfdeiI6fuarjZ65q2z8n9c/Sz9Tmjdc0cuZqVSqZW7XLF1D6Sn2tV5J/+NUiVS2dR2/XLamRs/6I0T4qlsilOhUK6sNWEc8yslgscnZ2UuC2Mer85QJ9H80t46Gh4dpz8KxyZOY5B46O7EZ0MmXOLF9fX50+fcpaSH/Uo/m9fesWrV/7p76f9Z2kiHPx8PBwvVI4v/oPHKw3GzZ+nocAOBzrefeKgdZlLi7O+qrHG+rSsqLy1vvvb+X0qZLpt2+6aMueE+o8ZIFNP9ZcfruypIdyeetodf5ygXYfPCu/jCn145j21m0ip0kN3DpahRt9qbMXb+p+aFiU6WUOnbikMkWj/3scCSPGhfRhw4apffv2mjRpkpydnTVs2DANHTr0qQLd3T3q7WT3ok7FCQcVHh6u+yEh0a4zkmSMQh6sHzV2gu4F/1eg+3ffPxrw2Sea+b2/MmXOEm0fwMtu7ort+nPrIZtlyyd31twV26I9OY4UGcZurhG/2jfvPi5XV2dly5RKJ85elSTlyppG0n8PC928+7gaViumpJ5uuhMUYm0TFhauc5duWts0e/1VWSwW65dhubKk0YUrtyiiOziy++Vz984dnTlzRnXqp9a9exFTRThZbJ9P4uTkLBP++AfVHTp4QE5OTkqRIqXdNjx4FHgyJ4uT3N0icjmJR8TUh5HTdUQKDzeyxOK5I5XajpHzQ88dqlupkHq2qabK7cbo/GOmbXFysqhAzgz6/QkPOkfCI7sRnUsXL+rmzZtKnerx2ftofn/vv0Bh4f/9rb7uzzWaOWO6ZvvPV9o0aeN9zICjm/vrdv257bDNsuUTO2rurzv0/bKt1mUZUvvot2+6aNeBM3p/0NwoD3yu1HasnJ0fyuWKBSNy+Z2xOn/5poKC76t4069sthnYqba8knjo45FLrEX0//17WrkfnK9HypU1jU5fjPp8BCScGBfSDx06pAUL/puPq2fPnvr88891+fJlpUmT5glbI7EaN2aUypWvoHTp0+vunTv6dcUv2rF9m6ZMm6GzZ87o999+1WtlysrXN4UuXbqo776dJnd3D5WrEHGreeYstsXymzcifgFky55DyZIle+7HAziKpJ5uNleF+WVMqcK5M+pGwF2duXhD12/Z3vJ1PzRMl64G6MipiKvLSxTMquIFsmrTrmO6GXhX2TKl1oBOdXTs9BVt3XtCkvTn1kPauf+0vhn4lnp9/aOcnCwa27ep/th8wHqV+oKV29WvfS1NG/S2vpj6q1ImT6qhH76p2T9v1r3g+5Kk6Yv+UsdmFTSqd2NNnrdeObOkVq93a2hyDOeRQ8Ihu198o74eroqVKit9hgy6cvmypkyaIGdnJ71eu668vb2VJUtWfTHoc330cR8lT55cf/75h7Zs3qgJk7+RJO3ZvUv/7N2jEiVLK2nSpNqzZ5e+Hj5MderWVzIfH0nSsqU/ydXVVXnz5ZMkrfljtZb+9KMGDB6SYMcNJAR72X3t5h31ea+GVqzfp4tXbyllci91aFpeGdL4aMnqXZKkrXtP6EbAXX07+G0NnfabgoLv652GZeSXMaV+++tfa7/ZM6eSl6e70qb0lqe7qwrnzihJOnD8ou6HhunQg7vLIr2SP7PCTbj2H7tgXdavfS1t++ekjp25ouTenurRuqqypPe1uRUdjonsfjncvXNHp0+ftr4+d/asDh44IB8fH/n4+GjqlImqVr2mUqZKpbNnzmjMqK+VOUtWlSlXXlLM8jt7jhw2+9y/b5+cnJyUK1fu53egQAKLkt0ZHj3vtn1o6KPn3RlS++j3aV11+sJ19Rv7s1L7/ve8gUvXAiVJh04+OZcf/m9J1mejPLx8zJw/NWdYG/2965jWbz+iGmXyqXb5AqrZYeKzvAWIYzEupN+9e9em8Onm5iYPDw/dvn2bQH+BXb9+TZ/166MrVy7Ly9tbuXPn0ZRpM/RambK6fPmSdv5vh36YM1sBtwKUMlVKFS/+qr73n6eUKR9/FRsA6ZX8WbXq2+7W1yM+biRJmrNsi94f8MMTt797774aVCmizzrWUVJPN128ekurNh3Q8OnfKeR+xKVGxhg1/vAbje7TRKtnfKg7QSFatXG/+o5eYu3nTlCI6nwwUaP7NNHGH3rr+q07+nH1Tg2c9Iu1zdlLN1W/82SN6NlQ2xf20/nLNzVp7jqNmrU6rt4OxBOy+8V36dJF9e31kW7evCnfFClU7JXimjN3oVKkSCFJmjh1msaNHqVuXTrq7t27ypI5i74Y+pXKP/jC283NTb+t/FVTJ09USEiIMmbMpFat26pVm3Y2+5k2dbLOXzgvF2dn+WXLrhEjx6h6zVrP/XiBhPRK/ixaNf2/aQlH9GwoSZqzbKu6Dl2gPH5p9XbdkkqZ3EvXb93Rjn9Pq9q746y3aV+7eUcNukzRwC51tfKbrnJ1cdaB4xfUpMd0/XPkvLXfKf1bqMKruayvt87vI0nKU2eg9Y6yJ/FN5qnJ/ZsrbcpkuhFwV7sOnFHldmN18MTFJ2+MBEV2vxz+/Xef3mvX2vp65IhhkqT6Dd7Up58P1OFDh7Xs56UKDAhUmjRp9FqZsurctbvc3CLubIlpfgMvu1fyZ9GqaV2tr0f0fFOSNGf5Vr0/cO4Tt69SOo9yZkmtnFlS69hvttMiehbv/pitns6ytXvVdehC9WpXXaM+bqjDpy6rRe/vtGn38TjdD56NxTx6T8JjODk5aciQITZPe+7Tp4969eqlVKlSWZd16/Z0c15zixngOHxLdEnoIQB4IGjX01+BQHYDLw/fkjx3BnAUQTvHP/W2ZDfw8vAtFbeFWABPL+h/42LULsaFdD8/P1ks9ufus1gsOn786b4pIdABx0EhHXAcz1JIJ7uBlweFdMBxPEshnewGXh4U0gHHEdNCeoyndjl58uTTjgUAACQAshsAgMSF7AYAwHE5PbkJAAAAAAAAAAAvLwrpAAAAAAAAAADYQSEdAAAAAAAAAAA7KKQDAAAAAAAAAGAHhXQAAAAAAAAAAOyIdSHd2dlZly9fjrL82rVrcnZ2jpNBAQCAuEN2AwCQuJDdAAA4nlgX0o0x0S4PDg6Wm5vbMw8IAADELbIbAIDEhewGAMDxuMS04fjx4yVJFotF3377rby8vKzrwsLCtGHDBuXNmzfuRwgAAJ4K2Q0AQOJCdgMA4LhiXEgfM2aMpIhvxqdOnWpzO5mbm5v8/Pw0derUuB8hAAB4KmQ3AACJC9kNAIDjinEh/cSJE5KkypUra8mSJfL19Y23QQEAgGdHdgMAkLiQ3QAAOK4YF9IjrV27Nj7GAQAA4gnZDQBA4kJ2AwDgeGL9sNFGjRpp+PDhUZaPGDFCTZo0iZNBAQCAuEN2AwCQuJDdAAA4nlgX0jds2KDatWtHWf76669rw4YNcTIoAAAQd8huAAASF7IbAADHE+tC+u3bt+Xm5hZluaurqwICAuJkUAAAIO6Q3QAAJC5kNwAAjifWhfRChQppwYIFUZbPnz9f+fPnj5NBAQCAuEN2AwCQuJDdAAA4nlg/bLR///5q2LChjh07pipVqkiS1qxZo3nz5mnRokVxPkAAAPBsyG4AABIXshsAAMcT60J6vXr1tHTpUg0dOlSLFy+Wp6enChcurD/++EMVK1aMjzECAIBnQHYDAJC4kN0AADgeizHGxFVn+/btU8GCBZ9q23uhcTUKAM/Kt0SXhB4CgAeCdk2M1/7JbuDF4FuyW0IPAcADQTvHx2v/ZDfwYvAt1T2hhwDggaD/jYtRu1jPkf6owMBATZs2TSVLllSRIkWetTsAABDPyG4AABIXshsAgIT31IX0DRs2qHXr1kqfPr1GjhypKlWqaMuWLXE5NgAAEIfIbgAAEheyGwAAxxGrOdIvXryoWbNmacaMGQoICFDTpk0VHByspUuX8uRwAAAcENkNAEDiQnYDAOCYYnxFer169ZQnTx7t3btXY8eO1fnz5zVhwoT4HBsAAHgGZDcAAIkL2Q0AgOOK8RXpK1euVLdu3fTBBx8oV65c8TkmAAAQB8huAAASF7IbAADHFeMr0v/++28FBgaqePHiKlWqlCZOnKirV6/G59gAAMAzILsBAEhcyG4AABxXjAvppUuX1vTp03XhwgV16NBB8+fPV4YMGRQeHq7Vq1crMDAwPscJAABiiewGACBxIbsBAHBcFmOMedqNDx06pBkzZmjOnDm6efOmqlevrmXLlj1VX/dCn3YUAOKab4kuCT0EAA8E7ZoYp/2R3cCLybdkt4QeAoAHgnaOj9P+yG7gxeRbqntCDwHAA0H/GxejdjG+Ij06efLk0YgRI3T27FnNmzfvWboCAADPAdkNAEDiQnYDAOAYnumK9LjEN+OA4+CKdMBxxPUV6XGJ7AYcB1ekA44jrq9Ij0tkN+A4uCIdcBzP5Yp0AAAAAAAAAABedBTSAQAAAAAAAACwg0I6AAAAAAAAAAB2UEgHAAAAAAAAAMAOCukAAAAAAAAAANhBIR0AAAAAAAAAADsopAMAAAAAAAAAYAeFdAAAAAAAAAAA7KCQDgAAAAAAAACAHRTSAQAAAAAAAACwg0I6AAAAAAAAAAB2UEgHAAAAAAAAAMAOCukAAAAAAAAAANhBIR0AAAAAAAAAADsopAMAAAAAAAAAYAeFdAAAAAAAAAAA7KCQDgAAAAAAAACAHRTSAQAAAAAAAACwg0I6AAAAAAAAAAB2UEgHAAAAAAAAAMAOCukAAAAAAAAAANhBIR0AAAAAAAAAADsopAMAAAAAAAAAYAeFdAAAAAAAAAAA7KCQDgAAAAAAAACAHRTSAQAAAAAAAACwg0I6AAAAAAAAAAB2UEgHAAAAAAAAAMAOCukAAAAAAAAAANhBIR0AAAAAAAAAADsopAMAAAAAAAAAYAeFdAAAAAAAAAAA7KCQDgAAAAAAAACAHRTSAQAAAAAAAACwg0I6AAAAAAAAAAB2UEgHAAAAAAAAAMAOCukAAAAAAAAAANhBIR0AAAAAAAAAADsopAMAAAAAAAAAYAeFdAAAAAAAAAAA7KCQDgAAAAAAAACAHRTSAQAAAAAAAACww2KMMQk9CLwYgoODNWzYMPXr10/u7u4JPRzgpcbnEUBM8LsCcBx8HgHEBL8rAMfB5/HlQyEdcSYgIEA+Pj66deuWkiVLltDDAV5qfB4BxAS/KwDHwecRQEzwuwJwHHweXz5M7QIAAAAAAAAAgB0U0gEAAAAAAAAAsINCOgAAAAAAAAAAdlBIR5xxd3fXgAEDeMAC4AD4PAKICX5XAI6DzyOAmOB3BeA4+Dy+fHjYKAAAAAAAAAAAdnBFOgAAAAAAAAAAdlBIBwAAAAAAAADADgrpAAAAAAAAAADYQSEdAAAAAAAAAAA7KKQjWm3bttUbb7yR0MOQJA0cOFBFixZN6GEADiWhPqN8HgHHRXYDjo3sBvAoshtwbGQ3HkUhPRFp27atLBaLLBaL3NzclDNnTg0ePFihoaHPfSzr1q2zjsViscjT01MFChTQtGnT4nxfH3/8sdasWRPn/QJxzZE/o2nTplWjRo10/PjxGPdhsVi0dOnS+Bsk8BJw5N8LZDfg2J9RshtIGI78e4HsBhz7M0p2v/hcEnoAiJ1atWpp5syZCg4O1q+//qrOnTvL1dVV/fr1i9I2JCREbm5u8TqeQ4cOKVmyZAoKCtLy5cv1wQcfKEeOHKpatWqc7cPLy0teXl5x1h8QnxzxM+rt7a0jR47o/fffV7169bR37145OzvH634B/McRfy+Q3cB/HPEzSnYDCcsRfy+Q3cB/HPEzSna/HLgiPZFxd3dXunTplDVrVn3wwQeqVq2ali1bJum/W06+/PJLZciQQXny5JEknTlzRk2bNlXy5MmVIkUKNWjQQCdPnrT2GRYWpo8++kjJkydXypQp1bt3bxljYjSeNGnSKF26dMqWLZu6deumbNmyaefOndb14eHhGjZsmLJlyyZPT08VKVJEixcvtq6P/PZuzZo1evXVV5UkSRKVKVNGhw4dsrZ59JaW0NBQdevWzTrePn36qE2bNja321SqVEndunVT7969lSJFCqVLl04DBw6MxTsNPB1H/IymT59eFSpU0Oeff679+/fr6NGj2r59u6pXr65UqVLJx8dHFStWtPns+vn5SZLefPNNWSwW6+tIc+bMkZ+fn3x8fNS8eXMFBgZa1wUHB6tbt25KkyaNPDw8VK5cOW3fvv0p3k3gxeCIvxfIbuA/jvgZJbuBhOWIvxfIbuA/jvgZJbtfDhTSEzlPT0+FhIRYX69Zs0aHDh3S6tWr9csvv+j+/fuqWbOmvL299ddff2njxo3y8vJSrVq1rNuNGjVKs2bN0nfffae///5b169f108//RSrcRhj9Ntvv+n06dMqVaqUdfmwYcP0/fffa+rUqfr333/Vo0cPvf3221q/fr3N9p9++qlGjRqlHTt2yMXFRe+8885j9zV8+HD5+/tr5syZ2rhxowICAqK9DWb27NlKmjSptm7dqhEjRmjw4MFavXp1rI4LeFaO8hmNHIsU8Y18YGCg2rRpo7///ltbtmxRrly5VLt2bWswRwbwzJkzdeHCBZtAPnbsmJYuXapffvlFv/zyi9avX6+vvvrKur5379768ccfNXv2bO3cuVM5c+ZUzZo1df369di/gcALyFF+L5DdQPQc5TMaORaJ7AYSmqP8XiC7geg5ymc0ciwS2f3CMkg02rRpYxo0aGCMMSY8PNysXr3auLu7m48//ti6Pm3atCY4ONi6zZw5c0yePHlMeHi4dVlwcLDx9PQ0v//+uzHGmPTp05sRI0ZY19+/f99kypTJuq/orF271kgySZMmNUmTJjUuLi7GycnJDBkyxNrm3r17JkmSJGbTpk0227777rumRYsWNv388ccf1vUrVqwwkkxQUJAxxpgBAwaYIkWKWNenTZvWfP3119bXoaGhJkuWLDbjrVixoilXrpzNfkuUKGH69Onz2GMCnpUjfkZv3LhhjDHm/PnzpkyZMiZjxow2+48UFhZmvL29zfLly63LJJmffvrJpt2AAQNMkiRJTEBAgHVZr169TKlSpYwxxty+fdu4uroaf39/6/qQkBCTIUMGm2MAXhaO+HuB7Ab+44ifUbIbSFiO+Hvh/+zdd3xN9x/H8ffNjiQIYkQQe+9Zm1arWrs2pfqzam9qa0uponbtUdRorSqlVFVr1qraM0bsFTJknN8f4XIluRKS5iZez8cjj0fu93zPOd8Tbt45n3vO95DdwFO2+B4lu18fzJGexPz0009yd3dXaGioIiIi1Lx5c4tbpwoXLmwx99OhQ4d0+vRpeXh4WGwnODhYZ86c0b179+Tv72/xabaDg4NKlSoVq1tY/vjjD3l4eCgkJER79uxRly5dlCZNGnXq1EmnT59WYGCgatSoYbHOo0ePVLx4cYu2IkWKmL/PlCmTJOn69evKmjWrRb979+7p2rVrKlOmjLnN3t5eJUuWVERERIzbfLLd69evv/CYgFdha+9RHx8fGYahwMBAFS1aVD/88IOcnJx07do1DR48WNu2bdP169cVHh6uwMBA+fn5vXCbvr6+FuN99r115swZhYaGqkKFCubljo6OKlOmjI4dO/bCbQPJka39XiC7AUu29h4lu4HEZ2u/F8huwJKtvUfJ7tcHhfQkplq1apo+fbqcnJzk7e0tBwfLf0I3NzeL1w8ePFDJkiW1ePHiKNvy8vJ65fFkz55dqVOnliQVLFhQu3fv1hdffKFOnTrpwYMHkqT169crc+bMFus5OztbvHZ0dDR/bzKZJClKQMfVs9t8st1X3SbwIrb2Hv3jjz+UMmVKpU+f3iKEW7durVu3bumbb75RtmzZ5OzsrDfeeMPidriY8N4C4sbWfi+Q3YAlW3uPkt1A4rO13wtkN2DJ1t6jZPfrg0J6EuPm5qZcuXLFun+JEiW0bNkypU+fXilTpoy2T6ZMmbR7925VrlxZUuRDRf7++2+VKFEizuOzt7dXUFCQJKlAgQJydnaWn5+fqlSpEudtRSdVqlTKkCGD9u7dax5veHi49u/fb/FgFCCx2Np79Nk/up/1559/atq0aapVq5akyAev3Lx506KPo6OjwsPDY30skpQzZ045OTnpzz//VLZs2SRJoaGh2rt3r3r06BGnbQHJha39Xnge2Y3Xna29R8luIPHZ2u+F55HdeN3Z2nuU7H598LDRZK5FixZKly6d6tatqz/++EPnzp3Ttm3b1K1bN126dEmS1L17d3355ZdavXq1jh8/rk8++UR3796N1favX7+uq1ev6sKFC1qxYoUWLVqkunXrSpI8PDzUp08f9ezZUwsWLNCZM2e0f/9+TZ48WQsWLHjpY+ratatGjx6tNWvW6MSJE+revbvu3Llj/kQdSEoS+j0ak9y5c2vRokU6duyYdu/erRYtWpgfivKEr6+vtmzZoqtXr+rOnTux2q6bm5s6deqkvn37auPGjTp69KjatWunwMBAffzxx680ZuB1QXYDto3sBvA8shuwbWQ34gtXpCdzKVKk0Pbt29W/f381aNBAAQEBypw5s958803zp3C9e/eWv7+/WrduLTs7O7Vt21b169fXvXv3Xrj9vHnzSoqcOypLlizq0KGDxbxUn332mby8vDR69GidPXtWqVOnVokSJfTpp5++9DH1799fV69e1Ycffih7e3u1b99e77zzjuzt7V96m0BiSej3aEzmzJmj9u3bq0SJEsqSJYtGjRqlPn36WPT5+uuv1atXL82aNUuZM2fW+fPnY7XtL7/8UhEREWrVqpUCAgJUqlQp/fLLL/L09Hzp8QKvE7IbsG1kN4Dnkd2AbSO7EV9MRmxmzQdsWEREhPLnz6/GjRvrs88+S+zhAACAFyC7AQBIWshuAOCKdCRBFy5c0KZNm1SlShWFhIRoypQpOnfunJo3b57YQwMAANEguwEASFrIbgCIijnSkeTY2dlp/vz5Kl26tCpUqKB//vlHv/76q/Lnz5/YQwMAANEguwEASFrIbgCIiqldAAAAAAAAAACwgivSAQAAAAAAAACwgkI6AAAAAAAAAABWUEgHAAAAAAAAAMAKCukAAAAAAAAAAFhBIR0AAAAAAAAAACsopAMAAAAAAAAAYAWFdAAAAAAAAAAArKCQDgAAAAAAAACAFRTSAQAAAAAAAACwgkI6AAAAAAAAAABWUEgHAAAAAAAAAMAKCukAAAAAAAAAAFhBIR0AAAAAAAAAACsopAMAAAAAAAAAYAWFdAAAAACSpDZt2sjd3T2xhwEAgE1o06aNfH19E3sYAGwEhXQgiZk/f75MJpP5y8HBQZkzZ1abNm10+fJli75Vq1a16PvsV758+SRJderUUYoUKRQQEBDjPlu0aCEnJyfdunVLkmLcpslkUseOHc3rcTIOAHidPZ/Zz3/t2rVLkvTgwQMNGzZMhQoVkpubm9KmTatixYqpe/fuunLlinl7w4cPl8lkkp2dnS5evBhlf/fv35erq6tMJpO6dOlisWz69Olq1KiRsmbNKpPJpDZt2iTosQMAYAuez2IXFxflyZNHXbp00bVr1xJ7eFYtWbJEEydOjJdthYeHy9vbWyaTSRs2bIi2z5O/M27evBkv+wSSI4fEHgCAlzNy5Ehlz55dwcHB2rVrl+bPn68dO3boyJEjcnFxMffz8fHR6NGjo6yfKlUqSZFF8nXr1mnVqlX68MMPo/QLDAzUmjVrVLNmTaVNm9bcXqNGjWj758mTJz4ODwCAZONJZj8vV65cCg0NVeXKlXX8+HG1bt1aXbt21YMHD/Tvv/9qyZIlql+/vry9vS3Wc3Z21tKlS9WvXz+L9h9//DHGMYwZM0YBAQEqU6aM/P394+fAAABIIp49f96xY4emT5+un3/+WUeOHFGKFCliXG/WrFmKiIj4D0f61JIlS3TkyBH16NHjlbe1detW+fv7y9fXV4sXL9a777776gMEXkMU0oEk6t1331WpUqUkSf/73/+ULl06jRkzRmvXrlXjxo3N/VKlSqWWLVvGuJ06derIw8NDS5YsibYwvmbNGj18+FAtWrSwaM+TJ4/V7QIAgEjPZvbzVqxYoQMHDmjx4sVq3ry5xbLg4GA9evQoyjq1atWKtpC+ZMkSvffee/rhhx+irPP777+br0bnbjEAwOvm+fPntGnTavz48VqzZo2aNWsWpf/Dhw/l5uYmR0fH/3qoCeK7775TiRIl1Lp1a3366afm4wMQN0ztAiQTlSpVkiSdOXMmTuu5urqqQYMG2rJli65fvx5l+ZIlS+Th4aE6derEyzgBAMBTT3K7QoUKUZa5uLgoZcqUUdqbN2+ugwcP6vjx4+a2q1evauvWrVGK8U9ky5ZNJpMp1uM6e/as3nnnHbm5ucnb21sjR46UYRixXh8AAFtWvXp1SdK5c+fMU5KeOXNGtWrVkoeHh/lCsufnSD9//rxMJpPGjRunqVOnKkeOHEqRIoXefvttXbx4UYZh6LPPPpOPj49cXV1Vt25d3b5922Lfa9as0XvvvSdvb285OzsrZ86c+uyzzxQeHm7uU7VqVa1fv14XLlwwT0vz7DhCQkI0bNgw5cqVS87OzsqSJYv69eunkJCQKMcaFBSkVatWqWnTpmrcuLGCgoK0Zs2aePxpAq8PrkgHkonz589Lkjw9PS3aw8PDo53jzNXV1fwJdIsWLbRgwQItX77cYk7V27dv65dfflGzZs3k6upqsX5wcHC0202ZMqWcnJxe9XAAAEg27t27FyUzTSaT0qZNq2zZskmSFi5cqMGDB8eq2F25cmX5+PhoyZIlGjlypCRp2bJlcnd313vvvffK4w0PD1fNmjVVrlw5jR07Vhs3btSwYcMUFhZm3h8AAEnZkw+yn0xfGhYWpnfeeUcVK1bUuHHjrE73IkmLFy/Wo0eP1LVrV92+fVtjx45V48aNVb16dW3btk39+/fX6dOnNXnyZPXp00dz5841rzt//ny5u7urV69ecnd319atWzV06FDdv39fX331lSRp0KBBunfvni5duqQJEyZIkvmOsoiICNWpU0c7duxQ+/btlT9/fv3zzz+aMGGCTp48qdWrV1uMde3atXrw4IGaNm2qjBkzqmrVqtHeCQfgxSikA0nUk5Py4OBg7d69WyNGjJCzs7Pef/99i37Hjx+Xl5dXlPU7dOigGTNmSIr8ND5TpkxasmSJRSF9xYoVCg0NjTKtiyTNmTNHc+bMidK+dOlSNW3a9FUPDwCAZOOtt96K0ubs7Kzg4GDVq1dPefPm1dChQzVnzhxVq1ZNlSpV0vvvv6/06dNHuz2TyaSmTZtq6dKl5sL24sWL1aBBAzk7O7/yeIODg1WzZk1NmjRJkvTJJ5+odu3aGjNmjLp166Z06dK98j4AAPgvPXv+/Oeff2rkyJFydXXV+++/r507dyokJESNGjWK9vli0bl8+bJOnTplfvZYeHi4Ro8eraCgIO3bt08ODpHlths3bmjx4sWaPn26OaOXLFlicaFax44d1bFjR02bNk2ff/65nJ2dVaNGDWXOnFl37tyJMqXqkiVL9Ouvv+r3339XxYoVze2FChVSx44d9ddff6l8+fLm9u+++07ly5dXlixZJElNmzbVJ598ohs3bkRbKwAQMwrpQBL1/Em5r6+vvvvuO/n4+ERpnzVrVpT1n+1nb2+vpk2basKECTp//rz5lrElS5YoQ4YMevPNN6OsX7duXYui+xOFCxd+mcMBACDZmjp1apSHcdvb20uKvENs9+7d+uKLL7R8+XLNnz9f8+fPl52dnT755BONGzcu2uJ48+bNNW7cOO3du1eenp7au3evRo0aFW9jfjbjTSaTunTpovXr1+vXX3/lA3MAQJLz/PlztmzZtHjxYmXOnNnc1qlTp1hvr1GjRuYiuiSVLVtWktSyZUtzEf1J+9KlS3X58mXlyJFDkiyK6AEBAQoJCVGlSpX07bff6vjx4ypatKjVfa9YsUL58+dXvnz5LO54ezJdzW+//WYupN+6dUu//PKL+ap2SWrYsKE6d+6s5cuXq3PnzrE+ZgAU0oEk68lJ+b179zR37lxt37492hNtNze3aK+Ee16LFi00YcIELVmyRJ9++qkuXbqkP/74Q926dTOf7D/Lx8cnVtsFAOB1V6ZMmRgfNipFPhh87NixGjt2rC5cuKAtW7Zo3LhxmjJlilKlSqXPP/88yjrFixdXvnz5tGTJEqVOnVoZM2Y0n0C/Kjs7O/PJ/hNPPgh4MpUcAABJyZPzZwcHB2XIkEF58+aVnd3TxwY6ODhEuSjNmqxZs1q8flJUf3LV9/Ptd+7cMbf9+++/Gjx4sLZu3ar79+9b9L93794L933q1CkdO3YsxqvJn3322bJlyxQaGqrixYvr9OnT5vayZctq8eLFFNKBOKKQDiRRz56U16tXTxUrVlTz5s114sQJ89xpcVGyZEnly5dPS5cu1aeffqqlS5fKMIxop3UBAAAJI1u2bGrbtq3q16+vHDlyaPHixdEW0qXIq9KnT58uDw8PNWnSxKIgAAAAnnrRh9rOzs5xytHoLjaz1v7kgd13795VlSpVlDJlSo0cOVI5c+aUi4uL9u/fr/79+ysiIuKF+46IiFDhwoU1fvz4aJc/W8xfvHixpOgfai5FPlz8+Q/PAcSMQjqQDNjb22v06NGqVq2apkyZogEDBrzUdlq0aKEhQ4bo8OHDWrJkiXLnzq3SpUvH82gBAMCLeHp6KmfOnDpy5EiMfZo3b66hQ4fK399fixYtird9R0RE6OzZsxbT0Zw8eVKSzNO/AQCAuNu2bZtu3bqlH3/8UZUrVza3nzt3LkrfmB5AnjNnTh06dEhvvvmm1YeUnzt3Tn/99Ze6dOmiKlWqWCyLiIhQq1attGTJEg0ePPgljwZ4/XDZCpBMVK1aVWXKlNHEiRMVHBz8Utt4cvX50KFDdfDgQa5GBwAggR06dMhiftMnLly4oKNHjypv3rwxrpszZ05NnDhRo0ePVpkyZeJ1XFOmTDF/bxiGpkyZIkdHx2ifmwIAAGLnyRXrT65Ql6RHjx5p2rRpUfq6ublFO9VL48aNdfny5WifhRYUFKSHDx9Keno1er9+/fTBBx9YfDVu3FhVqlQx9wEQO1yRDiQjffv2VaNGjTR//nx17NhRUuQca9999120/Z9/+nf27NlVvnx5rVmzRpKsFtJPnjwZ7XYzZMigGjVqmF+HhoZGe0t6mjRp9Mknn7z4oAAASOI2bNig48ePR2kvX768Nm/erGHDhqlOnToqV66c3N3ddfbsWc2dO1chISEaPny41W137949VmNYt26dDh06JCkymw8fPmzO5zp16qhIkSLmvi4uLtq4caNat26tsmXLasOGDVq/fr0+/fTTGOdjBQAAL1a+fHl5enqqdevW6tatm0wmkxYtWmRRWH+iZMmSWrZsmXr16qXSpUvL3d1dtWvXVqtWrbR8+XJ17NhRv/32mypUqKDw8HAdP35cy5cv1y+//KJSpUpp8eLFKlasWJR525+oU6eOunbtqv3796tEiRIJfehAskAhHUhGGjRooJw5c2rcuHFq166dJOnSpUtq1apVtP2fL6RLkcXzv/76S2XKlFGuXLli3NfmzZu1efPmKO1VqlSxKKQ/evRIQ4YMidIvZ86cFNIBAK+FoUOHRts+b948NWzYUAEBAdq0aZO2bt2q27dvy9PTU2XKlFHv3r1VrVq1eBnDDz/8oAULFphfHzhwQAcOHJAU+QDxZwvp9vb22rhxozp16qS+ffvKw8NDw4YNi/E4AABA7KRNm1Y//fSTevfurcGDB8vT01MtW7bUm2++qXfeecei7yeffKKDBw9q3rx5mjBhgrJly6batWvLzs5Oq1ev1oQJE7Rw4UKtWrVKKVKkUI4cOdS9e3flyZNH+/fv1/Hjx6M9F3+idu3a6tq1q7777jsK6UAsmYzoPvYCAAAAAAAAAACSmCMdAAAAAAAAAACrKKQDAAAAAAAAAGAFhXQAAAAAAAAAAKygkA4AAAAAAAAAgBUU0gEAAAAAAAAAsIJCOgAAAAAAAAAAVlBIBwAAAAAAAADACgrpSLKmTZsmk8mksmXLJvZQLPj6+spkMkX56tixo0U/f39/DRgwQNWqVZOHh4dMJpO2bdv20vutUaOGTCaTunTpYtEeEhKirl27ysvLSz4+Pvr888+jrHvp0iW5u7vrzz//fOn9S4r2uE0mkzJmzGjuE9/H/ay//vpLFStWVIoUKZQxY0Z169ZNDx48iPX6165dU4cOHZQ5c2a5uLjI19dXH3/8cbR9ly1bpjfeeENubm5KnTq1ypcvr61bt5qXX7x4USNGjFCZMmXk6empdOnSqWrVqvr1119f+TgBIKlK6tktSXfv3lX79u3l5eUlNzc3VatWTfv374/VfmLKSZPJpBo1aljso0WLFvL09FSOHDk0Z86cKNvat2+fUqRIoXPnzr30cZ8/fz7G8ZQrV87c78SJE+rZs6fKly8vFxcXmUwmnT9//qX3+6y1a9eqRIkScnFxUdasWTVs2DCFhYXFev0zZ86oefPmSp8+vVxdXZU7d24NGjTIok+bNm2iPcZ8+fJZ9Bs+fLjVf6NX/TsJAJKi1z27nxUaGqoCBQrIZDJp3LhxUfZBdscO2Y2kzCGxBwC8rMWLF8vX11d79uzR6dOnlStXrsQeklmxYsXUu3dvi7Y8efJYvD5x4oTGjBmj3Llzq3Dhwtq5c+dL7+/HH3+Mcf2vvvpKCxcu1KBBgxQQEKCRI0cqZ86catasmblP3759VadOHVWoUOGlx/BEjRo19OGHH1q0ubq6mr+Pz+N+1sGDB/Xmm28qf/78Gj9+vC5duqRx48bp1KlT2rBhwwvXv3jxovn4O3bsqMyZM+vKlSvas2dPlL7Dhw/XyJEj9cEHH6hNmzYKDQ3VkSNHdPnyZXOfNWvWaMyYMapXr55at26tsLAwLVy4UDVq1NDcuXP10UcfxctxA0BSktSzOyIiQu+9954OHTqkvn37Kl26dJo2bZqqVq2qv//+W7lz57a6j0WLFkVp27dvn7755hu9/fbb5rY+ffpo27ZtGjFihE6fPq127dopf/78Kl++vCTJMAx169ZNPXr0UPbs2V/2kM2aNWumWrVqWbR5eXmZv9+5c6cmTZqkAgUKKH/+/Dp48OAr71OSNmzYoHr16qlq1aqaPHmy/vnnH33++ee6fv26pk+f/sL1Dx48qKpVqypz5szq3bu30qZNKz8/P128eDFKX2dnZ82ePduiLVWqVBavGzRoEO3/yU8//VQPHjxQ6dKl43iEAJD0ve7Z/azJkyfLz88v2mVkN9mN14QBJEFnz541JBk//vij4eXlZQwfPjyxh2SWLVs247333nthv/v37xu3bt0yDMMwVqxYYUgyfvvttzjvLygoyPD19TVGjhxpSDI6d+5ssbxs2bLGiBEjzK9bt25tNG3a1Pz6jz/+MNzc3IyLFy/Ged/Pi27/z4uv437eu+++a2TKlMm4d++euW3WrFmGJOOXX36J1frZs2c3bt68abXfzp07DZPJZIwfP95qvyNHjhg3btywaAsODjby5ctn+Pj4vHA8AJDcJIfsXrZsmSHJWLFihbnt+vXrRurUqY1mzZq91L4//vhjw2QyWeRwhgwZjAULFphfV6lSxRgwYID59aJFiwxvb28jICDgpfb5xLlz5wxJxldffWW1361bt4z79+8bhmEYX331lSHJOHfu3Cvt2zAMo0CBAkbRokWN0NBQc9ugQYMMk8lkHDt2zOq64eHhRqFChYyyZcsagYGBVvu2bt3acHNze6kx+vn5GSaTyWjXrt1LrQ8ASRnZ/dS1a9eMVKlSmc+7n89OspvsxuuBqV2QJC1evFienp5677339MEHH2jx4sXR9rt796569uwpX19fOTs7y8fHRx9++KFu3rxp7hMcHKzhw4crT548cnFxUaZMmdSgQQOdOXPG3Mff31/Hjx9XaGhorMf46NEjPXz4MMblHh4eSpMmTay3F5OxY8cqIiJCffr0iXZ5UFCQPD09za/TpEmjwMBASZGfznfv3l39+vWTj4/PK48lNuJy3Ddv3tTx48fN443J/fv3tXnzZrVs2VIpU6Y0t3/44Ydyd3fX8uXLra5//PhxbdiwQX379lXatGkVHBwc47/1xIkTlTFjRnXv3l2GYcQ4dUzBggWVLl06izZnZ2fVqlVLly5dUkBAgNUxAUBykxyye+XKlcqQIYMaNGhgbvPy8lLjxo21Zs0ahYSExHpfUuT0az/88IOqVKlikcPWsvvhw4caMGCARo8eLXd39zjt72WlSZNGHh4eseob25/70aNHdfToUbVv314ODk9vkv3kk09kGIZWrlxpdf1NmzbpyJEjGjZsmFxdXRUYGKjw8HCr64SHh+v+/fuxOo4nli5dKsMw1KJFizitBwDJAdn91IABA5Q3b161bNky2uVkN9mN1wOFdCRJixcvVoMGDeTk5KRmzZrp1KlT2rt3r0WfBw8eqFKlSpo8ebLefvttffPNN+rYsaOOHz+uS5cuSYr8pfz+++9rxIgRKlmypL7++mt1795d9+7d05EjR8zbGjhwoPLnz28xdYc1W7duVYoUKeTu7i5fX19988038Xfwz/Dz89OXX36pMWPGWEyf8qzSpUtr5syZ+ueff7Rz504tXbpUZcqUkSTNmTNHN2/eVN++feNtTMHBwbp586bFV1wLC09MmTJF+fPnj3Z6lWf9888/CgsLU6lSpSzanZycVKxYMR04cMDq+k/mLc+QIYPefPNNubq6ytXVVe+++26UeeS2bNmi0qVLa9KkSfLy8pKHh4cyZcqkKVOmxOqYrl69qhQpUihFihSx6g8AyUVyyO4DBw6oRIkSsrOz/BO6TJkyCgwM1MmTJ+P0M/n555/Nc6o+q3Tp0ho/frxOnTqlX375RRs3bjRn96hRo5Q5c2a1atUqTvuyJjAwMEp2x6WI8azY/tyfZPPz2e3t7S0fH59YZ7ezs7NKlSolNzc3pUiRQk2bNtXt27ej9A8MDFTKlCmVKlUqpUmTRp07d47Vc1QWL16sLFmyqHLlyi/sCwDJDdkdac+ePVqwYIEmTpwok8kUbR+ym+zGayJRr4cHXsK+ffsMScbmzZsNwzCMiIgIw8fHx+jevbtFv6FDh5pvQ3teRESEYRiGMXfuXENStNN0POljGJG3FSmWt0LVrl3bGDNmjLF69Wpjzpw5RqVKlQxJRr9+/WJc52WnOPnggw+M8uXLm18rmqlVLl68aBQsWNCQZEgyKlWqZAQEBBh37941vLy8jO+//z5O+7TmyT6e/5o3b160/V903MOGDYvVz+XJdrZv3x5lWaNGjYyMGTNaXb9bt26GJCNt2rRGzZo1jWXLlhlfffWV4e7ubuTMmdN4+PChYRiGcfv2bXM/d3d346uvvjKWLVtm1KxZ05BkzJgxw+p+Tp06Zbi4uBitWrWy2g8Akpvkkt1ubm5G27Zto6y/fv16Q5KxcePGF+7rWQ0bNjScnZ2NO3fuWLQfPnzY8PHxMedow4YNjfDwcOPs2bOGq6ursXPnzjjtJyZPbg+P7ium7H3R7eGx/bk/2Y6fn1+UZaVLlzbKlStndf06deqYM7lFixbGypUrjSFDhhgODg5G+fLlLf4vDBgwwOjfv7+xbNkyY+nSpeYxVqhQweLW9OcdOXLkhX/DAUByRXY/HV+ZMmXM08DENLUK2U124/VAIR1JTs+ePY0MGTIYYWFh5rbevXtHaStYsKBRtGhRq9t67733jHTp0ln9RfyqIiIijHfeecdwcHCIcR7ylymkb9261TCZTMaePXvMbdEV0g3DMB49emQcOHDA+Pfff43w8HDDMCJ/jhUrVjQMI3Ke9DJlyhg+Pj5G165djZCQkDgc4VOSjLp16xqbN2+2+Lpy5Uq0/eNrjvSFCxcakozdu3dHWdaqVSsjVapUVtdv27atIckoWLCg+edjGIaxdOlSQ5Ixa9YswzAi51p78kfKsx9AhIeHGwUKFLA69/nDhw+NYsWKGZ6ensbly5fjeIQAkLQll+y2s7MzOnXqFKX/li1bDEnGqlWrYr2Pe/fuGS4uLkb9+vWjXR4UFGTs3bvXOHXqlLmtfv36RsuWLQ3DMIwffvjBKFKkiOHr62uMGDHC4uQztp6cjLdv3z5Kdt++fTvadeJrntUnc8xeu3YtyrJKlSq98P9B9erVDUlGzZo1LdpHjx5tUfiJyRdffGFIMpYuXRpjn4EDBxqSjEOHDlndFgAkR2R3pLlz5xqurq7m4rG1OcrJ7qJW1ye7kRwwtQuSlPDwcH3//feqVq2azp07p9OnT+v06dMqW7asrl27pi1btpj7njlzRoUKFbK6vTNnzihv3rwW83vFN5PJpJ49eyosLEzbtm2Ll22GhYWpW7duatWqVayeQu3o6KhixYqpQIECsrOz0/HjxzVt2jR98803un37tt577z3Vq1dPK1as0ObNm/XFF1+89Nh8fHz01ltvWXxlypTppbcXG0+mtYluCpng4OAYp715fv3GjRtb3PLXqFEjOTg46K+//rLo5+joqA8++MDcz87OTk2aNNGlS5eifYp7eHi4mjZtqqNHj2rlypXy9vaO4xECQNKVnLLb1dU1xqx5sjy2fvjhBwUHB8c4f6eLi4tKlSqlXLlySYq8fX3Tpk368ssvdeLECTVt2lQ9evTQ3LlzNW3aNM2fPz/2B/ic3LlzR8nuZ+d5TQjxld3NmjWzaG/evLkkmbM7Jj179pSdnZ35NvPnGYahJUuWqFChQipSpIjVbQFAckN2R7p//74GDhyovn37KkuWLC8cA9lNdiP5o5COJGXr1q3y9/fX999/r9y5c5u/GjduLEkxPvwksT0J3ejm/XoZCxcu1IkTJ9ShQwedP3/e/CVJAQEBOn/+vNUHdPbs2VMtW7ZUiRIltH79eqVJk0YDBw5UuXLl1K9fP5v9OcbkSaHe398/yjJ/f/8XFq6fLM+QIYNFu729vdKmTas7d+5Iinxgi4uLi9KmTSt7e3uLvunTp5ckc99ntWvXTj/99JPmz5+v6tWrx/KoACB5SE7ZnSlTphizRlKcPihdvHixUqVKpffff/+FfcPDw9W9e3cNGDBAmTNn1vLly1W+fHl99NFHqlatmjp06GCzP8eYJFR2W8vjZ7m6uipt2rQx/m32559/6sKFCzyoDMBrieyONG7cOD169EhNmjQxn3M/mff9zp07On/+vB49ehTtumR3VGQ3kgMK6UhSFi9erPTp02vFihVRvpo1a6ZVq1YpKChIkpQzZ06LB5dEJ2fOnDpx4sRLP5Qjts6ePSsp8ung8cHPz0+hoaGqUKGCsmfPbv6SIovs2bNn16ZNm6Jd96efftJff/2lUaNGSZKuXLliccW4t7d3rB/uYisKFSokBwcH7du3z6L90aNHOnjwoIoVK2Z1/ZIlS0pSlON+9OiRbt68af53s7OzU7FixXTjxo0ofzBduXJFUtR/4759+2revHmaMGFClE/eAeB1kJyyu1ixYtq/f78iIiIs+u7evVspUqRQnjx5YrVtf39//fbbb2rYsKGcnZ1f2H/69OkKCAhQnz59JEVmzrMnq0kxu59k8/PZfeXKFV26dOmlszumPH5eQECARcY/b/HixTKZTOar5ADgdUJ2R/Lz89OdO3dUsGBB8zl3pUqVJEU+QDR79uw6evRotOuS3VGR3UgWEntuGSC2AgMDDQ8Pj2gfFGIYhvHnn39azF0dnw89uXLlinHs2DHj0aNHVsd469Yti/niDCNyfvIKFSoYTk5Ohr+/f7TrvWiu8AsXLhjHjh0zvz527JixatWqKF+SjFq1ahmrVq2Kdl7ykJAQI3fu3MbYsWPNbfPmzTO8vLzM89VNmDDB8PX1tXqcMVEMc7TH5EXHfePGDePYsWPmh31aU7NmTSNTpkzG/fv3zW2zZ882JBkbNmwwtz18+NA4duyYcePGDXNbcHCwkT59eiNHjhxGUFCQuf3bb781JBnLly83t02YMMGQZMycOdPcFhQUZOTIkcMoUKCAxZjGjh1rSDI+/fTTF44fAJKj5Jbd33//vSHJWLFihbntxo0bRurUqY0mTZpYbOP06dPG6dOno93n+PHjDUnGli1brI7tyfjSpEljkUXDhg0zihcvbn7dvXt3o2rVqi/c1vOszfMakxfNsxrbn7thGEa+fPmMokWLWvz8Bw8ebJhMJuPo0aPmtrt37xrHjh0z7t69a27z9/c3nJ2djYoVK1o83+TJ3KhPniETFBRk8bfBE3379o3x/9qjR4+MtGnTGpUqVXrhMQBAckN2P83uv//+O8o595NzxDZt2hirVq2yyKZnx0d2k91IniikI8l4EoCrV6+Odnl4eLjh5eVl1K5d2zAMwwgICDAKFChg2NvbG+3atTNmzJhhjBo1yihXrpxx8OBBwzAMIywszKhataohyWjatKkxdepUY+zYscbbb79tsZ/YPsV63rx5Rs6cOY3+/fub91eoUCFDkjFq1Kgo/T/77DPjs88+M5o2bWpIMtq2bWtue1aVKlWM2Hzu9aJC9tixY43cuXNbPEz02rVrRooUKYxGjRoZEyZMMNKkSfPST7iObSE9tsc9bNiwWD+M9O+//zacnZ2N4sWLG9OnTzcGDRpkuLi4GG+//bZFv99++82QZAwbNsyifcGCBYYko3Tp0sakSZOMPn36GI6OjkalSpUs/kgIDAw0ChYsaDg6Ohp9+vQxJk2aZJQuXdqwt7c3fv75Z3O/H3/80ZBk5M6d21i0aFGUr6tXr77wmAAgqUtu2R0WFmaUK1fOcHd3N0aMGGFMnTrVKFiwoOHh4WEcP37com+2bNmMbNmyRbvPkiVLGt7e3hYnkTH55JNPjCpVqli0HT582DCZTEbHjh2N0aNHGy4uLsa0adNeuK3nxfZk/O7du+acrlmzpiHJ6N27t/HZZ58ZkydPtugb25+7YRjGunXrDJPJZFSvXt2YOXOm0a1bN8POzs5o166dRb958+YZkox58+ZZtD956FmNGjWMqVOnGu3btzdMJpPRrFkzi2NMnTq10alTJ+Obb74xvvnmG6NWrVrmh51F92+wbt06Q5IxY8aMFx4DACQ3ZHc2q/uOTXaS3WQ3ki8K6Ugyateubbi4uFi9OrlNmzaGo6OjcfPmTcMwIj8J7tKli5E5c2bDycnJ8PHxMVq3bm1ebhiRhdFBgwYZ2bNnNxwdHY2MGTMaH3zwgXHmzBlzn9gGy759+4zatWub9+fu7m5UrFjR4pPoZ0mK8etZ8VFIv3r1quHh4WGsXbs2yrINGzYY+fLlM1KnTm18+OGHsboCPK77f75fbI47LoV0wzCMP/74wyhfvrzh4uJieHl5GZ07d47ySXZMhXTDMIylS5caRYsWNZydnY0MGTIYXbp0ifaT8GvXrhmtW7c20qRJYzg7Oxtly5Y1Nm7cGO3YY/qK7TEBQFKWHLP79u3bxscff2ykTZvWSJEihVGlShVj7969UfrFdDJ+/PhxQ5LRq1cvq+MyjMiTbicnJ+PAgQNRls2fP9/w9fU10qZNa/Tq1SvKlXmxEduT8Sf9ovt6/hjjcjJuGIaxatUqo1ixYoazs7Ph4+NjDB48OMoVcTGdjEdERBiTJ0828uTJYzg6OhpZsmSJsv6dO3eMli1bGrly5TJSpEhhODs7GwULFjRGjRoV45V3TZs2NRwdHY1bt27F6hgAIDkhu7NZ3feLspPsjkR2I7kyGYZhPD/dCwAAAAAAAAAAiMTDRgEAAAAAAAAAsIJCOgAAAAAAAAAAVlBIBwAAAAAAAADACgrpAAAAAAAAAABYQSEdAAAAAAAAAAArKKQDAAAAAAAAAGCFQ2IP4InrAaGJPQQAj2Wr3DOxhwDgsaADUxJ7CDG69TAssYcA4DGfij0SewgAHrPl7L4TGJ7YQwDwmHeF7ok9BACPxTa7uSIdAAAAAAAAAAArKKQDAAAAAAAAAGAFhXQAAAAAAAAAAKygkA4AAAAAAAAAgBUU0gEAAAAAAAAAsIJCOgAAAAAAAAAAVlBIBwAAAAAAAADACgrpAAAAAAAAAABYQSEdAAAAAAAAAAArKKQDAAAAAAAAAGAFhXQAAAAAAAAAAKygkA4AAAAAAAAAgBUU0gEAAAAAAAAAsIJCOgAAAAAAAAAAVlBIBwAAAAAAAADACgrpAAAAAAAAAABYQSEdAAAAAAAAAAArKKQDAAAAAAAAAGAFhXQAAAAAAAAAAKygkA4AAAAAAAAAgBUU0gEAAAAAAAAAsIJCOgAAAAAAAAAAVlBIBwAAAAAAAADACgrpAAAAAAAAAABYQSEdAAAAAAAAAAArKKQDAAAAAAAAAGAFhXQAAAAAAAAAAKygkA4AAAAAAAAAgBUU0gEAAAAAAAAAsIJCOgAAAAAAAAAAVlBIBwAAAAAAAADACgrpAAAAAAAAAABYQSEdAAAAAAAAAAArKKQDAAAAAAAAAGAFhXQAAAAAAAAAAKygkA4AAAAAAAAAgBUU0gEAAAAAAAAAsIJCOgAAAAAAAAAAVlBIBwAAAAAAAADACgrpAAAAAAAAAABYQSEdAAAAAAAAAAArKKQDAAAAAAAAAGAFhXQAAAAAAAAAAKyIcyH92rVratWqlby9veXg4CB7e3uLLwAAYFvIbgAAkhayGwAA2+MQ1xXatGkjPz8/DRkyRJkyZZLJZEqIcQEAgHhCdgMAkLSQ3QAA2J44F9J37NihP/74Q8WKFUuA4QAAgPhGdgMAkLSQ3QAA2J44T+2SJUsWGYaREGMBAAAJgOwGACBpIbsBALA9cS6kT5w4UQMGDND58+cTYDgAACC+kd0AACQtZDcAALYnzlO7NGnSRIGBgcqZM6dSpEghR0dHi+W3b9+Ot8EBAIBXR3YDAJC0kN0AANieOBfSJ06cmADDAAAACYXsBgAgaSG7AQCwPXEupLdu3TohxgEAABII2Q0AQNJCdgMAYHviXEi/f/9+tO0mk0nOzs5ycnJ65UEBAID4Q3YDAJC0kN0AANieOBfSU6dOLZPJFONyHx8ftWnTRsOGDZOdXZyfZQoAAOIZ2Q0AQNJCdgMAYHviXEifP3++Bg0apDZt2qhMmTKSpD179mjBggUaPHiwbty4oXHjxsnZ2VmffvppvA8YAADEDdkNAEDSQnYDAGB74lxIX7Bggb7++ms1btzY3Fa7dm0VLlxY3377rbZs2aKsWbPqiy++INABALABZDcAAEkL2Q0AgO2J8z1gf/31l4oXLx6lvXjx4tq5c6ckqWLFivLz83v10QEAgFdGdgMAkLSQ3QAA2J44F9KzZMmiOXPmRGmfM2eOsmTJIkm6deuWPD09X310AADglZHdAAAkLWQ3AAC2J85Tu4wbN06NGjXShg0bVLp0aUnSvn37dPz4ca1cuVKStHfvXjVp0iR+RwoAAF4K2Q0AQNJCdgMAYHtMhmEYcV3p3Llzmjlzpk6cOCFJyps3rzp06CBfX9+XHsj1gNCXXhdA/MpWuWdiDwHAY0EHpsTLdhIiu289DIuXsQF4dT4VeyT2EAA8ZsvZfScwPF7GBuDVeVfonthDAPBYbLP7pQrpCYFCOmA7KKQDtiO+TsYTAoV0wHZQSAdshy1nN4V0wHZQSAdsR2yzO1ZTuxw+fFiFChWSnZ2dDh8+bLVvkSJFYrVjAACQcMhuAACSFrIbAADbFqtCerFixXT16lWlT59exYoVk8lkUnQXsptMJoWH8wk3AACJjewGACBpIbsBALBtsSqknzt3Tl5eXubvAQCAbSO7AQBIWshuAABsW6wK6dmyZYv2ewAAYJvIbgAAkhayGwAA2xarQvrzTp06pd9++03Xr19XRESExbKhQ4fGy8AAAED8IbsBAEhayG4AAGxLnAvps2bNUqdOnZQuXTplzJhRJpPJvMxkMhHoAADYGLIbAICkhewGAMD2xLmQ/vnnn+uLL75Q//79E2I8AAAgnpHdAAAkLWQ3AAC2xy6uK9y5c0eNGjVKiLEAAIAEQHYDAJC0kN0AANieOBfSGzVqpE2bNiXEWAAAQAIguwEASFrIbgAAbE+cp3bJlSuXhgwZol27dqlw4cJydHS0WN6tW7d4GxwAAHh1ZDcAAEkL2Q0AgO0xGYZhxGWF7Nmzx7wxk0lnz559qYFcDwh9qfUAxL9slXsm9hAAPBZ0YMorbyOhsvvWw7CXHRKAeOZTsUdiDwHAY7ac3XcCw192SADimXeF7ok9BACPxTa743xF+rlz5+I8GAAAkHjIbgAAkhayGwAA2xPnOdKfePTokU6cOKGwMK5GAwAgKSC7AQBIWshuAABsR5yvSA8MDFTXrl21YMECSdLJkyeVI0cOde3aVZkzZ9aAAQPifZBIGAf379PSRfN04thR3bp5Q1+M+0aVq75pXv771s1a88NynTh+VPfv3dPcxSuVO28+i210bd9GB/fvs2ir26CR+nw6zKLt53WrtWzxAl3yu6AUbu6q9tbb6tV/sCTJ7/w5jRs9UufPndHDBw+U1iu9arxTSx+17yQHB8u5AIHkokKJnOr54VsqUSCrMnmlUuOeM7Vu2+Fo+04a1FTtPqiovl+t1JQl26Isd3J00PZFfVQ0r4/KNhmtwycvR+mTI0s67Vo6QOEREcpUuZ+5vW71our78TvKmSWdHB3sddrvhr5ZtEVL1+8194npFqdPJ6zShIVb4njkSAxkd/Jx4O99WrJwrk4cO6qbN29o9NeTVKXa0+w2DEOzZ0zR2lUrFRAQoCJFi6vvp0OVJWs2cx+/C+c1ZeI4/XPogEJDQ5Urdx6169RVJUuXNfc5+u8/mj5pgk4cOyqTyaT8BQupc4/eyp0n8u8A/yuX1fD9t6OMb+b8JSpUpGgC/gSAxBMf2X18/Qhl805r0XfIpDUaN2+zJMnZyUGTBzVV8fxZlS97Bm3444ga95pl0T9jupT6slcDlSiQVTmzpNO0pb+r77gfLPq0rF1Ws0a2smgLDgmVZzmm70sqyO7kKzw8XLNnTNXGn9fp9q2bSueVXu/VrqeP2nWUyWQy9zt39oymfjNeB/bvVXhYuLLnyKnR4yYqYyZvi+0ZhqGeXTpo1187NGb8JFWp9pYk6dSJ41o4b7YOHdyve3fvKKN3ZjX4oImaNLf83QAkZ6+a3ZVK5tam2dFPwVOxxVj9fdRPubOl1+RBTZUvR0alcneV/417WrZhn76Y+bPCwiIkxS6X3Vyd9Hm3uqpdrYjSpHLT+Su3NG3p75q9ckc8/CQQX+JcSB84cKAOHTqkbdu2qWbNmub2t956S8OHDyfQk5DgoCDlyp1X79Wpr0F9e0RZHhQUpMLFSqhajXc09vPhMW6ndv0P9HGHLubXLi4uFsu//26Bli1eoE+691aBQoUVFBSkq1eumJc7ODjonffqKG++/HL3SKnTJ09o7BfDFGFEqEPnqOMCkgM3V2f9c/KyFq7ZqWXj28fYr061IipT2FdXrt+Nsc+oHnXlf+Oeiub1iXa5g4OdFo7+SH8eOKNyRS3n27x9L1BjZ2/UifPX9Cg0XLUqFdLM4S114/YD/brzmCTJ962BFuu8XaGgZgxrrlVbDsbuYJHoyO7kIzg4SLny5NX7dRtoYJ+of9R/t2COVixdrMEjR8nbO7NmTp+snp3ba/HKtXJ2dpYk9e3+iXyyZtPkGXPl7OKiZYsXqm/3zlqxdoPSpvNSYOBD9erSQRUrV1OfgUMen/BPUc/O7bX65y1yeOaBd5Omz1H2nDnNr1OlSp3gPwMgscRXdo+Y9pPm/fin+XXAwxDz9/Z2dgoKCdW0pdtU781i0a7v5Oigm3cC9OXsjeraolqM47gXEKSi9UeaX8ftyVhIbGR38rVo/mz9uPJ7DR05Wtlz5tLxf4/o8+GD5Obubi5yX7ropw5tW6p2vYZq16mz3NzcdfbMaTk9zvJnfb94oUUB/onjx/6VZ5o0Gv75GGXImFGHDx3Ql58Pl52dnRo1bZHgxwnYglfN7l2HzkY5Hx76yfuqViav/j7qJ0kKDQvX4p/26ODxi7oXEKjCeXw0dUgz2dmZNGzKOvN6L8rlMb0bqmrpPPpo0EJduHJLb72RX98MbCz/G/e0/vd/XvIngPgW50L66tWrtWzZMpUrV87il3XBggV15syZeB0cEla5CpVUrkKlGJfXfK+OpMirzqxxcXFR2nTpol0WcP+eZk+frC8nTFGpMuXM7bly5zV/7+2TRd4+WcyvM2by1oG/9+rwgf2xOg4gKdr051Ft+vOo1T7eXqk0vn8j1f5kqlZN7hRtn7crFNCb5fKrWd/ZqlmxYLR9hn9SWyfOXdNve05EKaT/8fcpi9dTl25Ti9plVb54DnMh/dqtAIs+tasW1u97T+n85VtWxw/bQXYnH29UqKQ3YshuwzC0fMkitflfB1WuWl2SNHTkaL1fo7K2b9uiGu/U0t07d3TR74IGDv1MufJEZnGnbr3044rvdfbMaaVN56UL58/p/r17atepizJkzCRJ+rj9J2rVpL6u+l+RzzNXt6dMnUpp03kl8FEDtiG+svvBw+Ao2fpEYPAjdR+1TJL0RrEcSu3hGqWPn/9t9fkq8gr01nXfiHEshowY9wPbR3YnX/8cOqjKVaqrQqUqkiRv78zatPFnHf33aaFsxpRvVL5iZXXt0cfc5pMla5RtnTxxTEsWzdf8xcv1Xo0qFstq12to8TqzTxYdOXxI27b+SiEdr41Xze7QsHCLLHVwsNP7VYto+ve/m9vOX75lcW7s539HlUvlVoXiOS229aJcLlc0u777abf5HH3uj3/q44YVVKpgNgrpNiTOc6TfuHFD6dOnj9L+8OHDaD8FRfK3acN6vf9mRX3YuJ5mTJmg4OAg87K9u3fKMCJ08/o1tfygthrUelNDB/TWtav+MW7v0kU/7d65Q8VKlPovhg/YJJPJpDmff6gJC7bo2Nmr0fZJn8ZD04Y008dDFiow6FG0faqUzqMGNYqrx5fLY7XfqmXyKI9veu34O/oTtPRpPFSzYiEtWL0zdgcCm0B2vx6uXL6kWzdvqlTZpx9cu3t4qEChIjpy+JAkKVXq1Mrqm10b1q9RUFCgwsLCtOaH5fJMk1Z58xeQJGXNll2pUqfWutU/KjT0kUKCg7Vu9Q/yzZ5DGb0zW+yzf88uqvVmJXVs21J//L71vztYwAbFJrslqfdHb+vSb2O0c2l/9fzwTdnbv/Rjq6xyd3XWiZ9H6tSGz7R8Qnvlz5ExQfaDhEF2J1+FixbT3j275HfhvKTIKVgOHdxv/qA8IiJCf+34XVmz+qr7J+30bvWKatuqiX7/7VeL7QQHBWnowL7qO2BwrD/UfvAgQClTporX4wGSsthm9xPvVymitKnctGjNrhj75MiSTjXK59cff5+2aH9RLu86dE7vVyksb6/I92jlUrmVO1t6/brr2EscGRJKnK9IL1WqlNavX6+uXbtKkjnEZ8+erTfeiPmKiGeFhIQoJCTEsu2RnfmWYyQdNWq+pwyZvJXOy0tnTp3UjMkTdPHCeX3x1TeSIk/qIyIitGjebHXrM0Du7u6aNX2yenVur/nf/yjHZ24P79S2hU4eP6ZHjx6pTv1G+rhjl5h2CyR7vT+qobDwCE1dui3GPjNHttSslTu0/6ifsmZKE2V5mlRumjWipT4avEABD4Nj3E5Kdxed+eULOTs6KDwiQt1HL9PW3cej7duydlkFBAZr9daDcT0kJKIEy+4we7Lbhty+dVOSlCaN5V1iadKm1e2bkctMJpMmTZ+tAb266a2KZWRnZydPzzQaP+Vb84m1m5ubpsycrwG9umr+7BmSJJ+s2TRhykw5OET+6ejqmkJde/VVkaIlZLIzaduWzRrQq5u+HD9JlapU/68OGbApscnuaUt/14FjF3Xn/kOVK5pDI7vWUUavVOr/9Y/xOpZTF66rw4jFOnLyslJ6uKpHqzf12/zeKvnBF7psZbo42I4Ey+5wB7I7kX34UTs9fPBQTeq/Jzt7e0WEh6tj5+6qWau2JOnO7VsKDAzUwnmz1aFzN3Xu3ku7/tyhAb27a+rM+SpRqrQkaeLXX6pw0eKq/MyzUqw5fPCAft20UeMnTU+wYwOSmthk97Na13tDm3ceizZLf5vfS8XyZZGLs6Nmr9yhkdPXm5fFJpd7jVmhqUOa6cymLxQaGq4II0KffLZUf+7nLiRbEudC+qhRo/Tuu+/q6NGjCgsL0zfffKOjR4/qr7/+0u+///7iDUgaPXq0RowYYdHWZ8Bg9f10aFyHg0RWp0Ej8/c5c+VR2nRe6tHpY12+5KfMPlkVERGhsLAwde87QGXKVZAkDftirOq9U1X79+1R2TcqmNcfPmqcAgMDdebkCU2b9LWWLpqvFq3b/ufHBCS24vmzqHOzqirffEyMfT5pVkUeKVz01dxNMfaZNqSZlm3c98LgDXgYorJNR8vd1VnVyubVmN4NdO7SrSjTvkjSh3XLadmGfQp5FBb7A0KiS6js7jtwiPoPIruTEsMwNO7Lz+WZJo2mz1koZ2cXrV29Uv16dNacRcuUzstLIcHBGj1yiIoUK64Ro79SRHiEliyapz7dO2nuomVydnFRak9PNWvZxrzdAgUL6+aNG1qyYB6FdLyWYpPdkjTpu6d3bhw5dUWPQsM0ZVAzDZm0Vo9C4y9bdx8+p92Hz5lf7zp0Vgd/GKKPP6igkdPWW1kTtiKhsrvfp0M0YNCwhBgyYmnLpo36ZcNPGjnqK2XPmUunThzXhHGjIx86WqeeIiIiJ06uXLW6mrVsLUnKkze/Dh86qFUrl6lEqdLavm2r9u3ZrYXf/2BtV2ZnTp9Sv55d9HH7TyzOwYHXWWyz+4nM6VOrxhv51bL/3GiXt+o/V+5uLiqSJ7NG9ainnh++qfELIu8kiU0uf9K0isoU9lXD7jPk539bFUvk0sQBkXOk/7b7xCseLeJLnAvpFStW1MGDB/Xll1+qcOHC2rRpk0qUKKGdO3eqcOHCsdrGwIED1atXL4u2e48S5pZG/LcKFIr8P3Dp4kVl9slqvsXMN/vTuaE8PdMoVerUUaZ3eTIPa/YcORUeEa6vvhihpi1by97e/j8aPWAbKhTPqfRp3HXy56cPInFwsNeXvRqoS4tqyvfeMFUtnUdli2TXvd0TLdb9c3E/fb9hn9oNXaQqZfLovSqF1aNV5FUqJpNJ9vZ2Ctj7jTp/vlQLH9+OZhiGzl6MvFr18MnLyps9o/q2fTtKIb1C8ZzKmz2jWg2Yl4BHj4SQUNn9IIzfz7YkTdrIK9Fv376pdF5Pb/G+feuWcufNJ0n6e89u/fXH7/pl2065ubtLkvrmH6q9u3bq559W68OP2mnTxvXyv3JFM+cvkZ1d5N9nI0aN1TtVymv771tV451a0e6/YKHC2rv7r4Q8RMBmxSa7o7P3n/NydLRXNu80OnXheoKNLywsQodOXFTOLDzTIKlIqOwODI9zCQDxbPLEcfrwo/+pRs3IPM2VO4/8/a9o4bxZeq9OPaX2TC17Bwf55rCcX9k3Rw4devwcsb/37tblSxdVo3I5iz4D+/RQ0eIlNX32AnPbuTOn1aVDW9Vt2Eht23VM4KMDko64ZneruuV0695D/fT74Wi3d+naXUnS8bNXZWdnp6mDm2nioi3mD8ee9Xwuuzg7akTX2mrSa5Y27vhXUuQH7kXy+kRevU4h3Wa8VIrmzJlTs2bNeumdOjs7R7mdLDgg9KW3B9tx6kTkdBBPHj5auGhxSZLfhfNKnyFy/qf79+7p3t27ypgpU4zbMR5fyW4YEZIo1OD1smT9Xm19LijXTeusJev3mIvfvceu1PCpP5mXZ/JKpZ+md1GrAfO095/zkqSqrb+Wvd3TDynfr1pEvdu8pWptxkd5Gvmz7EwmOTtFjYfW9d7Q30f99M9J6w8ghm1KiOwOfcidCbbEO7OP0qZLp317ditP3vySpIcPHujokcOq36iJJJmfY2Kys5xf187OTsbjP/KDg4NlZ2eymIPXZLKTyRSZzzE5efI4Dx7Fays22R2donl9FB4eoRu3E/ahoHZ2JhXM5a1fXvDANdiWhMju8MDwVx0WXlFwcJBMJssLCe3t7BTxOGMdHZ1UoEAh+V04Z9Hn4oXzypTJW5L04Uf/U536H1gsb9Gorrr37q9KVaqZ286eOaXO7duqVu266tSlRwIcDZB0xTW7P6xTTkt+2qOwsJj/Hn7Czs4kRwd72dmZoi2kP5/Ljg72cnJ0UIRh2Tc8PEJ2djwXw5a80sfRhmHot99+U1BQkMqXLy9PT8/4Ghf+A4GBgbp80c/82v/yZZ06cVwpU6VShoyZdP/ePV276q+bNyKvjnkS5GnSplPadOl0+ZKfNm/8WW9UqKSUqVLrzKmTmjx+jIqWKKVcufNKkrJm81XFKtU1adyX6jtomNzc3PXt1InK6ptdJUqVkSRt2vCTHBwclCNXbjk5Oun4sX/17dRvVP3td+Tg4CggOXJzdbK4Ksw3c1oVyZNZd+4H6uLVO7p976FF/9CwcF27ed98tdrFq3cslj8IjJz/8uzFG+Y51k6cu2bRp0SBrIowDB098/RukD5t39b+f/109tINOTs5qGbFgmr+Xhl1G/29xboebi5qUKO4Boxf9WoHjkRHdidtgYEPdckiuy/p5IljSpkylTJm8lbj5q20YPa3ypI1q7y9fTRz+mSl80qvylUj70wpVKSYPFKm1OdDP9VH7TtFTu3y40pduXxJ5StVliSVLvuGpk4cp3FffqZGTVoowjC0aN5s2ds7qESpspKkn9etloOjo7lg//vWX7V+zSoNGDJSQHL1qtldtkh2lS6UTb/vO6WAh8EqVyS7xvRpqKU/79XdgCDzevlyZJSTg708U7nJI4WziuSJfMjv4Wc+yH7S5pbCWek83VUkT2Y9CgvX8ccPShvYvqb2HD6vMxdvKLWHq3q2fktZM6XRvFXcNZIUkd3JS8XK1TR/zrfKmCmTsufMpZPHj2npdwv0fr0G5j4tWrfV4P69VKxEKZUsVUa7/tqhHdu3aeqs+ZKktOm8ov3wOmOmTPLO7CMpcjqXLu0/UtnyFdS8ZWvdunlDkmRnZy/PNFGfrwQkR6+a3U9ULZNH2X3SRZujTd8tpdCwcB05fUUhj8JUskBWfda1jlZu+ttcdH9RLgc8DNb2fac0qkc9BQWHys//tiqVzKUW75dR//Hx+xwVvJpYF9Lv3r2r7t27a//+/SpXrpy+/vpr1apVS3/9FfmPnj59em3atElFihRJsMEifp04ekTdOj6dg3zKhLGSpJrv19Wg4V9ox/bfNHrEYPPy4Z/2lSR91K6T2nboLAcHR+3bs0srli5ScFCQ0mfIqCrVa6j1xx0s9jN4xChNHj9G/Xp0lp2dScVKlNK4STPMRXJ7e3stXjBXF/3OS4ahDJm81aBxMzVu/mEC/wSAxFOiQDZtmt3d/Hpsn4aSpEVrd6n9sO/+s3G4uTjpm08bK3P61AoKCdXJ89fUdvACrdy036Jfo3dKyiSTlm/c95+NDa+O7E5+jh/9V13af2R+PWl8ZHbXql1Xg0eMUsvWHys4KEhjPh+uBwEBKlKshMZP+dZ8RWJqT0+Nn/Ktvp3yjbp2aKuwsDBlz5FLYyZMUe48kdO/+GbPobETp2ruzGlq36aFTHYm5cmbX+OnfGsxZcz8WTN01d9f9g72yuabXSO/HKfqb73zH/40gP/Wq2Z3yKNQNXqnpAZ1rCVnRwedv3JLkxf/pkmLtlr0Wz25k7J5pzW/3r1soCTJtXiXKG2SVLJAVjWtVVoXrtwy34bu6ZFC04Y2V4a0HrpzP0gHjvmpWpvx5kI7bBfZnfz17j9IM6dN0lejRurOndtK55Ve9T5orI/bdzL3qVr9LfUfNEwL5s7ShLGjlDWbr0Z/NVHFipeM9X62/vqL7ty5rY3r12nj+nXm9oyZvLX651/j9ZgAWxVf591t6pXXzoNndPL8tSjLwsIj1KtNDeXOll4mk0l+/rc1fdl2TX7muSixyeUPB8zVyK51NX9Ua3mmTCE//9saPvUnzVqx42UOHQnEZBhG1HsMovG///1P27dvV+vWrbVu3brIW4ANQxMnTpSdnZ369esnd3d3rVu37sUbi8Z1pnYBbEa2yj0TewgAHgs6MOWl103o7L7F1C6AzfCp2COxhwDgMVvO7jtM7QLYDO8K3V/cCcB/IrbZHetCeubMmbVkyRJVqVJFly9fVpYsWbR161ZVrVpVkrRnzx7VqVNHV6++3FUOFNIB20EhHbAdr3IyntDZTSEdsB0U0gHbYcvZTSEdsB0U0gHbEdvstntxl0jXrl1Tnjx5JEWGu4uLi7JkyWJenjVrVt24cSOOwwQAAAmF7AYAIGkhuwEAsF2xLqRHRETI3t7e/Nre3l4m09Mnxz77PQAASHxkNwAASQvZDQCA7Yr1w0Ylafbs2XJ3d5ckhYWFaf78+UqXLp0kKSAgIP5HBwAAXgnZDQBA0kJ2AwBgm2I9R7qvr2+sPv0+d+7cSw2EOdIB28Ec6YDteJV5VhM6u5kjHbAdzJEO2A5bzm7mSAdsB3OkA7Yjttkd6yvSz58//7JjAQAAiYDsBgAgaSG7AQCwXbGeIx0AAAAAAAAAgNcRhXQAAAAAAAAAAKygkA4AAAAAAAAAgBUU0gEAAAAAAAAAsIJCOgAAAAAAAAAAVsS5kG5vb6/r169Hab9165bs7e3jZVAAACD+kN0AACQtZDcAALYnzoV0wzCibQ8JCZGTk9MrDwgAAMQvshsAgKSF7AYAwPY4xLbjpEmTJEkmk0mzZ8+Wu7u7eVl4eLi2b9+ufPnyxf8IAQDASyG7AQBIWshuAABsV6wL6RMmTJAU+cn4jBkzLG4nc3Jykq+vr2bMmBH/IwQAAC+F7AYAIGkhuwEAsF2xLqSfO3dOklStWjX9+OOP8vT0TLBBAQCAV0d2AwCQtJDdAADYrlgX0p/47bffEmIcAAAggZDdAAAkLWQ3AAC2J84PG23YsKHGjBkTpX3s2LFq1KhRvAwKAADEH7IbAICkhewGAMD2xLmQvn37dtWqVStK+7vvvqvt27fHy6AAAED8IbsBAEhayG4AAGxPnAvpDx48kJOTU5R2R0dH3b9/P14GBQAA4g/ZDQBA0kJ2AwBge+JcSC9cuLCWLVsWpf37779XgQIF4mVQAAAg/pDdAAAkLWQ3AAC2J84PGx0yZIgaNGigM2fOqHr16pKkLVu2aOnSpVqxYkW8DxAAALwashsAgKSF7AYAwPbEuZBeu3ZtrV69WqNGjdLKlSvl6uqqIkWK6Ndff1WVKlUSYowAAOAVkN0AACQtZDcAALbHZBiGEV8bO3LkiAoVKvRS614PCI2vYQB4Rdkq90zsIQB4LOjAlATd/qtk962HYfE8GgAvy6dij8QeAoDHbDm77wSGx/NoALws7wrdE3sIAB6LbXbHeY705wUEBGjmzJkqU6aMihYt+qqbAwAACYzsBgAgaSG7AQBIfC9dSN++fbs+/PBDZcqUSePGjVP16tW1a9eu+BwbAACIR2Q3AABJC9kNAIDtiNMc6VevXtX8+fM1Z84c3b9/X40bN1ZISIhWr17Nk8MBALBBZDcAAEkL2Q0AgG2K9RXptWvXVt68eXX48GFNnDhRV65c0eTJkxNybAAA4BWQ3QAAJC1kNwAAtivWV6Rv2LBB3bp1U6dOnZQ7d+6EHBMAAIgHZDcAAEkL2Q0AgO2K9RXpO3bsUEBAgEqWLKmyZctqypQpunnzZkKODQAAvAKyGwCApIXsBgDAdsW6kF6uXDnNmjVL/v7+6tChg77//nt5e3srIiJCmzdvVkBAQEKOEwAAxBHZDQBA0kJ2AwBgu0yGYRgvu/KJEyc0Z84cLVq0SHfv3lWNGjW0du3al9rW9YDQlx0GgHiWrXLPxB4CgMeCDkyJ1+3FZ3bfehgWr2MD8PJ8KvZI7CEAeMyWs/tOYHi8jg3Ay/Ou0D2xhwDgsdhmd6yvSI9O3rx5NXbsWF26dElLly59lU0BAID/ANkNAEDSQnYDAGAbXumK9PjEFemA7eCKdMB2xPdVbfGJK9IB28EV6YDtsOXs5op0wHZwRTpgO/6TK9IBAAAAAAAAAEjuKKQDAAAAAAAAAGAFhXQAAAAAAAAAAKygkA4AAAAAAAAAgBUU0gEAAAAAAAAAsIJCOgAAAAAAAAAAVlBIBwAAAAAAAADACgrpAAAAAAAAAABYQSEdAAAAAAAAAAArKKQDAAAAAAAAAGAFhXQAAAAAAAAAAKygkA4AAAAAAAAAgBUU0gEAAAAAAAAAsIJCOgAAAAAAAAAAVlBIBwAAAAAAAADACgrpAAAAAAAAAABYQSEdAAAAAAAAAAArKKQDAAAAAAAAAGAFhXQAAAAAAAAAAKygkA4AAAAAAAAAgBUU0gEAAAAAAAAAsIJCOgAAAAAAAAAAVlBIBwAAAAAAAADACgrpAAAAAAAAAABYQSEdAAAAAAAAAAArKKQDAAAAAAAAAGAFhXQAAAAAAAAAAKygkA4AAAAAAAAAgBUU0gEAAAAAAAAAsIJCOgAAAAAAAAAAVlBIBwAAAAAAAADACgrpAAAAAAAAAABYQSEdAAAAAAAAAAArKKQDAAAAAAAAAGAFhXQAAAAAAAAAAKygkA4AAAAAAAAAgBUU0gEAAAAAAAAAsIJCOgAAAAAAAAAAVjgk9gCeSOnqmNhDAPCEnX1ijwBAEuDmbDN/RgBwck3sEQBIAlyd+DsfsBku7ok9AgBxxBXpAAAAAAAAAABYQSEdAAAAAAAAAAArKKQDAAAAAAAAAGAFhXQAAAAAAAAAAKygkA4AAAAAAAAAgBUU0gEAAAAAAAAAsIJCOgAAAAAAAAAAVlBIBwAAAAAAAADACgrpAAAAAAAAAABYQSEdAAAAAAAAAAArKKQDAAAAAAAAAGAFhXQAAAAAAAAAAKygkA4AAAAAAAAAgBUU0gEAAAAAAAAAsIJCOgAAAAAAAAAAVlBIBwAAAAAAAADACgrpAAAAAAAAAABYQSEdAAAAAAAAAAArKKQDAAAAAAAAAGAFhXQAAAAAAAAAAKygkA4AAAAAAAAAgBUU0gEAAAAAAAAAsIJCOgAAAAAAAAAAVlBIBwAAAAAAAADACgrpAAAAAAAAAABYQSEdAAAAAAAAAAArKKQDAAAAAAAAAGAFhXQAAAAAAAAAAKygkA4AAAAAAAAAgBUU0gEAAAAAAAAAsIJCOgAAAAAAAAAAVlBIBwAAAAAAAADACgrpAAAAAAAAAABYQSEdAAAAAAAAAAArKKQDAAAAAAAAAGAFhXQAAAAAAAAAAKygkA4AAAAAAAAAgBUU0gEAAAAAAAAAsIJCOgAAAAAAAAAAVlBIBwAAAAAAAADACgrpAAAAAAAAAABYQSEdAAAAAAAAAAArKKQDAAAAAAAAAGAFhXQAAAAAAAAAAKygkA4AAAAAAAAAgBUU0gEAAAAAAAAAsOKVCumGYcgwjPgaCwAASGBkNwAASQvZDQCAbXipQvrChQtVuHBhubq6ytXVVUWKFNGiRYvie2wAACCekN0AACQtZDcAALbFIa4rjB8/XkOGDFGXLl1UoUIFSdKOHTvUsWNH3bx5Uz179oz3QQIAgJdHdgMAkLSQ3QAA2B6TEcd7xLJnz64RI0boww8/tGhfsGCBhg8frnPnzr3UQILDXmo1AAnAs2z3xB4CgMeC/v7mlbdBdgPJn2eFvok9BACPBe3+6pW3QXYDyZ9npQGJPQQAjwXt/DJW/eI8tYu/v7/Kly8fpb18+fLy9/eP6+YAAEACI7sBAEhayG4AAGxPnAvpuXLl0vLly6O0L1u2TLlz546XQQEAgPhDdgMAkLSQ3QAA2J44z5E+YsQINWnSRNu3bzfP1fbnn39qy5Yt0QY9AABIXGQ3AABJC9kNAIDtifMV6Q0bNtTu3buVLl06rV69WqtXr1a6dOm0Z88e1a9fPyHGCAAAXgHZDQBA0kJ2AwBge+L8sNGEwkNPANvBw0YB2xEfDxtNKGQ3YDt42ChgO+LjYaMJhewGbAcPGwVsR2wfNhrnqV0kKSIiQqdPn9b169cVERFhsaxy5covs0kAAJCAyG4AAJIWshsAANsS50L6rl271Lx5c124cEHPX8xuMpkUHh4eb4MDAACvjuwGACBpIbsBALA9cS6kd+zYUaVKldL69euVKVMmmUymhBgXAACIJ2Q3AABJC9kNAIDtiXMh/dSpU1q5cqVy5cqVEOMBAADxjOwGACBpIbsBALA9dnFdoWzZsjp9+nRCjAUAACQAshsAgKSF7AYAwPbE6or0w4cPm7/v2rWrevfuratXr6pw4cJydHS06FukSJH4HSEAAIgzshsAgKSF7AYAwLaZjOefXBINOzs7mUymKA85MW/k8bJXeehJcNhLrQYgAXiW7Z7YQwDwWNDf37zUemQ38HrxrNA3sYcA4LGg3V+91HpkN/B68aw0ILGHAOCxoJ1fxqpfrK5IP3fu3CsNBgAA/LfIbgAAkhayGwAA2xarQnq2bNnM32/fvl3ly5eXg4PlqmFhYfrrr78s+gIAgMRBdgMAkLSQ3QAA2LY4P2y0WrVqun37dpT2e/fuqVq1avEyKAAAEH/IbgAAkhayGwAA2xPnQvqTOdmed+vWLbm5ucXLoAAAQPwhuwEASFrIbgAAbE+spnaRpAYNGkiKfMBJmzZt5OzsbF4WHh6uw4cPq3z58vE/QgAA8FLIbgAAkhayGwAA2xXrQnqqVKkkRX4y7uHhIVdXV/MyJycnlStXTu3atYv/EQIAgJdCdgMAkLSQ3QAA2K5YF9LnzZsnSfL19VWfPn24nQwAABtHdgMAkLSQ3QAA2K5YF9KfGDZsWEKMAwAAJBCyGwCApIXsBgDA9sS5kC5JK1eu1PLly+Xn56dHjx5ZLNu/f3+8DAwAAMQfshsAgKSF7AYAwLbYxXWFSZMm6aOPPlKGDBl04MABlSlTRmnTptXZs2f17rvvJsQYAQDAKyC7AQBIWshuAABsT5yvSJ82bZpmzpypZs2aaf78+erXr59y5MihoUOH6vbt2wkxRvyH/t63V/PnztGxo0d048YNTZg0VdXffEuSFBoaqimTJmrHH9t16dJFebi7q+wb5dW9Z2+lT5/BvI1unTvqxPHjun37llKmTKWyb7yhHr36WPT5ZePPmjPzW124cF6enmnUtHkLtWn7v//8eIHEUqF4TvX8sLpK5M+iTF6p1Lj3bK3b9k+0fScNbKx2H1RQ33E/asrS3yVJWTOl0cD/vaOqpXMrQ1oP+d+8r6U/79OYOZsUGhYuSRrUvqYGd4h6ovUwKETpKvYzv07l7qrhnd9T3epFlCalm/z8b6vv16v0y59H4zxW2CayO/maM+tbbdm8SefOnZWzi4uKFSuuHr36yDd7DknS5cuXVOvtN6Nd96vxE/X2O+/qxPHjmjt7pg4c+Ft379yRd+bMatS4qVq0am3ue+PGdX09doz+/feILvpdUPMWrdRv4KD/5BgBW1GhWHb1bFlVJfJljszDvvO1bvu/0fad1L+B2jV4Q30nrNGU73eY2/u1qa53K+RXkTzeehQarkxvDbVYr3DuTOrzYTWVL5pdaVO56YL/bc1etUtTlz3dxswhTdTq/VJR9nn07FWVbPa1JMk9hbOGdXhHdaoUkpenuw6dvKw+49fo72OX4uNHgf8A2Z28WTvvlqTpUydr44b1unr1qhwdHVWgQEF16d5TRYoUtdjO9t+36dvpU3Xq5Ak5OTurVKnSmjh5miRpzaofNXTwwGj3v3X7X0qbNm3CHSBgIyoUy66eLSqrRN7MyuSVUo37L9S67Uej7TupXz21q19OfSeu05Rlf0qSsmb01MC21VW1ZM7I8+4b97X0lwMaM/8383n3Ez2aV1LbumWUNaOnbt17qG9/2KWxC36TJJUvkk2fd35XebJ5KYWLk/yu3tGc1Xs0+Zm/EdrVL6t2DcopWyZPSdKxs9c0au4Wbdp1MiF+NHhJcS6k+/n5qXz58pIkV1dXBQQESJJatWqlcuXKacqUKfE7QvyngoIClTdvXtVr0FC9unexWBYcHKzjx46qfcdOyps3n+7fv68xo79Q9y6dtHT5j+Z+pcuU0//ad1Q6Ly9dv3ZN48eNVZ+e3bVw8feSpB1//K5P+/dV/08Hq3z5ijp79oxGDhssZ2cXNWvR8j89XiCxuLk66Z+Tl7Vw7W4tG/dxjP3qVCuiMoWz6cr1uxbteX3Ty87OpC6jlunMxZsqmDOTpg5uKjdXJw2cuEaSNHHRVs3+4U+L9X6e3ll/H/Uzv3Z0sNf6aZ/o+p0Ateg3T5ev31PWTJ66FxAU57HCdpHdyde+vXvUpFkLFSxcWOFh4Zr8zXh1bPexfly7XilSpFDGjJm0ZdsOi3VWrlimBfPmqGLFypKko0ePKE3aNBr15VfKmDGTDh7cr8+GD5Wdnb05lx89eiTPNJ5q36GTFi2c/18fJmAT3Fyd9M+pK1q4bq+WjW0dY786VQqpTKFsunL9XpRlTo4O+nHLYe3+54Ja1ykTZXnxfD66ceeBPhq2VJeu3VW5Ir6aOrChwsMjNGPlX5KkPuPXaMjUn83rODjYafd3PfXjlsPmtumffqACOTOq7fCl8r95X81qltD6Ke1Vouk4Xblx/1V+DPiPkN3Jm7XzbknKls1XAwcNlY9PFgWHBOu7hfPVqV1brduwWWnSpJEk/brpF40YNkRde/RUmbLlFB4WrtOnnxbc3nm3lipUrGSx3SGDBujRo0cU0fHacHNx1D+n/LXwp31a9mWrGPvVqVJQZQpm1ZUbltmd19dLdiaTuoxZpTOXbqlgjgyaOrBB5Hn35KdZ/HXP2nqzbG4NnPyzjpy5qjQpU8gzpat5+cPgUM1YuVP/nPbXw6BQlS+aTVP6N9DDoEeau2aPJOnyjfsaMm2jTl+8KZPJpJa1SmjF2A9VrvUkHTt3PZ5/MnhZcS6kZ8yYUbdv31a2bNmUNWtW7dq1S0WLFtW5c+dkGEZCjBH/oYqVqqhipSrRLvPw8NC3s+dZtA0cNEQtmjaS/5UryuTtLUlq1bqNebm3d2a1/bidenTrrNDQUDk6OuqntWtVrfqbatykmSTJJ0sWtW3XQfPmzlLT5i1kMpkS5uAAG7Lpr2Pa9Ncxq328vVJpfN+Gqt1lulZ9095i2eadx7V553Hz6/OXbynPoq1q90EFcyH9YdAjPQx6Op9m4dzeKpAzk7qNWm5ua123nDxTpVDVthMUFhYhSfLzt7zKKTZjhW0ju5Ov6TPnWLwe+cWXqlbpDR07+q9Kliote3t7pfPysuizdcuvervmu0rh5iZJqt/gA4vlPlmy6PDBg9ry6yZzIT1zZh/1HzhYkrR61Q8JdTiATdu084Q27TxhtY+3V0qN71NXtbvN1qrxbaMs/3zWJklSy/eiXlEuSQvX7bV4ff7KbZUtnE11qxU2F9LvPwzW/YfB5j61KxeUp4erFv0Uua6Ls4PqVSusRv3m68+D5yRJX8zerFqVCqhdgzc04ttfYnnESExkd/Jm7bxbkmq9X9vidZ9+A7Xqh5U6dfKEypZ7Q2FhYRrz5Rfq2aevGjRsZO6XM1cu8/cuLi5ycXExv759+7b27N6t4Z99Ho9HAti2TbtOvvCKbm+vlBrfq45q95ijVV9/ZLFs866T2vzM+uev3FaeJX+oXf2y5kJ63mxeategnEq2mKBTfjclSRf871hs59DJKzp08or5td/VO6pXtZAqFPU1F9J/3mF5zj38201q16CcyhTKSiHdhsR5jvTq1atr7dq1kqSPPvpIPXv2VI0aNdSkSRPVr18/3gcI2/bgwQOZTCZ5pEwZ7fJ7d+9q/fp1KlqsuBwdHSVFXtXm5Oxs0c/F2UXXrl7VlSuXE3zMQFJgMpk057OWmrBoq46dvRqrdVK6u+j2/cAYl39U7w2dPH9Nfx48a257r3Ih7T58XhP7N9L5TZ9r37IB6vtRDdnZ8YFWckJ2vz4ePL5iMWWqVNEuP/rvEZ04fixK8fx5AQ8ClCpV6vgeHpCsmUwmzRneTBO++13Hzl2Lt+2mcnPRHSv53rpOGW3de1p+V+9Kkhzs7eXgYK/gkDCLfsEhoSpfNHu8jQsJi+zGE6GPHumHFcvk4eGhPHnzSpKOHT2q69euyc7OTo0b1tObVSrqkw7/06lTMRcM161dLVdXF9V4u+Z/NXTA5plMJs0Z2kQTFm+PdbE6pZuLbt9/egf3exXz69zl26pVIb+O/dBPx3/sr2kDG1pckf68onm8VbZwNv1x4Fy0y+3sTGr0VhG5uThp9z9+0fZB4ojzFekzZ85URETkVYudO3dW2rRp9ddff6lOnTrq0KFDrLYREhKikJAQizbD3lnOzxVXYdtCQkI0cfw4vVvrPbm7u1ssm/D1V/p+6WIFBwWpSNFimjxthnlZ+QoV9dXY0dq9a6dKlykrP78LWrhgriTp5o0bypzZ5z89DsAW9W7zpsLCIzT18ZzoL5LDJ506Na1svhr9ec5ODmrybkl9Pf9Xi/bsPmlVNVNufb/hb9XvNkM5s3hp4oBGcnSw16hZG1/5OGAbyO7XQ0REhMaOGaVixUsod+480fZZ9cNK5ciRU8WKl4hxOwcP7NemjRs0edq3CTVUIFnq/WHVyOxetuPFnWOpXOFs+qBGUdXvNTfa5ZnSpdQ7b+RVm6FLzG0PAkO06/B5DWz7lk6cv65rtwPU+O3iKlsom85cuhlvY0PCIrvx+7bf1L9PLwUHBymdl5dmzJorT8/IaV0uXbooSZoxdYr69Bsg78yZtXD+PP2vTSutXf+LUqVOHWV7q39YqXdrvW9xlTrwuuvdqorCwsM1dfmfL+4sKYdPWnVqVF4DJ683t/lmTqOsGVOrQfXC+t/I5bKzs9PYHu9ryRct9W7XWRbrn14zUOlSu8nB3k6fz/lV85+7E61gzgzaNvMTuTg56EHQIzUZsEjHz3M1ui2J0xXpu3bt0pAhQzRw4EBt3BhZYGnatKkmTZqkrl27ysnJKVbbGT16tFKlSmXx9dWY0XEfPRJNaGio+vbqLsMwNGjoiCjL27T9WMtWrtKMWXNlZ2enwQP7m29BbNiosZo2a6Gun3RQqWKF1KpZE9V89z1JkskuzjdJAMlO8Xw+6ty0itoPWxyr/t5eqbR2Skf9+OtBzVu1M9o+dasVkYebi777yTKo7Uwm3bjzQJ2/+F4Hjl/Sys0HNHbuJv3vg/KvfBywDWT362PU5yN05tQpjR03IdrlwcHB2vDzT6rXMOar0U+dOqkeXT9Rh06dVb5CxYQaKpDsFM+XWZ2bVFL7kcvibZsFcmTQ8q/a6IvZm7Vld/RXmbZ4r6TuPgjW2t8tH3radvj3Mpmks+uH6N4fo9W5cQUt33RQERFMCZIUkN2QpNJlymr5D6u1cPH3qlCxkvr27qFbt25JkozHH7L8r31HvfX2OypQsJBGfjFaJpNJmzZFvRjm0MEDOnv2jOpb+RsAeN0Uz5tZnRtXUPvPV8Sqv7dXSq2d8JF+3PqP5q19el5tZzLJxdlRH49crj8PndcfB86q0xcrVbVUTuXOms5iG292nKEKbSer69hV6tKkghrXsHyA8MkLN1W29SRV/t80zVq1S7OGNFI+3/SvfrCIN7G+In3lypVq0qSJXF1d5ejoqPHjx2vMmDHq06dPnHc6cOBA9erVy6LNsOdT8aQiNDRUfXv3kP+VK5o1b0GUq9ElydMzjTw908jXN7ty5Mipt9+sosOHDqposeIymUzq2buvuvXopZs3byqNp6d2744s/vn4ZPmvDwewORWK51T6NO46uX64uc3BwV5f9qynLs2rKF/tkeb2TOlSauO3XbTr0Dl1/jzmk/c29d7Qhj/+1fXbARbtV2/eV2hYuMWJ9fFz15QpXSo5OthHeRI5khay+/Ux6vOR2v77Ns1d8J0yZMwYbZ/NmzYqKChYtevUi3b5mdOn1f7jNmrYqInad/wkAUcLJD8VimVXek83nVzzqbnNwcFeX3arrS5NKilf/bgVL/NlT6+fp3bQ3NW7NWbelhj7ta5dWks3/B0lr89dvqW3O81QChdHpXRz0dVbAVr0eQudu3I7hi3BVpDdeCJFihTKmi2bsmbLpiJFi6n2u29r9Y8r9XG7Dubnn+TImdPc38nJSZl9suiqv3+Ubf34wwrlzZdfBQoW+s/GD9i6CsV8I7N71QBzm4ODvb7s+p66NKmofA3GmNszpfPQxinttOsfP3X+8keL7Vy9FaDQsHCdvvj0rq/jFyKvIs+SIbV53nTp6dzp/565pvRpPDTo47e0fPMh8/LQsHCdvRT5gdmBE5dVMr+POjepoK5jVsXjkeNVxLqQPnr0aLVr105Tp06Vvb29Ro8erVGjRr1UoDs7R72dLDgshs6wKU+K6H4XLmj2vIVKndrzhes8uSXx0aNHFu329vbKkCGDJGnDz+tVtFhx8xPIgdfZkp/3auseyyvP1k3pqCU/79PCtbvNbd5eqbTx2y46cOyi2o9YEuODp7J5p1GVUrn0Qa/ZUZbtPHROTWqWkMlkMq+fO1t6+d+4RxE9GSC7kz/DMDT6i8+0dctmzZm/yOoH0qt//EFVq1WPNmtPnz6ldm1bq06deuravWdCDhlIlpb8vF9b95yyaFv3TTst2fC3Fv60L07byp89gzZM66DF6//W8BkxT7NWqUQO5cripflrF8TYJzA4VIHBoUrt4aq3yuXVoCnrY+wL20B2IyYRRoT5nLpAwUJycnLS+fPnVKJk5MOLQ0NDdeXKZWXK5G2xXuDDh9q0cYO69ej9n48ZsGVLNhzQ1r2nLdrWTWyrJRsOaOH6p9nt7ZVSG6e004Hjl9X+8xVRzrt3Hr4gRwd7Zc+cRucuR35gnTtL5IddT55fEh07O5OcnayXZe1MdnJ2jPOs3EhAsf7XOHHihJYtWyZ7e3tJUu/evTV06FBdv35d6dNzm0FyEfjwofz8nj7I4PKlSzp+7JhSpUqldF5e6tOzm44dO6rJU79VRHi4bt64IUlKlSqVHJ2cdPjwIf37zz8qXqKkUqZKqYt+fpo2+RtlyZJVRYsVlyTduXNbmzf9otKlyygk5JHWrP5Bm3/ZqDnzv0uUYwYSg5urk3I+DldJ8vVOqyJ5MuvO/UBdvHpHt+9ZPlQsNCxc127e16nHn2x7e6XSLzO7ys//tgZOXCMvz6d3hly7ZXnVeeu65XT15n398ufRKOOYtXKHOjaupK/7NNC0ZduVK6uX+n5UQ9O+fzo3+4vGCttFdid/oz4boQ0//6SJk6fJLYWbOZfdPTws5kD1u3BBf+/bq6nTZ0bZxqlTJ9WubWuVr1BRrVp/ZN6Gnb29RdH9+LFjkqTAwIe6c+e2jh87JkdHR+XMlSshDxGwGW6uTsrp8/QWbV/vNCqS2zsyD6/djfLA79CwcF27HaBTfjfMbVkypJZnyhTKkjG17O1MKpI7suB15tJNPQx6pAI5MmjD1I76dfcJTVqyXRnSeEiSwiMidPPuQ4vtt6lTRnuOXNDRs1EfbPpW2TwymUw6eeG6cmZJp1Fd39fJC9e18Lm5WGF7yO7Xg7Xz7lSpU2v2zBmqWq260nl56e6dO/p+6WJdv3ZNNd6JfFCou7u7GjVuqulTJytjxkzy9vbW/HlzJElvv2P5MNGNG39WeHi43qtd5787QMBGRGZ3WvPryOzO9Di770WT3RGPszvyKnJvr5T6ZWp7+V29o4FTfpZXajdz32u3H0iStu49rf3HL+nbQR+o78SfZGcyaWKfuvp190nzVeodGpbTxWv3dOLxfOcVi2dXj+aVNG35X+btjez0jn7ZeVIXr96Vh5uTmrxdTJVLZFftHtE/JwWJI9aF9MDAQKVMmdL82snJSS4uLnrw4AGBnoz8++8R/e+jD82vx42NvA21Tt366ti5i7b9tlWS1LhhXYv1Zs9bqNJlysrVxUVbft2k6VMnKygoUOm8vFShYiWN7fCJxVx+69as1vivxsqQoaJFi2n2/EUqXKTIf3CEgG0oUSCrNs3san49tnd9SdKidbvVfviSmFYzq14ur3Jl9VKurF46s3GkxTLXkt3N35tMJrV6v4wWrdsT7byol67dVZ0u0zW2d33t/b6/rty4p6lLf9fXC54+lPRVx4rEQ3Ynf8uXLZUkfdymlUX7yM9Hq279BubXq1f9oAwZMuqNaOY9/3XTL7pz+7bWr1ur9evWmtu9vTNrw+at5tdNPqhn/v7ov//q5/U/RekDJGcl8vto0/RO5tdje0YWpRb9tE/tP4vd3OhD2r+jVu+XMr/e/V3kHSBvd5quP/afVf3qRZQ+jbuav1tSzd8tae534cpti+lhUrq5qF61wuozPvqHjKdyd9HIT2opc/pUun0/UGt++0fDpm9UWHhE7A8YiYLsfj1YO+8ePGyEzp07q7VrVununTtKnTq1ChYqrHkLFytXrtzmdXr26Sd7BwcNGthPIcHBKlykqGbNXaCUqVJZ7Gv1jz/ozbdqWPy/Al4XJfL5aNO09ubXY7u/L0latP7vWM2NXr10buXKkk65sqTTmbWfWixzfSNyShjDMPRB3wUa36uuNk/roIfBj7Rp5wkNeOaBpHYmO43s+I58vdMoLDxCZy/f0uCpGzR79R5zHy9Pd80Z2lgZ03ro3oNgHTnjr9o95ka5ah6Jy2TENBfAc+zs7PT5559bzIfdv39/9e3bV+nSPb0yo1u3bi81EG4xA2yHZ9nuL+4E4D8R9Pc3L70u2Q28Pjwr9E3sIQB4LGj3Vy+9LtkNvD48Kw14cScA/4mgnV/Gql+sC+m+vr4ymUzWN2Yy6ezZs7Ha8fMIdMB2UEgHbMerFNLJbuD1QSEdsB2vUkgnu4HXB4V0wHbEtpAe66ldzp8//7JjAQAAiYDsBgAgaSG7AQCwXXaJPQAAAAAAAAAAAGwZhXQAAAAAAAAAAKygkA4AAAAAAAAAgBUU0gEAAAAAAAAAsIJCOgAAAAAAAAAAVsS5kG5vb6/r169Hab9165bs7e3jZVAAACD+kN0AACQtZDcAALYnzoV0wzCibQ8JCZGTk9MrDwgAAMQvshsAgKSF7AYAwPY4xLbjpEmTJEkmk0mzZ8+Wu7u7eVl4eLi2b9+ufPnyxf8IAQDASyG7AQBIWshuAABsV6wL6RMmTJAU+cn4jBkzLG4nc3Jykq+vr2bMmBH/IwQAAC+F7AYAIGkhuwEAsF2xLqSfO3dOklStWjX9+OOP8vT0TLBBAQCAV0d2AwCQtJDdAADYrlgX0p/47bffEmIcAAAggZDdAAAkLWQ3AAC2J84PG23YsKHGjBkTpX3s2LFq1KhRvAwKAADEH7IbAICkhewGAMD2xLmQvn37dtWqVStK+7vvvqvt27fHy6AAAED8IbsBAEhayG4AAGxPnAvpDx48kJOTU5R2R0dH3b9/P14GBQAA4g/ZDQBA0kJ2AwBge+JcSC9cuLCWLVsWpf37779XgQIF4mVQAAAg/pDdAAAkLWQ3AAC2J84PGx0yZIgaNGigM2fOqHr16pKkLVu2aOnSpVqxYkW8DxAAALwashsAgKSF7AYAwPbEuZBeu3ZtrV69WqNGjdLKlSvl6uqqIkWK6Ndff1WVKlUSYowAAOAVkN0AACQtZDcAALbHZBiGEV8bO3LkiAoVKvRS6waHxdcoALwqz7LdE3sIAB4L+vubBN0+2Q0kD54V+ib2EAA8FrT7qwTdPtkNJA+elQYk9hAAPBa088tY9YvzHOnPCwgI0MyZM1WmTBkVLVr0VTcHAAASGNkNAEDSQnYDAJD4XrqQvn37dn344YfKlCmTxo0bp+rVq2vXrl3xOTYAABCPyG4AAJIWshsAANsRpznSr169qvnz52vOnDm6f/++GjdurJCQEK1evZonhwMAYIPIbgAAkhayGwAA2xTrK9Jr166tvHnz6vDhw5o4caKuXLmiyZMnJ+TYAADAKyC7AQBIWshuAABsV6yvSN+wYYO6deumTp06KXfu3Ak5JgAAEA/IbgAAkhayGwAA2xXrK9J37NihgIAAlSxZUmXLltWUKVN08+bNhBwbAAB4BWQ3AABJC9kNAIDtinUhvVy5cpo1a5b8/f3VoUMHff/99/L29lZERIQ2b96sgICAhBwnAACII7IbAICkhewGAMB2mQzDMF525RMnTmjOnDlatGiR7t69qxo1amjt2rUvta3gsJcdBYD45lm2e2IPAcBjQX9/E6/bI7uB5MmzQt/EHgKAx4J2fxWv2yO7geTJs9KAxB4CgMeCdn4Zq36xviI9Onnz5tXYsWN16dIlLV269FU2BQAA/gNkNwAASQvZDQCAbXilK9LjE5+MA7aDK9IB2xHfV6THJ7IbsB1ckQ7Yjvi+Ij0+kd2A7eCKdMB2/CdXpAMAAAAAAAAAkNxRSAcAAAAAAAAAwAoK6QAAAAAAAAAAWEEhHQAAAAAAAAAAKyikAwAAAAAAAABgBYV0AAAAAAAAAACsoJAOAAAAAAAAAIAVFNIBAAAAAAAAALCCQjoAAAAAAAAAAFZQSAcAAAAAAAAAwAoK6QAAAAAAAAAAWEEhHQAAAAAAAAAAKyikAwAAAAAAAABgBYV0AAAAAAAAAACsoJAOAAAAAAAAAIAVFNIBAAAAAAAAALCCQjoAAAAAAAAAAFZQSAcAAAAAAAAAwAoK6QAAAAAAAAAAWEEhHQAAAAAAAAAAKyikAwAAAAAAAABgBYV0AAAAAAAAAACsoJAOAAAAAAAAAIAVFNIBAAAAAAAAALCCQjoAAAAAAAAAAFZQSAcAAAAAAAAAwAoK6QAAAAAAAAAAWEEhHQAAAAAAAAAAKyikAwAAAAAAAABgBYV0AAAAAAAAAACsoJAOAAAAAAAAAIAVFNIBAAAAAAAAALCCQjoAAAAAAAAAAFZQSAcAAAAAAAAAwAoK6QAAAAAAAAAAWEEhHQAAAAAAAAAAKyikAwAAAAAAAABgBYV0AAAAAAAAAACsoJAOAAAAAAAAAIAVFNIBAAAAAAAAALCCQjoAAAAAAAAAAFZQSAcAAAAAAAAAwAqTYRhGYg8CyUNISIhGjx6tgQMHytnZObGHA7zWeD8CiA1+VwC2g/cjgNjgdwVgO3g/vn4opCPe3L9/X6lSpdK9e/eUMmXKxB4O8Frj/QggNvhdAdgO3o8AYoPfFYDt4P34+mFqFwAAAAAAAAAArKCQDgAAAAAAAACAFRTSAQAAAAAAAACwgkI64o2zs7OGDRvGAxYAG8D7EUBs8LsCsB28HwHEBr8rANvB+/H1w8NGAQAAAAAAAACwgivSAQAAAAAAAACwgkI6AAAAAAAAAABWUEgHAAAAAAAAAMAKCukAAAAAAAAAAFhBIR3RatOmjerVq5fYw5AkDR8+XMWKFUvsYQA2JbHeo7wfAdtFdgO2jewG8DyyG7BtZDeeRyE9CWnTpo1MJpNMJpOcnJyUK1cujRw5UmFhYf/5WLZt22Yei8lkkqurqwoWLKiZM2fG+7769OmjLVu2xPt2gfhmy+/RDBkyqGHDhjp79myst2EymbR69eqEGyTwGrDl3wtkN2Db71GyG0gctvx7gewGbPs9SnYnfw6JPQDETc2aNTVv3jyFhITo559/VufOneXo6KiBAwdG6fvo0SM5OTkl6HhOnDihlClTKigoSOvWrVOnTp2UM2dOvfnmm/G2D3d3d7m7u8fb9oCEZIvvUQ8PD506dUrt27dX7dq1dfjwYdnb2yfofgE8ZYu/F8hu4ClbfI+S3UDissXfC2Q38JQtvkfJ7tcDV6QnMc7OzsqYMaOyZcumTp066a233tLatWslPb3l5IsvvpC3t7fy5s0rSbp48aIaN26s1KlTK02aNKpbt67Onz9v3mZ4eLh69eql1KlTK23atOrXr58Mw4jVeNKnT6+MGTMqe/bs6tatm7Jnz679+/ebl0dERGj06NHKnj27XF1dVbRoUa1cudK8/Mmnd1u2bFGpUqWUIkUKlS9fXidOnDD3ef6WlrCwMHXr1s083v79+6t169YWt9tUrVpV3bp1U79+/ZQmTRplzJhRw4cPj8NPGng5tvgezZQpkypXrqyhQ4fq6NGjOn36tPbu3asaNWooXbp0SpUqlapUqWLx3vX19ZUk1a9fXyaTyfz6iUWLFsnX11epUqVS06ZNFRAQYF4WEhKibt26KX369HJxcVHFihW1d+/el/hpAsmDLf5eILuBp2zxPUp2A4nLFn8vkN3AU7b4HiW7Xw8U0pM4V1dXPXr0yPx6y5YtOnHihDZv3qyffvpJoaGheuedd+Th4aE//vhDf/75p9zd3VWzZk3zel9//bXmz5+vuXPnaseOHbp9+7ZWrVoVp3EYhqGNGzfKz89PZcuWNbePHj1aCxcu1IwZM/Tvv/+qZ8+eatmypX7//XeL9QcNGqSvv/5a+/btk4ODg9q2bRvjvsaMGaPFixdr3rx5+vPPP3X//v1ob4NZsGCB3NzctHv3bo0dO1YjR47U5s2b43RcwKuylffok7FIkZ/IBwQEqHXr1tqxY4d27dql3Llzq1atWuZgfhLA8+bNk7+/v0UgnzlzRqtXr9ZPP/2kn376Sb///ru+/PJL8/J+/frphx9+0IIFC7R//37lypVL77zzjm7fvh33HyCQDNnK7wWyG4ierbxHn4xFIruBxGYrvxfIbiB6tvIefTIWiexOtgwkGa1btzbq1q1rGIZhREREGJs3bzacnZ2NPn36mJdnyJDBCAkJMa+zaNEiI2/evEZERIS5LSQkxHB1dTV++eUXwzAMI1OmTMbYsWPNy0NDQw0fHx/zvqLz22+/GZIMNzc3w83NzXBwcDDs7OyMzz//3NwnODjYSJEihfHXX39ZrPvxxx8bzZo1s9jOr7/+al6+fv16Q5IRFBRkGIZhDBs2zChatKh5eYYMGYyvvvrK/DosLMzImjWrxXirVKliVKxY0WK/pUuXNvr37x/jMQGvyhbfo3fu3DEMwzCuXLlilC9f3sicObPF/p8IDw83PDw8jHXr1pnbJBmrVq2y6Dds2DAjRYoUxv37981tffv2NcqWLWsYhmE8ePDAcHR0NBYvXmxe/ujRI8Pb29viGIDXhS3+XiC7gads8T1KdgOJyxZ/L5DdwFO2+B4lu18fzJGexPz0009yd3dXaGioIiIi1Lx5c4tbpwoXLmwx99OhQ4d0+vRpeXh4WGwnODhYZ86c0b179+Tv72/xabaDg4NKlSoVq1tY/vjjD3l4eCgkJER79uxRly5dlCZNGnXq1EmnT59WYGCgatSoYbHOo0ePVLx4cYu2IkWKmL/PlCmTJOn69evKmjWrRb979+7p2rVrKlOmjLnN3t5eJUuWVERERIzbfLLd69evv/CYgFdha+9RHx8fGYahwMBAFS1aVD/88IOcnJx07do1DR48WNu2bdP169cVHh6uwMBA+fn5vXCbvr6+FuN99r115v/s3Xd0VEUfxvFn00MIoUNCC016r9KLKEWqgjQpKgLSRLpIsQGiSC9SRF4EpKgoIB0RpCMoKEVKIHRIKAmkEJJ5/wBW1iRLAonZwPdzTs5h586dnbu6++z+9u7cEycUFRWlqlWrWre7urqqYsWKOnz48EPHBp5Ejva6QHYDthztOUp2AynP0V4XyG7AlqM9R8nupweF9FSmdu3amj59utzc3OTn5ycXF9v/hF5eXja3b968qXLlymnBggWxxsqSJctjzydv3rxKnz69JKlYsWLatWuXPv74Y3Xv3l03b96UJK1atUo5cuSw2c/d3d3mtqurq/XfFotFkmIFdGI9OOb9cR93TOBhHO05unXrVqVLl05Zs2a1CeGOHTsqODhYEydOVJ48eeTu7q5nn33W5udw8eG5BSSOo70ukN2ALUd7jpLdQMpztNcFshuw5WjPUbL76UEhPZXx8vJSgQIFEty/bNmyWrx4sbJmzap06dLF2cfX11e7du1SjRo1JN29qMhvv/2msmXLJnp+zs7OCg8PlyQVLVpU7u7uCgwMVM2aNRM9Vlx8fHyULVs27dmzxzrf6Oho7du3z+bCKEBKcbTn6INvuh+0bds2TZs2TQ0bNpR098IrQUFBNn1cXV0VHR2d4GORpPz588vNzU3btm1Tnjx5JElRUVHas2eP3n777USNBTwpHO114d/IbjztHO05SnYDKc/RXhf+jezG087RnqNk99ODi40+4dq1a6fMmTOradOm2rp1qwICArR582b17t1bZ8+elST16dNHY8aM0fLly3XkyBG99dZbun79eoLGv3z5si5evKjTp09r6dKlmj9/vpo2bSpJ8vb2Vv/+/dW3b1/NmzdPJ06c0L59+zR58mTNmzfvkY+pV69eGj16tH744QcdPXpUffr00bVr16zfqAOpSXI/R+NTsGBBzZ8/X4cPH9auXbvUrl0760VR7vP399fGjRt18eJFXbt2LUHjenl5qXv37howYIDWrFmjQ4cOqUuXLgoLC9Prr7/+WHMGnhZkN+DYyG4A/0Z2A46N7EZS4Yz0J1yaNGm0ZcsWDRo0SC1atFBoaKhy5MihunXrWr+F69evny5cuKCOHTvKyclJr732mpo3b64bN248dPxChQpJurt2VK5cudS1a1ebdak+/PBDZcmSRaNHj9bJkyeVPn16lS1bVu++++4jH9OgQYN08eJFdejQQc7OznrzzTf1wgsvyNnZ+ZHHBFJKcj9H4zNnzhy9+eabKlu2rHLlyqVRo0apf//+Nn3GjRund955R7NmzVKOHDl06tSpBI09ZswYxcTE6NVXX1VoaKjKly+vtWvXKkOGDI88X+BpQnYDjo3sBvBvZDfg2MhuJBWLSciq+YADi4mJUZEiRdSqVSt9+OGHKT0dAADwEGQ3AACpC9kNAJyRjlTo9OnTWrdunWrWrKnIyEhNmTJFAQEBatu2bUpPDQAAxIHsBgAgdSG7ASA21khHquPk5KSvvvpKFSpUUNWqVXXw4EFt2LBBRYoUSempAQCAOJDdAACkLmQ3AMTG0i4AAAAAAAAAANjBGekAAAAAAAAAANhBIR0AAAAAAAAAADsopAMAAAAAAAAAYAeFdAAAAAAAAAAA7KCQDgAAAAAAAACAHRTSAQAAAAAAAACwg0I6AAAAAAAAAAB2UEgHAAAAAAAAAMAOCukAAAAAAAAAANhBIR0AAAAAAAAAADsopAMAAAAAAAAAYAeFdAAAAAAAAAAA7KCQDgAAAAAAAACAHRTSAQAAAAAAAACwg0I6AAAAAAAAAAB2UEgHUrkTJ06oa9euypcvnzw8PJQuXTpVrVpVEydOVHh4uE3f6Oho+fn5yWKxaPXq1XGON3LkSFksFutfmjRplDt3bjVu3Fhz585VZGRkrH06depks0/atGmVL18+vfzyy/r2228VExOTLMcOAMCT4KuvvrJm6K+//hpruzFGuXLlksVi0YsvvmhtfzB7nZyc5Ofnp+eff16bN2+ONUZ0dLTmzp2rWrVqKWPGjHJ3d5e/v786d+6svXv3JufhAQDwxHoww+P6GzVqlCwWi2bPnh3vGOvXr5fFYtGkSZMkxf58/eCfh4fHf3VoAOLgktITAPDoVq1apZYtW8rd3V0dOnRQ8eLFdfv2bf36668aMGCA/vrrL82cOdPaf9OmTbpw4YL8/f21YMECNWjQIN6xp0+frrRp0yoyMlLnzp3T2rVr9dprr2nChAlauXKlcuXKZdPf3d3d+uYgPDxcp0+f1ooVK/Tyyy+rVq1a+uGHH5QuXbrkeSAAAHgCeHh4aOHChapWrZpN+y+//KKzZ8/K3d091j716tVThw4dZIxRQECApk2bpjp16mjVqlXWnA8PD1eLFi20Zs0a1ahRQ++++64yZsyoU6dOacmSJZo3b54CAwOVM2fO/+Q4AQB40nzwwQfKmzdvrPZ69epp7NixWrhwod5444049124cKGcnZ3VunVra9uDn68f5OzsnHSTBpBoFNKBVCogIECtW7dWnjx5tGnTJvn6+lq39ejRQ8ePH9eqVats9vn6669VtmxZdezYUe+++65u3bolLy+vOMd/+eWXlTlzZuvt4cOHa8GCBerQoYNatmypnTt32vR3cXFR+/btbdo++ugjjRkzRkOGDFGXLl20eB8fMXkAAHXmSURBVPHixz1sAACeWA0bNtTSpUs1adIkubj88zZ94cKFKleunIKCgmLt88wzz9jkb/PmzVWyZElNmDDBWkgfMGCA1qxZo/Hjx+vtt9+22X/EiBEaP3588hwQAABPiQYNGqh8+fJxbnv55Zc1d+5cnT9/Xn5+fjbbIiIi9P3336tevXrKmjWrtT2uz9cAUh5LuwCp1NixY3Xz5k3NmTPHpoh+X4ECBdSnTx/r7fDwcH3//fdq3bq1WrVqpfDwcP3www+Jus927drpjTfe0K5du7R+/foE7TN48GA9//zzWrp0qf7+++9E3R8AAE+TNm3aKDg42CZjb9++rWXLlqlt27YJGqNEiRLKnDmzAgICJElnz57VF198oXr16sUqokt3z2zr378/Z6MDAJBM2rdvr5iYGH3zzTextq1atUo3btxQu3btUmBmABKLQjqQSq1YsUL58uVTlSpVEtT/xx9/1M2bN9W6dWtlz55dtWrV0oIFCxJ9v6+++qokad26dYnaxxiT4OI7AABPI39/fz377LNatGiRtW316tW6ceOGzc+97bl27ZquXbumTJkyWfe/c+eONb8BAEDSu3HjhoKCgmz+goODJUk1atRQzpw5tXDhwlj7LVy4UGnSpFGzZs1ibfv3eEFBQQoJCUnuQwFgB4V0IBUKCQnRuXPnVKJEiQTv8/XXX6tKlSrWtc1bt26tdevW6cqVK4m67+LFi0u6e5HT5NwHAICnUdu2bbV8+XLrBcMXLFigmjVrxvop+H0REREKCgrSlStXtHv3brVs2VLR0dFq2bKlJOnw4cOSlKj3DAAAIHGee+45ZcmSxeYvR44ckiQnJye1adNGv/32m82vtENCQvTTTz+padOmSps2rc14t27dijVelixZ1KpVq//0uADYYo10IBW6/y20t7d3gvoHBwdr7dq1NmugvvTSS+rRo4eWLFmiHj16JPi+7wd8aGhosu4DAMDTqFWrVnr77be1cuVK1a9fXytXrtSkSZPi7T9nzhzNmTPHetvDw0PvvPOOdRmXxL5nAAAAiTd16lQ988wzNm0PXhi0ffv2+vTTT7Vw4UKNHDlSkvTtt98qIiIizmVdPDw8tGLFiljtD17HDMB/j0I6kAqlS5dOUsIL04sXL1ZUVJTKlCmj48ePW9srVaqkBQsWJKqQfvPmTUmJ+0D+KPsAAPA0ypIli5577jktXLhQYWFhio6O1ssvvxxv/6ZNm6pnz56yWCzy9vZWsWLFbC4kntj3DAAAIPEqVqwY78VGJalkyZIqXry4Fi1aZC2kL1y4UJkzZ9YLL7wQq7+zs7Oee+655JougEdEIR1IhdKlSyc/Pz/9+eefCep/fy30qlWrxrn95MmTypcvX4LGun+fBQoUSFD/R90HAICnVdu2bdWlSxddvHhRDRo0UPr06ePtmzNnTrsftAsXLixJOnjwoEqXLp3EMwUAAAnVvn17DR48WHv37lXOnDn1888/q2vXrnJxoTQHpBaskQ6kUi+++KJOnDihHTt22O0XEBCg7du3q2fPnlq6dKnN3+LFi+Xm5hbnRU/iM3/+fEmK81tze/tYLBbVq1cvwfsAAPC0at68uZycnLRz5061bdv2scZq0KCBnJ2d9fXXXyfR7AAAwKNo06aNLBaLFi5cqMWLFys6OjrOZV0AOC6+9gJSqYEDB2rBggV64403tGnTJmXLls1m+4kTJ7Ry5UrrT7kHDhxovdDog2bPnq0FCxbovffee+h9Lly4ULNnz9azzz6runXrJmieY8aM0bp169S6dWsVLFgwQfsAAPA0S5s2raZPn65Tp06pcePGjzVWrly51KVLF82YMUOTJ09Wr169bLbHxMRo/PjxeuWVV5QzZ87Hui8AABC/3Llzq3r16lq8eLH8/PyUN29eValSJaWnBSARKKQDqVT+/Pm1cOFCvfLKKypSpIg6dOig4sWL6/bt29q+fbuWLl2qTp06acuWLSpdunScRXRJatKkiXr16qV9+/apbNmy1vZly5Ypbdq0un37ts6dO6e1a9dq27ZtKlWqlJYuXRprnDt37ljPdouIiNDp06f1448/6sCBA6pdu7ZmzpyZPA8EAABPoI4dOybZWOPGjdOJEyfUu3dvfffdd3rxxReVIUMGBQYGaunSpTpy5Ihat26dZPcHAMDTZvXq1Tpy5Eis9ipVqtgso9q+fXu9+eabOn/+vIYOHRrveA9+vv635s2b21wPBcB/h0I6kIo1adJEBw4c0KeffqoffvhB06dPl7u7u0qWLKlx48apQoUKmjlzpoYNGxbvGI0bN1avXr309ddf2xTSu3fvLunu1cIzZ86s0qVL68svv1Tbtm3l7u4ea5zIyEi9+uqrkqQ0adIoa9asKleunIYPH279iToAAPjvpUmTRqtXr9ZXX32lefPm6cMPP1RYWJj8/PxUp04dLViwQDly5EjpaQIAkGoNHz48zva5c+faFNJffvll9erVS5GRkXaXdXnw8/W/BQQEUEgHUojFGGNSehIAAAAAAAAAADgqThEFAAAAAAAAAMAOCukAAAAAAAAAANhBIR0AAAAAAAAAADsopAMAAAAAAAAAYAeFdAAAAAAAAAAA7KCQDgAAAAAAAACAHRTSAQAAAAAAAACwg0I6Uq1p06bJYrGoUqVKKT2VeJ04cUIeHh6yWCzau3dvrO3r169XtWrVlCZNGmXIkEEvv/yyTp06leDxY2JiNH36dJUuXVqenp7KlCmT6tSpoz/++MPa5/r162rXrp0yZMigfPnyac6cObHG2bt3r9KkSaOAgIBHOk5JOnXqlCwWS5x/lStXtvY7evSo+vbtqypVqlgfm8Qcsz0//vijypYtKw8PD+XOnVsjRozQnTt3Erz/iRMn1LZtW2XNmlWenp4qWLCghg4datNn1qxZqlmzprJlyyZ3d3flzZtXnTt3jnUMX331VbyPh8Vi0YIFC5LikAEgVXnas7tTp05xZkLhwoVt+pHdSZvdCX3cjxw5ooEDB6p06dLy9vaWr6+vGjVqFOf/BwDwtHgSsvu3337Tiy++qOzZsytt2rQqWbKkJk2apOjo6IeOndDPf5GRkerVq5eyZMminDlz6qOPPoo11tmzZ5U2bVpt27btkY9VUrzZnT17dmufCxcuaPDgwapdu7a8vb1lsVi0efPmx7rf+7Zv3259L5Q9e3b17t1bN2/eTPD+ly5dUteuXZUjRw55eHjI399fr7/+uk2fkSNHxnmMHh4ecY45Z84cFSlSRB4eHipYsKAmT578WMcIxMclpScAPKoFCxbI399fu3fv1vHjx1WgQIGUnlIsffv2lYuLiyIjI2NtW7lypZo2baqyZctqzJgxCgkJ0cSJE1WtWjXt379fWbJkeej4r732mhYsWKAOHTqoZ8+eunXrlvbv36/Lly9b+/Tv31+bN2/W+++/r+PHj6tLly4qUqSIqlSpIkkyxqh37956++23lTdv3sc+5jZt2qhhw4Y2bQ8ey44dOzRp0iQVLVpURYoU0e+///7Y9ylJq1evVrNmzVSrVi1NnjxZBw8e1EcffaTLly9r+vTpD93/999/V61atZQjRw7169dPmTJlUmBgoM6cOWPTb//+/cqbN6+aNGmiDBkyKCAgQLNmzdLKlSv1xx9/yM/PT5JUo0YNzZ8/P9b9jB8/Xn/88Yfq1q2bJMcNAKkJ2S25u7tr9uzZNm0+Pj42t8nupM1uKWGP++zZszVnzhy99NJLeuutt3Tjxg198cUXqly5stasWaPnnnvu8Q4YAFKh1J7dv/32m6pUqaKCBQtq0KBBSpMmjVavXq0+ffroxIkTmjhxot2xE/r579NPP9X//vc/DR06VKGhofrggw+UP39+tWnTxjrWgAED1KRJE1WtWvWxj7levXrq0KGDTZunp6f130ePHtUnn3yiggULqkSJEtqxY8dj36d0N3vr1q2rIkWK6PPPP9fZs2f12Wef6dixY1q9evVD9z9z5oz1+Lt166YcOXLo/Pnz2r17d5z9p0+frrRp01pvOzs7x+rzxRdfqFu3bnrppZf0zjvvaOvWrerdu7fCwsI0aNCgRzxSIB4GSIVOnjxpJJnvvvvOZMmSxYwcOTKlpxTLmjVrjJubm3nvvfeMJLNnzx6b7UWLFjUFChQwkZGR1rbff//dODk5mXfeeeeh4y9evNj6GNiTLVs2M2/ePOvtmjVrmsGDB1tvz58/3/j5+ZnQ0NCEHlqcAgICjCTz6aef2u0XHBxsQkJCjDHGfPrpp0aSCQgIeKz7Nubu41mqVCkTFRVlbRs6dKixWCzm8OHDdveNjo42xYsXN5UqVTJhYWGJvu+9e/caSWb06NF2+4WFhRlvb29Tr169RN8HAKR2ZLcxHTt2NF5eXg/tR3YnbXYn9HHfu3dvrMc0KCjIZMmSxVStWvWh+wPAk+ZJyO4uXboYNzc3ExwcbNNeo0YNky5duke6z7g+/1WqVMm8//771tsdO3Y0rVu3tt7eunWr8fLyMmfOnHmk+3yQJNOjRw+7fUJCQqzHvHTpUiPJ/Pzzz4993w0aNDC+vr7mxo0b1rZZs2YZSWbt2rUJ2j9v3rwmKCjIbr8RI0YYSebKlSt2+4WFhZlMmTKZRo0a2bS3a9fOeHl5matXrz50TkBisLQLUqUFCxYoQ4YMatSokV5++eV4l8m4fv26+vbtK39/f7m7uytnzpzq0KGDgoKCrH0iIiI0cuRIPfPMM/Lw8JCvr69atGihEydOWPtcuHBBR44cUVRUVILmFxUVpT59+qhPnz7Knz9/rO1Xr17VoUOH1Lx5c7m5uVnbS5UqpSJFiuibb7556H18/vnnqlixopo3b66YmBjdunUrzn7h4eHKkCGD9XbGjBkVFhYmSbp165YGDx6s0aNH23zLm5wyZswob2/vBPVN6ON+6NAhHTp0SG+++aZcXP75oc1bb70lY4yWLVtmd/9169bpzz//1IgRI+Tp6amwsLAE/czvPn9/f0l3/3+zZ8WKFQoNDVW7du0SPDYAPCnI7n9ER0crJCQk3u1kd/Jk98Me93LlysV6TDNlyqTq1avr8OHDdscGgCdRas9uSQoJCZGHh4fSp09v0+7r62tzBndixPX5z152x8TEqE+fPho4cKBy5sz5SPeZWN7e3sqYMWOC+gYFBenIkSPW+cYnJCRE69evV/v27ZUuXTpre4cOHZQ2bVotWbLE7v5HjhzR6tWrNWDAAGXKlEkREREP/W9tjFFISIiMMXFu//nnnxUcHKy33nrLpr1Hjx66deuWVq1aZXd8ILEopCNVWrBggVq0aCE3Nze1adNGx44d0549e2z63Lx5U9WrV9fkyZP1/PPPa+LEierWrZuOHDmis2fPSrr7gerFF1/U+++/r3LlymncuHHq06ePbty4oT///NM61pAhQ1SkSBGdO3cuQfObMGGCrl27pvfeey/O7fd/chZXcKdJk0bnz5/XxYsX4x0/JCREu3fvVoUKFfTuu+/Kx8dHadOmVb58+WKFV4UKFfT555/r2LFjWrt2rdasWaOKFStKkkaNGqUcOXLo1VdfTdBxJURYWJiCgoJs/hL6RujfEvq479+/X5JUvnx5m3Y/Pz/lzJnTuj0+GzZskHT3Z9/ly5eXl5eX0qRJo9atW+vq1atx7hMcHKzLly9r79696ty5syQ9dLmWBQsWyNPTUy1atLDbDwCeRE97dt8XFhamdOnSycfHRxkzZlSPHj1irStKdid9difkcY/PxYsXlTlz5gT1BYAnSWrPbkmqVauWQkJC1LVrVx0+fFinT5/WjBkz9N1332nIkCEJfiwe9vmvQoUKmjlzpg4ePKgdO3Zo0aJF1uyeM2eOgoKCNGDAgATf38NERETEyu64lrZJiClTpqhIkSLxLq9y38GDB3Xnzp1Y2e3m5qbSpUsnOLuzZcumunXrytPTU56enmrQoEG8117Jly+ffHx85O3trfbt2+vSpUs22+N7P1GuXDk5OTk9dE5AoqXsCfFA4t3/GdX69euNMcbExMSYnDlzmj59+tj0Gz58eLxLn8TExBhjjPnyyy+NJPP555/H28eYuz/LUgJ/xnzhwgXj7e1tvvjiC2OMMXPnzo31E7Po6GiTPn16U7duXZt9g4KCjJeXl5Fk9u7dG+997Nu3z0gymTJlMtmyZTPTpk0zCxYsMBUrVjQWi8WsXr3a2vfAgQMmZ86cRpKRZF566SUTHR1tTp48aTw9Pc2OHTseekwJcf/n4XH9xfcTsof9PDyhj/v9cQIDA2Ntq1ChgqlcubLd/Zs0aWJ9PNu1a2eWLVtmhg0bZlxcXEyVKlVs/l+4z93d3Xp8mTJlMpMmTbJ7H8HBwcbNzc20atXKbj8AeBKR3XcNHjzYDBo0yCxevNgsWrTIOseqVavaLG9Cdidtdif0cY/Lli1bjMViMcOGDbPbDwCeNE9CdhtjzJ07d0zPnj2Nq6urNeOcnZ3N9OnTH3ofD3rY578zZ86YYsWKWftUr17dhIaGmuvXr5ssWbKYb775JlH3Z0982T137tw4+z9saZf7y6g8bOmX++Ns2bIl1raWLVua7Nmz292/d+/e1sevfv36ZvHixebTTz81adOmNfnz5ze3bt2y9p0wYYLp2bOnWbBggVm2bJnp06ePcXFxMQULFrRZVqZHjx7G2dk5zvvLkiWLzfI6QFKgkI5Up2/fviZbtmzmzp071rZ+/frFaitWrJgpVaqU3bEaNWpkMmfO/NAPUYnRoUMHU6pUKRMdHW2MiT/QBw0aZCSZwYMHm7///tvs3bvX1KlTxxrwW7dujfc+tmzZYg3LnTt3WttDQ0NN5syZY63jGR4ebvbs2WOOHTtmbWvevLlp3769McaYb7/91pQsWdL4+/ub999/P87C8cPc/zD+5ptvmvXr19v8xbcuWVKts/rBBx8YSebSpUuxtlWvXv2h/x/UqVPHSDL169e3aR89erTNm8cHbdq0yfz0009m3LhxpkyZMg9dH/2LL74wkswPP/zw8AMCgCcM2R2/jz/+2EgyixYtsmknu0vZ3f9RsvtB8T3uD7p06ZLJmTOnyZcv32OvRw8Aqc2Tkt3GGDN+/Hjz4osvmnnz5pnFixebZs2aGRcXF/P9998n+P4S8vnv9u3bZv/+/eavv/6yzqtv376mWrVqxpi766RXrFjR5MyZ0/Tq1cvmmiuJIck0bdo0VnafP38+zv5JtUb6//73PyPJ7Nq1K9a2V1991fj4+Njd/7XXXjOSTLFixayPjzHGLFq0yEgys2bNsrv/ggULYq1N/9prrxlPT884++fKlcs0bdrU7phAYlFIR6py584d4+vra1q3bm2OHTtm/VuyZEmsi1t4eHiYdu3a2R2vcOHCSXrxqB07dhiLxWI2bdpkbYsv0CMjI83rr79unJycrEXx559/3nTr1s1IMvv374/3fvbs2WMkmbx588ba1rlzZ+Pq6mr3TcrGjRuNl5eXOXv2rDly5IhxdXU1X375pdm0aZPJli2b+fLLLxN97Am9YNmDkurD+OOe1daoUSMjyebCbsYYc/r0aSPJ5qIxcTl+/Ljx8PAwkydPjrdPjRo1TMaMGc3t27ftjgUATxqy276wsDDj5ORkXn/9dbv9yG5bj5vdD3vcb968aSpUqGB8fHzMwYMHH3I0APBkeZKye/To0SZ79uyxvhCtVauW8fPze6TifkI+/xljzOHDh427u7v57bffTHBwsEmXLp0ZNWqU2bFjhylcuLAZPnx4ou/bmIRdbPRBSVVIf9wz0nv06BFnRt+5c8e4uLiYzp07P3QO2bNnt/l1IGek47/GGulIVTZt2qQLFy7om2++UcGCBa1/rVq1kqR4L37yXxk4cKCqV6+uvHnz6tSpUzp16pT1AisXLlxQYGCgta+bm5tmz56t8+fPa8uWLTp69KjWrl2rGzduyMnJSQUKFIj3fvz8/CTdXVvs37JmzaqoqKh4Lz4aHR2tPn36aPDgwcqRI4eWLFmiKlWqqHPnzqpdu7a6du2a4o9jYvn6+kq6+xj/24ULF6yPV3ziezyzZs0qSbp27Zrd/fPnz68yZcrE+7gFBgZq69atatmypVxdXe2OBQBPGrLbPk9PT2XKlCnea3JIZHdcHje77T3ut2/fVosWLXTgwAH98MMPKl68uN2xAOBJ8yRl97Rp01SnTp1YF5Nu0qSJzp8/H+/a3PY87PPffX379lX79u1VtmxZrVq1ShkzZtSQIUNUuXJlDRw4MMUfx8RKrux2dnZWpkyZHprdkpQrVy6b7Pb19VV0dLQuX75s0+/27dsKDg5+6JyAxHJJ6QkAibFgwQJlzZpVU6dOjbXtu+++0/fff68ZM2bI09NT+fPnt7lwSVzy58+vXbt2KSoqKkkKnIGBgTp9+rTy5s0ba1uTJk3k4+Njc2Vv6W6I3A+S6Ohobd68WZUqVYoV9A/y8/NT9uzZ47wIy/nz5+Xh4SFvb+84950+fbpCQ0PVv39/a/8Hw8XPzy/BF3dxFKVLl5Yk7d2713pBF+nusZ09e1Zvvvmm3f3LlSunWbNmxTru8+fPS5KyZMny0DmEh4fHe3GXRYsWyRijdu3aPXQcAHjSkN32hYaGKigoyG7WkN2xPW52x/e4x8TEqEOHDtq4caOWLFmimjVrJvSQAOCJ8SRl96VLlxQdHR2r3/2Lat+5c+eR5mDv858krVy5Utu3b9exY8ck3c2n+4VoKXVmd/HixeXi4qK9e/dav1SR7hatf//9d5u2uJQrV06SYh337du3H/peSJKMMTp16pTKlCljbXvw/UTDhg2t7Xv37lVMTIx1O5BkUvqUeCChwsLCjLe3t3nttdfi3L5t2zYjyXoRj6S86Mn58+fN4cOHH7osx9q1a833339v89erVy8jyXz22Wdm5cqVdvcfM2aMkWSWLVtm0378+HFz/Phxm7Y+ffoYSWbdunXWtitXrph06dKZhg0bxjl+cHCwyZgxo1myZIm1bcSIEaZMmTI249aqVcvuPOOSHD8PT+jjbszdnwuWKlXKZr2+9957z1gsFnPo0CFr2/Xr183hw4fN9evXrW0XLlww7u7uplq1ajZrtQ0ZMsRIMrt37zbGGBMVFRXnmrG7du0yzs7O5tVXX41zbiVLljS5c+d+pPVrASA1I7v/ye7w8HATEhISa/8BAwbEe8zGkN3GPF52J/Zxf+utt4wk68XrAOBp86Rld/HixU3GjBlNUFCQte3OnTumXLlyxtvb2+a+/p3dj/r5LzIy0hQsWNCMHTvW2jZ37lyTJUsW61Iy48ePN/7+/naPMz5K4qVdrly5Yg4fPmxzsc/41K9f3/j6+tpk6+zZs40ks3r1amvbrVu3zOHDh82VK1esbRERESZr1qwmX758Jjw83Np+/3piD77XuXz5cqz7njp1aqz/j8LCwkzGjBnNiy++aNO3ffv2Jk2aNCY4OPihxwQkBoV0pBrffPONkWSWL18e5/bo6GiTJUsW07hxY2PM3QtvFi1a1Dg7O5suXbqYGTNmmFGjRpnKlSub33//3RhzN0Br1aplJJnWrVubqVOnmrFjx5rnn3/e5n4Sc/Xwf4tvrbb58+ebZs2amc8//9zMnDnTtGrVykgyb7zxRqwx8uTJY/LkyWPTdvHiRePr62u8vb3NiBEjzOeff26eeeYZ4+npaT2+f3vrrbdMzZo1bdoOHDhgLBaL6datmxk9erTx8PAw06ZNS/RxJvTD+PXr182HH35oPvzwQ1O/fn0jyfTr1898+OGHsdaYS8zjvmLFCmOxWEydOnXMzJkzTe/evY2Tk5Pp0qWLTb/7/z3+fUXz+xc9q1evnpk6dap58803jcViMW3atLH2uXbtmvHy8jKvvfaaGTdunJkxY4bp0aOHSZMmjcmYMaP5+++/Y83r4MGD1gvTAcDThuzOY70dEBBg0qdPb7p3724mTpxoJk6caBo2bGi9YOaDxeAHkd2Pl92JedzHjx9vJJlnn33WzJ8/P9bfzZs3H3pMAJDaPWnZ/fXXXxtJJn/+/OaTTz4xkyZNMs8++6yRZD766CObvv/O7kf5/GeMMWPHjjUFCxa0uZjopUuXTJo0aUzLli3N+PHjTcaMGc3AgQMTfZzGJLyQfj+7W7dubSSZ1157zdr2oBEjRiR4DfXffvvNuLu7mzJlypjp06eboUOHGg8PD/P888/b9Pv555+NJDNixAib9nnz5hlJpkKFCmbSpEmmf//+xtXV1VSvXt3mi3VPT0/TqVMnM27cODN16lTTpk0bY7FYTOnSpWMV/O8X2F9++WUza9Ys06FDByPJfPzxxw89HiCxKKQj1WjcuLHx8PCw+y1pp06djKurq/Xb5uDgYNOzZ0+TI0cO4+bmZnLmzGk6duxo8210WFiYGTp0qMmbN69xdXU12bNnNy+//LI5ceKEtU9yBPquXbtMjRo1TIYMGYyHh4cpVaqUmTFjRpxnLcdVSDfGmBMnTpjmzZubdOnSGU9PT1OnTh3rGVj/duDAAePm5hbnhdC++uor4+/vbzJlymTeeecdmwBLqIR+GL/fL66/fx9jYh/377//3pQuXdq4u7ubnDlzmvfeey/W2QzxfRiPiYkxkydPNs8884xxdXU1uXLlirV/ZGSk6dOnjylZsqRJly6dcXV1NXny5DGvv/56vHMcPHiwkWQOHDiQoGMAgCcJ2Z3HevvatWumffv2pkCBAiZNmjTG3d3dFCtWzIwaNSreM+/I7rseJ7sT87jfn3t8f497gVUASA2etOw2xpg1a9aYmjVrmsyZMxs3NzdTokQJM2PGjFj9/p3dj/L57+LFi8bb29v8+OOPsbatXr3aFC5c2KRPn9506NAhQWeAxyWhhXR7mfagxBTSjTFm69atpkqVKsbDw8NkyZLF9OjRI9avv+IrpBtjzKJFi0ypUqWMu7u7yZYtm+nZs2es/d944w1TtGhR4+3tbVxdXU2BAgXMoEGD4vyVmTHGzJw50xQqVMi4ubmZ/Pnzm/Hjx/OLcCQLizHGxL3oCwAAAAAAAAAAcErpCQAAAAAAAAAA4MgopAMAAAAAAAAAYAeFdAAAAAAAAAAA7KCQDgAAAAAAAACAHRTSAQAAAAAAAACwg0I6AAAAAAAAAAB2uKT0BO67ER6T0lMAcE/2Kr1TegoA7gnfPyWlpxCv0AiyG3AUWZ8luwFH4cjZHR6V0jMAcF/Gij1TegoA7klodnNGOgAAAAAAAAAAdlBIBwAAAAAAAADADgrpAAAAAAAAAADYQSEdAAAAAAAAAAA7KKQDAAAAAAAAAGAHhXQAAAAAAAAAAOygkA4AAAAAAAAAgB0U0gEAAAAAAAAAsINCOgAAAAAAAAAAdlBIBwAAAAAAAADADgrpAAAAAAAAAADYQSEdAAAAAAAAAAA7KKQDAAAAAAAAAGAHhXQAAAAAAAAAAOygkA4AAAAAAAAAgB0U0gEAAAAAAAAAsINCOgAAAAAAAAAAdlBIBwAAAAAAAADADgrpAAAAAAAAAADYQSEdAAAAAAAAAAA7KKQDAAAAAAAAAGAHhXQAAAAAAAAAAOygkA4AAAAAAAAAgB0U0gEAAAAAAAAAsINCOgAAAAAAAAAAdlBIBwAAAAAAAADADgrpAAAAAAAAAADYQSEdAAAAAAAAAAA7KKQDAAAAAAAAAGAHhXQAAAAAAAAAAOygkA4AAAAAAAAAgB0U0gEAAAAAAAAAsINCOgAAAAAAAAAAdlBIBwAAAAAAAADADgrpAAAAAAAAAADYQSEdAAAAAAAAAAA7KKQDAAAAAAAAAGAHhXQAAAAAAAAAAOygkA4AAAAAAAAAgB0U0gEAAAAAAAAAsINCOgAAAAAAAAAAdlBIBwAAAAAAAADADgrpAAAAAAAAAADYkehC+qVLl/Tqq6/Kz89PLi4ucnZ2tvkDAACOhewGACB1IbsBAHA8LondoVOnTgoMDNSwYcPk6+sri8WSHPMCAABJhOwGACB1IbsBAHA8iS6k//rrr9q6datKly6dDNMBAABJjewGACB1IbsBAHA8iV7aJVeuXDLGJMdcAABAMiC7AQBIXchuAAAcT6IL6RMmTNDgwYN16tSpZJgOAABIamQ3AACpC9kNAIDjSfTSLq+88orCwsKUP39+pUmTRq6urjbbr169mmSTAwAAj4/sBgAgdSG7AQBwPIkupE+YMCEZpgEAAJIL2Q0AQOpCdgMA4HgSXUjv2LFjcswDAAAkE7IbAIDUhewGAMDxJLqQHhISEme7xWKRu7u73NzcHntSAAAg6ZDdAACkLmQ3AACOJ9GF9PTp08tiscS7PWfOnOrUqZNGjBghJ6dEX8sUAAAkMbIbAIDUhewGAMDxJLqQ/tVXX2no0KHq1KmTKlasKEnavXu35s2bp/fee09XrlzRZ599Jnd3d7377rtJPmEAAJA4ZDcAAKkL2Q0AgONJdCF93rx5GjdunFq1amVta9y4sUqUKKEvvvhCGzduVO7cufXxxx8T6AAAOACyGwCA1IXsBgDA8ST6N2Dbt29XmTJlYrWXKVNGO3bskCRVq1ZNgYGBjz87AADw2MhuAABSF7IbAADHk+hCeq5cuTRnzpxY7XPmzFGuXLkkScHBwcqQIcPjzw4AADw2shsAgNSF7AYAwPEkemmXzz77TC1bttTq1atVoUIFSdLevXt15MgRLVu2TJK0Z88evfLKK0k7UwAA8EjIbgAAUheyGwAAx2MxxpjE7hQQEKCZM2fq6NGjkqRChQqpa9eu8vf3f+SJ3AiPeeR9ASSt7FV6p/QUANwTvn9KkoyTHNkdGkF2A44i67NkN+AoHDm7w6OSZGoAkkDGij1TegoA7klodj9SIT05UEgHHAeFdMBxJNWH8eRAIR1wHBTSAcfhyNlNIR1wHBTSAceR0OxO0NIuBw4cUPHixeXk5KQDBw7Y7VuyZMkE3TEAAEg+ZDcAAKkL2Q0AgGNLUCG9dOnSunjxorJmzarSpUvLYrEorhPZLRaLoqOjk3ySAAAgcchuAABSF7IbAADHlqBCekBAgLJkyWL9NwAAcGxkNwAAqQvZDQCAY0tQIT1Pnjxx/hsAADgmshsAgNSF7AYAwLElqJD+b8eOHdPPP/+sy5cvKybG9kJjw4cPT5KJAQCApEN2AwCQupDdAAA4lkQX0mfNmqXu3bsrc+bMyp49uywWi3WbxWIh0AEAcDBkNwAAqQvZDQCA40l0If2jjz7Sxx9/rEGDBiXHfAAAQBIjuwEASF3IbgAAHI9TYne4du2aWrZsmRxzAQAAyYDsBgAgdSG7AQBwPIkupLds2VLr1q1LjrkAAIBkQHYDAJC6kN0AADieRC/tUqBAAQ0bNkw7d+5UiRIl5OrqarO9d+/eSTY5AADw+MhuAABSF7IbAADHYzHGmMTskDdv3vgHs1h08uTJR5rIjfCYh3cC8J/IXoU35oCjCN8/5bHHSK7sDo0guwFHkfVZshtwFI6c3eFRjzojAEktY8WeKT0FAPckNLsTfUZ6QEBAoicDAABSDtkNAEDqQnYDAOB4El1Iv+/27dsKCAhQ/vz55eLyyMMgBe37bY++nveljhz+S0FXrmjs55NVq85zNn0CTp7QlInjtO+3PYq+E628+fLrk3ETld3XTzduXNfM6VO0a8c2Xbp4QekzZFTN2nXV7a3eSuvtLUm6fv2ahr87UMePHdWN69eVIWMm1axVR9179VXatGklSe8PG6JVK5bHml/efPm1+LuVyf44ACmhatn86tvhOZUtmlu+WXzUqu9Mrdh8wLp9aNeGavlCWeXMnkG3o6K1/3CgRk5ZoT1/nrb2KV04pz7q00zliuVWdLTR8o2/a9C4b3Ur/LYkqX3jSpr1watx3n/uOoN15dpNSZKbq4vefbOB2jSqoGyZvHUxKESjZq7W/37YKUlqWqeUBrz+gvLnyixXF2cdD7yiifM3atGqPcn18CCZkN1PpsuXLmnyhHHavm2LIiIilDNXbo34YJSKFisuSSpfqkic+/Xu218dOr2u8+fOafbMadq7e5eCg4OUOUtWNWzUWK916SpXVzdJ0qlTARr94UgFnDyhmzdDlSVLVr3QsJHe7NpDLv9abgB4Uj0sux80aWhrdXm5mgZ8ukxTFm62tj8su0s8k0P9O9dTldL5lSm9l06fv6rZy37V1EX/jDHz/fZ6tUnlWPd56MQFlXv5Y0nSkVXvK49fplh9Zizeor5jljzGo4D/Gtn9ZPpt7x7NmztHhw/9qStXrujziVNVp+7dz+JRUVGaOnmCft26RWfPnpF32rSqVLmKevftp6xZs1nHmPXFdG3d8ov+PnpYLq6u+nXHXpv7uH79mt4d1F/H/j6q69evK2PGTKpVp6569XnH+lkceNI9LLvjytR12w6pac9pscZyc3XRlvn9VapQTlV6ZbQO/H1O0t3P7u91axir/63wSGWu0s96u2fbWurSsrpyZc+g4Ou39P2G/Ro2+UdF3r4jiexOLRKdxGFhYerVq5fmzZsnSfr777+VL18+9erVSzly5NDgwYOTfJJIHhHh4Sr4TCE1btZCg96J/XPgs2cC1aVzOzVp9pLe7N5TXl5pdfLEcbm5u0uSgq5cVtCVy+rzzkDlzZdfFy6c15iPRiroymWN+WyiJMnJyUk1atVRtx59lCFDBp05E6hPR3+oGzdu6KMxn0mS+g18Vz36vGO93+joaLVr1Ux169X/Dx4FIGV4ebrr4N/n9L8fdmjx52/G2n789GX1/WSpAs4GydPdVb3a19GKaT1VvOn7Crp2U75ZfLRqRi8tW7dPfccsUTovD3064CXN+uBVtR0wR5K0bN0+rd9+yGbcme+/Kg93V2sRXZK+HvuasmX0Vrf3F+hE4BX5ZvGRk8Vi3X71RpjGzl6jo6cu6XZUtBpWL66ZI9vrytWb2rDjcDI9QkhKZPeTKyTkhl7v1Fbly1fSxKkzlSFDRp0JPK106dJZ+6zZuMVmn+2/btWHI99TneeelySdOnVSJsbo3WHvK2fu3Dpx/Jg+fn+4wsPD9Xa/gZIkFxcXNWrcVIWLFJW3t7f+/vuoPn5/uEyMUY/eff+7AwZS0MOy+74mtUuqYgl/nb983aY9IdldpkguXbkaqs7vzdPZi9dUuVQ+TX2vjaJjYjRj8d3ncv9Pl2nYpB+s47o4O2vX4iH6bv1+a1u19p/K2emfLC9awE8/zehl0weOjex+soWHh+mZQoXUrPlLeudt2+U9IiIidPjQIXXp2l2FChVWSEiIxo75WG/37K6FS76z9ouKilK9F+qrVOnS+v67ZbHuw8nipFq166pHr7eVIWNGnQkM1OiP39eNGzc0Zuy4ZD9GwBEkJLvXbvtLXUd8bb19v7D9b6PebqoLV26oVKGcNu0T/rdBs5dttWn76Yve+u2vf06Ce6V+eX3Yu6m6jVygHX+cVME8WTXrg1dlJA0ad/d5TXanDokupA8ZMkR//PGHNm/erPr1/yl0Pvfccxo5ciSBnopUqVZDVarViHf79CkTVLVaDfXuO8DaljNXbuu/8xd4Rp+Mm2SzrXvPtzVi6EDduXNHLi4uSpfORy+3amPt4+uXQy+3aqP58760tqX19raewS5JmzdtUGhIiBo3bf7Yxwg4qnXbDmndtkPxbl+8xvaMkkHjvlPn5lVUvKCfNu/+Ww2qF1fUnWi9PXqJ7l/qotfHi7V36bvKlyuzTp4JUkRklCIi/1kIM3OGtKpV8Rl1e3+Bta1elSKqXq6Air44UtdCwiRJgReu2tz31t+O2dyeumiz2jWupCpl8lFITyXI7ifXvC9nK1s2X434cJS1LUdO2zf3mTNnsbn9y+ZNKl+hknLmzCVJqlK1uqpUrW7dnjNnLp0+FaBvl3xjLaTnzJnL2l+6m+e/7dmt/ft+S/JjAhzVw7Jbkvyy+OjzQS3V+K2p+n5yd5ttCcnu+78Gu+/UuWBVKplXTeuUshbSQ25GKORmhLVP41ollSGdp+b/uMPaFvTAF+aS1L9zcZ0IvBIr0+G4yO4nW7XqNVWtes04t3l7e+uL2XNt2ga/O0zt27TUhQvn5evrJ0l6q+fdk+F+WP5drDEkKZ2Pj1q1bmu97eeXQ61eaat5c+ckxSEAqUJCsvv27Tu6FBxqt8/zVYuqbuUiajNgtupXK2az7Vb4besvy6S7vy4rmt9XvT/+xtpWuVRe7fj9pPVzfuCFq1qyZq8qFPe39iG7UwenxO6wfPlyTZkyRdWqVZPlgTMWixUrphMnTiTp5JByYmJitG3rL8qdx1+9ur+hF2pXVef2r2jzpg1297t5M1ReadPG+7PDK5cv6+eN61W2XIV4x/hx+beqWOlZ+frleKxjAJ4Uri7Oer1FVV0PDdPBez8fc3dzUVRUtB68XnR45N3wrlI6f5zjtHuxosIibuv7Db9b2xrVLKF9hwL1TqfndGLtRzqwfLhG920uD/f4l2qoVfEZPeOfVb/+xmt+akF2P7m2/PKzihQrpkH931a9WlXVtlULff9t/D/9DA4O0q9bf1HT5i/ZHffmzVCl8/GJd/uZwNPasf1XlS1f/pHnDjxpLBaL5nzUQePnbdThkxdjbX+U7JYkn7Qe1i+749Kx2bPatOuoAi9ci3O7q4uzWjesoHk/7IhzOxwT2Y0H3bx5UxaLRd7e6R7eOR6XL1/Sxg3rVa58/J/FgadR9fIFdXrjaP3x/TBNfPcVZfTxstmeNaO3pg1ro9eH/U9hDxTM49O5eRX9feqStu3/57V65x8BKlM0l8oXyyNJ8s+RSS9ULaY1v/4V5xhkt+NK9BnpV65cUdasWWO137p1yybg7YmMjFRkZKRtW4yr3O8tGYKUd/VqsMLCwjTvy9nq1qO3evXppx3bf9Wgfr01fdZXKlu+Yqx9rl+7pi9nTVezFq1ibXtvcD/9snmTIiMiVL1mbQ0d8WGc93vl8mXt2LZVH4z6NMmPCUhtGlQvrv+N6aw0Hq66GBSiF7tNUfD1W5KkzbuP6pN3Wqhvh7qasnCzvDzd9FHvppKk7FniLn51bPasFq/ea3OWet4cmVWldH5FRN7RK+/MUqYMXpo45O6bh64j//l5W7q0Hjqx9mO5u7ooOiZGfUYv1qZdR5Lx6JGUkiu7bxuyO6WdO3tG3y75Ru1e7aTOr7+pQ3/9qc8+GSVXVze92KRZrP4rf1wurzReql23Xrxjngk8rcWLFujtdwbE2vZahzY6cviQbt++reYvtVK3t2IvDQc8rfp1rqc70TE265k/6FGyu3KpvHr5+XJq3nt6nNt9s/johapF1endr+KdV5PaJZXe21Nfr9iVqONBykqu7I5xcie7U5nIyEhNHP+Z6jds9Ehrmw8e8I42/7xRERERqlmrtkZ88HEyzBJIndZvP6wfNv2hU+eClS9nZr3fq7F+mNJdNTuOU0zM3S++Z37QXrOW/ap9hwKV2zej3fHc3Vz0SoPyGjd3vU374jV7lSmDlzbO7SuLLHJ1ddbMpVv16Zfr4hyH7HZciT4jvXz58lq1apX19v0Qnz17tp599tkEjTF69Gj5+PjY/H3+6ZjETgXJyNx7wahRq47avtpJzxQuoo6vdVG1GrX03bLFsfrfvHlTfXt1U958BfRmtx6xtr/df7DmL/pWn02YqrNnAjXhs7j/e69asVxpvb1Vq07dpD0gIBX6Zc/fqtR6tGp3+lzrth/S12NfU5YMd988Hz55UV2Gz1fvV+vq6o7PdWrDKJ06F6yLQSEyMTGxxqpUMq+K5PPVvOW232g7OVlkjFHnoV9p71+ntfbXQxo07ju1b1zR5qz00FuRqtR6tKq1H6uRU1fok34tVL1cweR9AJBkkiu7x5HdKS4mxqhwkaLq0buvChcpqhYvt1KzFi317dJv4uz/4/LvVL/hi/EWUS5fuqReb72p5+q9oOYvxf5ifNTYz/X1N9/qozGfadvWX2yWagOeZmWK5FKPNrX05gNrrP5bYrO7aH5fLRn/pj6e+ZM27oz7y+t2jSvpemi4fvw57oueSlLHZlW0dtshXbhyI/EHhhSTXNn96Sejk2W+SB5RUVEa2K+PjDEaOuz9Rxqj/6AhWrTkO02YPE1nzpzRZ2P5fwC4b+na37Tql4P66/h5rdh8QC16z1D54v6qUf7uZ9232tSUdxqPeAve/9a0Til5p/GIVQCvXq6gBrz2gvqMXqxn236iV96ZqQbVimlwl7ivDUh2O65En5E+atQoNWjQQIcOHdKdO3c0ceJEHTp0SNu3b9cvv/ySoDGGDBmid955x6YtIib+ZQTw30ufIb2cXVyUN7/tz0z98+bTH/v32bTdunVLfd7qojReaTT288lycY393zJz5izKnDmL/PPmUzofH73Zub1ef7O7Mmf55ywLY4xWLP9WDRo1kaurW/IcGJCKhEXc1skzQTp5Jki7D57SwR+Gq2PzKvrsXogvXrNXi9fsVdaM3roVHiljpN7t6yjgbHCssTo1f1a/Hzmj/YfP2LRfDArR+cs3bNZaPRJwUU5OTsqRLb1OBF6RdPf5efJMkCTpwN/nVChvdg147XnWa0slkiu7bxuyO6VlzpJZefPZZnXefPm0aUPsN/v79+3V6VMBGj328zjHunL5srq90VElS5XW0OEfxNkne3ZfSVK+/AUUEx2tjz8cofYdOsvZ2fkxjwRI3aqWya+sGdPq75/+ee64uDhrzDst1LNdbRVuNEJSwrO7cL7s+umLXvry2+36ZPbaeO+3Y9PKWrRqt6LuRMe5PbdvBtWpVEit+89KgqPEfym5sjvGibPRU4u7RfS3deH8ec38ct4jnY0u/fNZPG++/PLx8VHnDu30Zre3lCVL7F88AE+7U+eCdeVaqPLnyqLNu/9WrQrPqFLJvLqxa4JNv20LBuqb1XvVZfh8m/ZOzapo9dY/dfmq7ZrrI95qpEWrduur7++e2PbX8fNK4+muqe+10Sez19os+0Z2O7ZEF9KrVaum33//XWPGjFGJEiW0bt06lS1bVjt27FCJEiUSNIa7e+yfk5nw2GdhIOW4urqpaNHiCjwVYNMeePqUst+7uIl090z03m+9ITdXN42bMC1BPxOMuXfGze3bUTbt+/bu0ZkzgQ9dtxV4WjlZLHJ3jf2yfT+kOzStrIjbUbHOWvPydNNL9cpq+OQfY+274/eTavFcGXl5ulkvkFIwT1ZFR8fo3KXr9ufilugIQQpJruwOjSC7U1qp0mV1+tQpm7bTp0/J188vVt8fvv9WRYoW0zOFCsfadvnSJXV7o6MKFy2mER+MkpPTw3+0GGNidOfOnbtn0lJIx1Nu4ao92rTrqE3bimk9tHDV7lgXEJXsZ3eRfNm1emZvLVixSyOnroj3PquXK6gCubPqq+Xxr5/6apNndflqqFZvjXsNVjiu5Mru8Kh4OsOh3C+iBwae1qwv/6f06TMkybj3l6q4ffvh6zwDT6McWdMrk4+XLgaFSJL6jV2mkVNXWrf7ZvHRyuk99ergudpz8JTNvnn8MqlmhYJ6+e2Zscb19HCzPv/uu18bs1ikB+roZLeDe6QqSP78+TVrFt+MpHZhYbd0NjDQevv8ubP6+8hhpfPxUXZfP7Xv9JqGDuynMmXLq1yFStqx/Vf9umWzps+eJ+leEb3764qIiNAHH4/VzVs3dfPW3asMZ8iQUc7Oztq29RddDQ5W0eLF5enppZMnjmnyhM9UqnRZ+eWwvZjoj8uXqXiJkspf4Jn/7kEAUoiXp5vy58pive2fI5NKPpND10LCFHz9lga98YJW/XJQF4NuKFP6tOraqob8sqbXd+v/+UVIt1dqaOcfJ3Uz7LbqVi6sUW8307DJP+jGzXCb+3r5hXJycXbSolV7Ys1j8eo9GtKlvma+314fzvhJmdJ7adTbzTXvhx3WtdT7v/a89v0VqJNnr8jdzUX1qxVT20YV1Xt03EtHwDGR3U+mtu076rWObfXl7C9U7/n6+uvPg/p+2VINHW778++bN29qw7q1ervfwFhjXL50SV3f6CBfXz+9/c5AXbt21botc+a7r1OrV62Qi4uLChR8Rq5ubjr815+aOnG8nn++QZy/RAOeRPay+8zFa7p645ZN/6g70boUFKJjpy9b2x6W3UXz+2r1zN7asP2wJn29SdkyeUuSomOMgq7dtBm/U7NntftAgA6duBDnfC0Wizo0rawFK3cpOpovPlMjsvvJFRZ2S4EPfBY/d+6sjhw5LB8fH2XOnEUD3umtw4cOadLULxQTE62goLu/EvXx8bH+evvChfO6ceOGLl44r5joaB05cliSlDt3bqVJ46WtW35RcHCQihcvIc80aXTi+HFNGDdWpcuUVY4cOf/7gwZSgL3svnrjloZ2bajlG3/XxaAQ5cuVWR/3aaYTZ4K0fvvd59OZi7YX8r4Zdve6EyfPXNG5y9dttnVsVlkXg0K0dlvsAvhPW/5U7/a19cfRs9p98JTy58qi4d1f1E9bDtoU2Mlux/dYpxMaY/Tzzz8rPDxcVapUUYYMSfMtKf4bh//6S927dLTenjDuE0lSo8bNNOLD0apdp54GvzdC8+bM1Lixo5Q7T16N+WyiSpcpJ0k6eviQ/jx4dz3GFo1fsBl7+aoN8suRQ+4eHlr+3VKN/2yMoqJuK2u27Kpdt546du5i0/9maKg2bVyvfgOGJOchAw6jbNE8Wje7j/X22P53f4kx/8ed6vXxNyrkn03tG1dSpvReunojTHv/Oq3nXhuvwycvWvcpXzyP3uvWSGnTuOnoqUvq+fGiOIvlnZo9qx82/RGrwC5Jt8Jvq1H3Kfp8UEtt+3qgrt64pW/X77P51t3Lw00T322lHFnTKzwySn+fuqTX3punZev2xRoPjo/sfrIUK15Cn30+SVMmjdfsL6bJL0dO9Rs4WA0aNbbpt27NTzIyqt+gUawxdu3crjOBgToTGKiGz9ey2bb3j7sfIpydnTVv7mwFnj4lYyRfX1+1atNObdt3jDUe8KSyl9321kZ/0MOyu/lzZZQ1o7favlhRbV+saG0/fT7YujyMdPci4M3qllb/T5fFe191KhVSbt+Mmrc89hnxSF3I7ifPX3/+qS6vdbDeHndv3fLGTZur21s9tfnnTZKkV15uarPfrC//pwoVK0mSpk2ZpBU/fG/d1vrlZjZ9PDzc9d2ypfps7GhF3b6tbNl9Vfe5eur8+pvJeWiAQ7GX3b1HLVbxgjnUrnElpff21IUrN7RhxxF9MG2lbkfdSdT9WCwWvdq4sub/uCvWmeeSNGb2GhljNOKtF+WX1UdB125q1ZY/NXKK7S/PyG7HZzEPLsRjx/Xr19WnTx/t27dPlStX1rhx49SwYUNt375dkpQ1a1atW7dOJUuWfKSJ3GBpF8BhZK/SO6WnAOCe8P1THnnf5M5ulnYBHEfWZ8luwFE4cnaztAvgODJW7JnSUwBwT0Kz++ELYN7Tv39/7dixQ61bt9bBgwdVv359RUdHa8eOHdq1a5eKFCmioUOHPvKEAQBA0iK7AQBIXchuAAAcV4LPSM+RI4cWLlyomjVr6ty5c8qVK5c2bdqkWrVqSZJ2796tJk2a6OLFi/YHigdnpAOOgzPSAcfxOGe1JXd2c0Y64Dg4Ix1wHI6c3ZyRDjgOzkgHHEeSn5F+6dIlPfPM3YtA5siRQx4eHsqVK5d1e+7cuXXlypVEThMAACQXshsAgNSF7AYAwHEluJAeExMjZ2dn621nZ2dZLBbr7Qf/DQAAUh7ZDQBA6kJ2AwDguFwS03n27NlKmzatJOnOnTv66quvlDlzZklSaGho0s8OAAA8FrIbAIDUhewGAMAxJXiNdH9//wR9+x0QEPBIE2GNdMBxsEY64DgeZ53V5M5u1kgHHAdrpAOOw5GzmzXSAcfBGumA40hodif4jPRTp0496lwAAEAKILsBAEhdyG4AABxXgtdIBwAAAAAAAADgaUQhHQAAAAAAAAAAOyikAwAAAAAAAABgB4V0AAAAAAAAAADsoJAOAAAAAAAAAIAdiS6kOzs76/Lly7Hag4OD5ezsnCSTAgAASYfsBgAgdSG7AQBwPIkupBtj4myPjIyUm5vbY08IAAAkLbIbAIDUhewGAMDxuCS046RJkyRJFotFs2fPVtq0aa3boqOjtWXLFhUuXDjpZwgAAB4J2Q0AQOpCdgMA4LgSXEgfP368pLvfjM+YMcPm52Rubm7y9/fXjBkzkn6GAADgkZDdAACkLmQ3AACOK8GF9ICAAElS7dq19d133ylDhgzJNikAAPD4yG4AAFIXshsAAMeV4EL6fT///HNyzAMAACQTshsAgNSF7AYAwPEk+mKjL730kj755JNY7WPHjlXLli2TZFIAACDpkN0AAKQuZDcAAI4n0YX0LVu2qGHDhrHaGzRooC1btiTJpAAAQNIhuwEASF3IbgAAHE+iC+k3b96Um5tbrHZXV1eFhIQkyaQAAEDSIbsBAEhdyG4AABxPogvpJUqU0OLFi2O1f/PNNypatGiSTAoAACQdshsAgNSF7AYAwPEk+mKjw4YNU4sWLXTixAnVqVNHkrRx40YtWrRIS5cuTfIJAgCAx0N2AwCQupDdAAA4nkQX0hs3bqzly5dr1KhRWrZsmTw9PVWyZElt2LBBNWvWTI45AgCAx0B2AwCQupDdAAA4HosxxiTVYH/++aeKFy/+SPveCI9JqmkAeEzZq/RO6SkAuCd8/5RkHf9xsjs0guwGHEXWZ8luwFE4cnaHRyXxZAA8sowVe6b0FADck9DsTvQa6f8WGhqqmTNnqmLFiipVqtTjDgcAAJIZ2Q0AQOpCdgMAkPIeuZC+ZcsWdejQQb6+vvrss89Up04d7dy5MynnBgAAkhDZDQBA6kJ2AwDgOBK1RvrFixf11Vdfac6cOQoJCVGrVq0UGRmp5cuXc+VwAAAcENkNAEDqQnYDAOCYEnxGeuPGjVWoUCEdOHBAEyZM0Pnz5zV58uTknBsAAHgMZDcAAKkL2Q0AgONK8Bnpq1evVu/evdW9e3cVLFgwOecEAACSANkNAEDqQnYDAOC4EnxG+q+//qrQ0FCVK1dOlSpV0pQpUxQUFJSccwMAAI+B7AYAIHUhuwEAcFwJLqRXrlxZs2bN0oULF9S1a1d988038vPzU0xMjNavX6/Q0NDknCcAAEgkshsAgNSF7AYAwHFZjDHmUXc+evSo5syZo/nz5+v69euqV6+efvzxx0ca60Z4zKNOA0ASy16ld0pPAcA94funJOl4SZndoRFkN+Aosj5LdgOOwpGzOzwqSacG4DFkrNgzpacA4J6EZneCz0iPS6FChTR27FidPXtWixYtepyhAADAf4DsBgAgdSG7AQBwDI91RnpS4ox0wHFwRjrgOJL6rLakxBnpgOPgjHTAcThydnNGOuA4OCMdcBz/yRnpAAAAAAAAAAA86SikAwAAAAAAAABgB4V0AAAAAAAAAADsoJAOAAAAAAAAAIAdFNIBAAAAAAAAALCDQjoAAAAAAAAAAHZQSAcAAAAAAAAAwA4K6QAAAAAAAAAA2EEhHQAAAAAAAAAAOyikAwAAAAAAAABgB4V0AAAAAAAAAADsoJAOAAAAAAAAAIAdFNIBAAAAAAAAALCDQjoAAAAAAAAAAHZQSAcAAAAAAAAAwA4K6QAAAAAAAAAA2EEhHQAAAAAAAAAAOyikAwAAAAAAAABgB4V0AAAAAAAAAADsoJAOAAAAAAAAAIAdFNIBAAAAAAAAALCDQjoAAAAAAAAAAHZQSAcAAAAAAAAAwA4K6QAAAAAAAAAA2EEhHQAAAAAAAAAAOyikAwAAAAAAAABgB4V0AAAAAAAAAADsoJAOAAAAAAAAAIAdFNIBAAAAAAAAALCDQjoAAAAAAAAAAHZQSAcAAAAAAAAAwA4K6QAAAAAAAAAA2EEhHQAAAAAAAAAAOyikAwAAAAAAAABgB4V0AAAAAAAAAADsoJAOAAAAAAAAAIAdFNIBAAAAAAAAALCDQjoAAAAAAAAAAHa4pPQE7nN3paYPOIx0WVN6BgBSAVcXshtwGGkzpvQMAKQCFktKzwCAVaacKT0DAInEJ2AAAAAAAAAAAOygkA4AAAAAAAAAgB0U0gEAAAAAAAAAsINCOgAAAAAAAAAAdlBIBwAAAAAAAADADgrpAAAAAAAAAADYQSEdAAAAAAAAAAA7KKQDAAAAAAAAAGAHhXQAAAAAAAAAAOygkA4AAAAAAAAAgB0U0gEAAAAAAAAAsINCOgAAAAAAAAAAdlBIBwAAAAAAAADADgrpAAAAAAAAAADYQSEdAAAAAAAAAAA7KKQDAAAAAAAAAGAHhXQAAAAAAAAAAOygkA4AAAAAAAAAgB0U0gEAAAAAAAAAsINCOgAAAAAAAAAAdlBIBwAAAAAAAADADgrpAAAAAAAAAADYQSEdAAAAAAAAAAA7KKQDAAAAAAAAAGAHhXQAAAAAAAAAAOygkA4AAAAAAAAAgB0U0gEAAAAAAAAAsINCOgAAAAAAAAAAdlBIBwAAAAAAAADADgrpAAAAAAAAAADYQSEdAAAAAAAAAAA7KKQDAAAAAAAAAGAHhXQAAAAAAAAAAOygkA4AAAAAAAAAgB0U0gEAAAAAAAAAsINCOgAAAAAAAAAAdlBIBwAAAAAAAADADgrpAAAAAAAAAADYQSEdAAAAAAAAAAA7KKQDAAAAAAAAAGAHhXQAAAAAAAAAAOygkA4AAAAAAAAAgB0U0gEAAAAAAAAAsINCOgAAAAAAAAAAdlBIBwAAAAAAAADADgrpAAAAAAAAAADY8ViFdGOMjDFJNRcAAJDMyG4AAFIXshsAAMfwSIX0//3vfypRooQ8PT3l6empkiVLav78+Uk9NwAAkETIbgAAUheyGwAAx+KS2B0+//xzDRs2TD179lTVqlUlSb/++qu6deumoKAg9e3bN8knCQAAHh3ZDQBA6kJ2AwDgeCwmkb8Ry5s3r95//3116NDBpn3evHkaOXKkAgICHmkiEXceaTcAySBD3Q9SegoA7gn/Zfhjj0F2A0++DLUf/7UCQNII3/r476XJbuDJl6H+mJSeAoB7wjcMTlC/RC/tcuHCBVWpUiVWe5UqVXThwoXEDgcAAJIZ2Q0AQOpCdgMA4HgSXUgvUKCAlixZEqt98eLFKliwYJJMCgAAJB2yGwCA1IXsBgDA8SR6jfT3339fr7zyirZs2WJdq23btm3auHFjnEEPAABSFtkNAEDqQnYDAOB4En1G+ksvvaRdu3Ypc+bMWr58uZYvX67MmTNr9+7dat68eXLMEQAAPAayGwCA1IXsBgDA8ST6YqPJhYueAI6Di40CjiMpLjaaXMhuwHFwsVHAcSTFxUaTC9kNOA4uNgo4joRebDTRS7tIUkxMjI4fP67Lly8rJibGZluNGjUeZUgAAJCMyG4AAFIXshsAAMeS6EL6zp071bZtW50+fVr/PpndYrEoOjo6ySYHAAAeH9kNAEDqQnYDAOB4El1I79atm8qXL69Vq1bJ19dXFoslOeYFAACSCNkNAEDqQnYDAOB4El1IP3bsmJYtW6YCBQokx3wAAEASI7sBAEhdyG4AAByPU2J3qFSpko4fP54ccwEAAMmA7AYAIHUhuwEAcDwJOiP9wIED1n/36tVL/fr108WLF1WiRAm5urra9C1ZsmTSzhAAACQa2Q0AQOpCdgMA4Ngs5t9XLomDk5OTLBZLrIucWAe5t+1xLnoSceeRdgOQDDLU/SClpwDgnvBfhj/SfmQ38HTJUPvRXisAJL3wrY/2XprsBp4uGeqPSekpALgnfMPgBPVL0BnpAQEBjzUZAADw3yK7AQBIXchuAAAcW4IK6Xny5LH+e8uWLapSpYpcXGx3vXPnjrZv327TFwAApAyyGwCA1IXsBgDAsSX6YqO1a9fW1atXY7XfuHFDtWvXTpJJAQCApEN2AwCQupDdAAA4nkQX0u+vyfZvwcHB8vLySpJJAQCApEN2AwCQupDdAAA4ngQt7SJJLVq0kHT3AiedOnWSu7u7dVt0dLQOHDigKlWqJP0MAQDAIyG7AQBIXchuAAAcV4IL6T4+PpLufjPu7e0tT09P6zY3NzdVrlxZXbp0SfoZAgCAR0J2AwCQupDdAAA4rgQX0ufOnStJ8vf3V//+/fk5GQAADo7sBgAgdSG7AQBwXAkupN83YsSI5JgHAABIJmQ3AACpC9kNAIDjSXQhXZKWLVumJUuWKDAwULdv37bZtm/fviSZGAAASDpkNwAAqQvZDQCAY3FK7A6TJk1S586dlS1bNu3fv18VK1ZUpkyZdPLkSTVo0CA55ggAAB4D2Q0AQOpCdgMA4HgSfUb6tGnTNHPmTLVp00ZfffWVBg4cqHz58mn48OG6evVqcswRyeS3vXv01ZdzdPjQn7py5YrGT5qqOnWfs27fsH6dli75Rof/+ks3blzX4mXLVbhIEZsxPhg5XLt2bteVy5eVJk0alSpdRm+/01958+W39hkz6iP9vn+fjh/7W/ny5deS736wGeNUwEl9+P4InTx5QjdDQ5Ula1Y1bPiiur7VU66ursn7IAAppGrJ3OrbporKPuMr38zeajV0sVb8etS6fWinmmpZp5hyZk2n23eitf/oBY2c/bP2HD4nScqd3UdDOtRQrbL+ypYxrS4EhWrR+oP6ZP5WRd2Jsbmvt195Vq81Lqvc2XwUfCNMXyzfq7Ff/2rd3vq54urbpooK5MykG7citG7Xcb07fYOuhoRLktZO6KAaZfxjHcPqHcfUYvCiZHh0kNTI7ifXnFlfaOP6dQoIOCl3Dw+VvpfD/nnzWfssW7JYq39aqcOH/tKtW7e0dccepUuXLtZYW37ZrC+mT9Wxv4/Kzd1d5ctX0ITJ06zb/zx4QBPHj9PhQ39JFouKFy+pvv0GqFDhwv/JsQIprWqpPOrbpprKFvKVb+Z0avXuQq3YeiTOvpP6NVaXZhU0YNJqTVm6w9o+8NUaavDsMypZMLtuR0XLt+HoWPuWK+ynD7s9rzLP+MpI2nv4nIZOW6uDJy5JkgrmyqTJ/ZuosH8W+Xi560JwqBavP6iP5/6sO9F33wN0blxO7V4oraL5skqS9h89rxEzN2jvvfcRcHxk99MlIXkedOWKPh83Vju3b9etsFvy98+rLm9203PPv2Az1sPyHHiaVC2RS31bVVLZgtnufu4e/q1WbD8WZ99JfV5Ql8ZlNGDaBk35bq+1fekHL6lUgazKkt5L10Ij9PO+U3pv9mZdCL4pSRraoZre61At1ni3wm8rc+PPJUlF8mTW8E7VVaZgduXJ7hPrPv6tf+vK+vCNWpry7R4NmL7xcR4CJLFEF9IDAwNVpUoVSZKnp6dCQ0MlSa+++qoqV66sKVOmJO0MkWzCw8NUqFAhNWvxkt7p0zPO7WXKlNULLzTQ+yPei3OMokWLqdGLjZXd11chN25o+tTJ6tbldf20bqOcnZ2t/Zo1f0kHD/6hY0ePxhrDxcVVjZs2U5EixeSdzlt/Hzmi90cOU4wx6v32O0l3wIAD8fJ008Hjl/S/n/Zr8UevxNp+/Gyw+k5crYDz1+Tp7qpeLStpxWftVLztFAXdCFOh3Jnl5GRRz89W6cS5qyqWN6umDnhRXh5uGjJ9vXWccb1fUN3y+TVk2nr9efKyMqbzVIZ0ntbtzxbPpdnvNtPAqeu0atvfypHFW5PeaaRpA15U62FLJUmthy2Rm+s/z+eM6dJo95yu+m7zoWR8hJCUyO4n1949u/VKm3YqVqKEou9Ea/LEz9Wty+v67sdVSpMmjSQpIiJcVapWV5Wq1TVpwrg4x9mwbq3eHzFMvd7uq4qVKiv6TrSOH//buj3s1i291bWLatauo6HDRuhOdLSmT5ms7m++rrUbN/PFN54KXh5uOnj8ov63ap8Wj2oTb78m1YuoYrGcOn8lJNY2N1dnfbf5L+3664w6Niob+z483fTDZx20atsR9Rm3Qi7OThr2eh39OK6DCr40TneiYxR1J0YL1v6u34+e142bESpRILumDmwqJyeLRszcIEmqUdpfSzYc0M4/zyji9h31a1dNK8Z1ULkOU3Q+KDTpHhQkG7L76ZKQPB/67iCFhoRo4pTpypAhg35atUID+r2thUu+VZEiRSU9PM+Bp42Xh6sOnryk/605oMXvt4i3X5Oqz6hiEb84M3LLH4H6dNEOXQy+Kb/M3hrdtbYWDm+m2n2+liRNWLJLs1fst9nnp09b67ejF62303i4KODCdX33yxF90r2u3TmXK5RdrzcqrQMnLifmUPEfSXQhPXv27Lp69ary5Mmj3Llza+fOnSpVqpQCAgJkjEmOOSKZVKteU9Wq14x3e+MmzSRJ586djbfPy63+KQDmyJFTPXu/rZYtmur8uXPKlTu3JGnwu3eL8NemXo2zkJ4zVy7lzJXLetvPL4f27Nmtfb/F/+0ckNqt23Vc63Ydj3f74g1/2tweNHWdOr9YVsXzZ9PmfQFav/uE1u8+Yd1+6sJ1PbN4h7o0LW8tpBfKk1ldmpZXuU4zdOxMsCTp9MXrNuNWKpZTpy9e17Rvd1u3z1nxm/q1qWrtcy00wmaflnWKKywyikJ6KkJ2P7mmz5xjc/uDj8eodvVndfjQXypXvoIkqX2HTpKkPbt3xTnGnTt39MmYj9W3/wC1eKmltT1/gQLWfwcEnNSNG9fVo2dvZff1lSR1e6uHXm7eRBfOn1fuPHmS8rAAh7Ru1zGt2xX3WWz3+WX21udvN1Tjfv/T92Pbx9r+0Zc/S5LaNygd5/6FcmdWJp80+nDOJp29fLcQ//Hcn7V3Xk/lzp5eJ89d1akL13TqwjXrPoGXbqjG+gOqWvKf52HnD7+1Gbf7Jz+oWc2iqlUunxau/SNBx4uURXY/XRKS53/s36+hw0eoRMmSkqQ3u72lr/83T4f/+ktFihRNUJ4DT5t1e05q3Z6Tdvv4ZUqrz3s+p8aDl+j7j1vG2j752z3WfwdeDtFn3+zUkvdfkouzk+5Ex+hWRJRuRURZ+5TIl1VF/bOo94S11rbfjl60FtY/fKNWvHPx8nDV3CFN9Nb41Rrcrmq8/ZByEr1Gep06dfTjjz9Kkjp37qy+ffuqXr16euWVV9S8efMknyBSj7CwMP3w/XfKkTOnsmfP/sjjBJ4+re2/blX5e28YgKedq4uTXm9cTtdDI3TwxMV4+6Xz8rAuxyJJjao8o4Dz19Tw2YI6/E0vHfmmt6YNeFEZvD2sfXb9dVY5s/rohUp332BnzeCl5jWLao2dQkHHRqW1dNOfCnvgzQIcG9n99Lh574zFdD4+Cd7n8KFDunzpkpycnNTqpWaqW7Oa3ur6ho4d++cMNv+8eZU+fXp9/90yRd2+rYiICH3/7TLly5dffjlyJPlxAKmRxWLRnPde0vhF23T41JVHGuPvwCAFXb+ljo3KydXFWR5uLurUqJwOn7oc68vw+/LlyKh6lQpo6++n4h03jburXF2cdS00PN4+cCxk99MtrjwvVaaM1q5ZrRvXrysmJkarf1qlyNuRKl+hoqSE5TkAWxaLNGdwY41fsluHTwc9tH8Gbw+1rltMOw+dtS6n9m+dG5bS32eCte3P+E9Kjc+E3s9rza4T+nnf6UTvi/9Gos9InzlzpmJi7v7P0qNHD2XKlEnbt29XkyZN1LVr1wSNERkZqcjISJs24+wud3f3xE4HDmDxogUaP+4zhYeHyT9vXn0xa65c3dwSPU6Hdq11+NBfun37tl5q+Yre6tUnGWYLpB4Nni2o/w1/SWk8XHUxOFQv9v9awTfi/gCcL0cGdW9RwWZZF3/f9MqdLb1a1CqqN0b9ICcni8b2fF4LP2ipBn3nS5J2/HlGnT/6TvNHviQPNxe5ujhr5bajenv86jjvp3xhPxXPl03dP1mR9AeMZEN2Px1iYmI09pNRKl2mrAoWfCbB+509e0aSNGPqFPUfOFh+OXLof1/N1RudXtWPq9bKJ316eXml1eyv5qtvrx6aOePuOqu58+TR9Jlz5OKS6LeTwBOpX7tquhMdo6nLdj7yGDfDb+uF3nO1ZFQbDel495ejx88Gq0m//yn6Xx/Yf572hko/4ysPd1fN/mGPPpizKd5xP+r+vC4EhWrTXvtn5cFxkN1Pr/jy/NNxEzSwX1/VqFpJLi4u8vDw0PiJU6y/CktIngOw1a915bvZ/b39FRE+eqOWujUtKy9PN+06dE4t3lsaZz93V2e9Uqeoxn2T+PcCLWsVUemC2VTtrXmJ3hf/nUSdkb5z504NGzZMQ4YM0Zo1ayRJrVu31qRJk9SrVy+5JbB4Onr0aPn4+Nj8ffpJ7AvtIHVo+GITLf72e30572vlyeOvAf3ejvWGLSHGfjZe3yz9XmPGjtPWLZs1b+6ch+8EPMF+2X9Kld74QrV7fKl1u0/o65EvKUv6NLH6+WX21o9j2+m7zYc0d+U/a7M5OVnk4e6i10ct17YDgdr6+2l1/2SFapXNq4K5MkmSCufJrM961dfoeVtUpcssNe6/QHmyp9fkfo3inFPHRmV08MQl7T1yPnkOGkmO7H56jProfZ04dkxjPxufqP3MvULNG/cuWFa0WHF98PFoWSwWrVt39/+ZiIgIjRw2VKXLlNX8hYs17+tFKlDgGfXs3lURERH2hgeeCmWe8VWPlyvrzVHfP9Y4Hm4umjG4mXYcDFTNbjNV563ZOhRwWd+NbS8PN9svrV4duUTPvjFDHUcuVYNnn1HfNnH/BLx/u+pqWbe4Xhm6SJG37zzW/PDfILufbvHl+dTJExUaGqKZc77SwsXf6tWOnTWw39s69vfd5VMTkucA/lGmYDb1aF5eb3666qF9xy/Zpcrd5qrRwG8UHROj2YNejLNf02rPyDuNm75edzBRc8mZxVuf9nhOnUetUGRUdKL2xX8rwacQLVu2TK+88oo8PT3l6uqqzz//XJ988on69++f6DsdMmSI3nnH9iKSxplvxVMrb29veXt7K08ef5UsWUrVqlTUpg3r1aBR3C8s8bm/5mr+AgUUHROtD0cOV4dOr9lctBR4moRFROnkuWs6ee6adh86p4MLeqhjozL6bME2ax/fTGm1ZkIH7fzrjHp8ttJm/4vBNxV1J1rHz161th2593O1XNl8dOxMsAa0r6Ydf57R+G92SJL+PHlZYRG3tXFKZ70/+2ddvHrTum8aD1e1rFNMH365ORmPGkmJ7H56jProA235ZbO+nPe1siVyebXMWbJIkvLlz29tc3NzU46cuXTxwgVJ0k+rVuj8+XOav3CxnJzunocxZuxnqlalon7etFENGsb95RvwtKhayl9ZM3jp72X/vE66uDhrTI8X1LNlZRVulbAvuF6pV1K5s6dXzW6zrOtgd3x/mS78NESNqxfW0o3/XEPl/hrqR05dkZOzRVMHNNGEb7YpJuaf9bPfbl1V/dpVU6O+8/TniUtJcahIZmT30y2+PD8TGKhvFn6tb39YqQIFCkqSChUurH2/7dU3ixZo2IgPEpTnAP5RtUQuZU3vpb8XvmVtc3F20piuddSzRQUVbj/d2h4cEq7gkHAdP3dNRwODdfybHqpUxE+7DtueYNapQSmt3nlCl6+HJWouZQpmV7YMXtoxo7PNXKqVyKVuzcrJp8GnNvmOlJPgQvro0aPVpUsXTZ06Vc7Ozho9erRGjRr1SIHu7h7752QRnBzxRDCSZIxu3779eOPEGN25c0cxMTEU0oF7nCwWubv+87Ltl9lbayZ00P6/L+jNMT/q39ed2nHwjFxdnJXXL4MCzt+9KNn9M9ED762zmsbdNdbabtH3AtpisR2vRa2icnd10aL1ift2HSmH7H7yGWM0+uMPtWnjes35ar5y5sz18J3+pWix4nJzc9OpUwEqW668JCkqKkrnz5+Tr6+fpLtnpDtZnGR54IXB4uQkiyzWM+CAp9nCtb9r094TNm0rxnXQwrV/6H8/7UvwOGk8XBVjjM3FJO/fdvp3MD/AyWKRq4uznCwWxdx9R6532lbTwFdrqEm//2nfUX5JllqQ3U+nh+V5RMTd5R2dLLaLCjg5Ocvce++ekDwH8I+FG/7Upn2nbNpWjHlFCzf8qf+tif8zr5PT3Tx2+9cvxfJk91HN0nn08rBliZ7Lz/tPq9wbs23aZg5opKOBwRq3eCdFdAeS4EL60aNHtXjxYmtRs1+/fho+fLguX76srFmzJtsEkXzCbt1SYGCg9fa5s2d15PBh+fj4yNfPTzeuX9eFCxd05cplSdKpUwGSpMyZMytzliw6e+aM1q75Sc9WqaoMGTLq0qWL+nL2TLm7e6hajZrWcQNPn1ZYWJiCgq4oIjJCRw4fliTlz59frm5uWrXyR7m4uKhgwUJyc3PTX38d1MQJ4/R8/QZydXX9Dx8R4L/j5emq/DkyWm/7+6ZXyQLZdO3eN92DXq2uVduO6mLwTWXySaOuzcvLL3M6fbf5kKS7RfS1Ezso8OINDZm23mbJl0tXb0mSNv12UvuOntcXg5powJS1crJYNOHtBtqw54T1LPVV2//WtAEvqkvTclq/+4R8M3nr017Pa8+hc7oQfFMP6tSojFb8esTmgqZwbGT3k2/Uh+9r9U8rNWHyNHml8VLQlbsXOEzr7S0Pj7sXFg66ckVBQUE6cy/zjx/7W2nSeMnX11c+6dMrbdq0atmqtaZPnazs2X3l5+enr+4tr/b8C/UlSc8+W0XjPxurUR++rzbtXlWMidGXs2fKxcVZFSpVSoEjB/57Xp5u/8ruDCpZILuuhYTrzOUbsfIx6k60Ll29qWNngq1tubL6KEM6T+XKll7Ozk4qWeDuGacnzl3VrfDb2rjnhEZ1f14T3nlR07/dKSeLRf3bV9ed6Bj9sv/ue/HW9Uoq6k60/jx5SZG3o1WusJ8+7FpPyzb9af1yvF/bahr2eh11+mCZTl+8rmwZ00q6uwb7rfDHO+EFyYvsfjo9LM/98+ZT7tx59OH7w/VO/0FKnz69Nm3aoJ07tmnytC/u9k1AngNPGy8PV+XPkcF62983vUrmz6proRE6czlEV0NslyiMuhOjS1dv6di9z8sVCvuqXCFfbf/zrK6HRiivXwaN6FRdJ85d065D52z27Vi/pC5evam1e2Jfj8TVxUlF8mSWJLm5OMkvs7dK5s+qm+G3dfL8dd0Mv61Dp2wvdnorIkpXQ8JjtSNlWYz59zmMcXNyctLFixdtwtvb21t//PGH8uXL99gT4Zvx/96e3bv0RucOsdqbNG2uD0eN0Q/ff6fh7w2Jtb3bWz3VvUcvXb58Se8Pf0+HDv2lkBshypQ5k8qVK6+u3XvIP+8//0+83ulV7d2zO9Y4P63bqBw5cmrN6p/01ZezdfpUgIyRfP389GLjJmrfoRMXwkkhGep+kNJTeOJVL51H6yZ2jNU+f/Xv6vX5Ks0b1kIViuRQJp80uhoSrr1HzuuT+Vv12721ydvXL6VZQ5rGObZnzX/++/lmSqvP+zRQ3Qr5dCs8Sut2H9fgqet0LfSfNwzdW1TQG03Kyd83g27cjNDmfQF674uNOh8Uau1TMFcmHfi6hxr1+5oLlf3Hwn8Z/sj7kt1PvlLFCsXZ/sFHo9W0eQtJ0vSpkzVj2hS7faKiojRpwudaueIHRUZEqETJUhow+F3rz8clacf2bZoxbYpOHD8mi8VJhYsUUa8+fVWyVOmkPzAkWobaj/5agYSpXtpf6ya/Fqt9/ur9ca6NfmRJX01ZulNTlu6wts18t7lebVAmVt/ne32prb+fkiTVKZ9fQzvXUtG8WRVjjP44dlEjZ27Q7kNnJUkv1ymuvm2rqWCuTLJICrx0Q4vW/aHJS3ZY10A/sqSv8vhmiHU/H335sz6e+/MjHD0SI3zro7+XJrufTgnJ89OnT2ni5+O0f/9vCgsLU+5cudWh82tq3KSZtX9C8hyOI0P9MSk9hSde9VK5tW5c21jt89cejHNt9CNfd9eU7/Zoynd3Lz5aLG8WffbWcyqRP6u8PFx1Mfim1u09qU++3q7zD5x4ZrFIfy94SwvW/6mRc7fEGjd3Nh8dXdA9VvuWPwL1Qr+Fcc597bi2OnD8kgZM35jg48WjC98wOEH9ElVI/+ijj5Q2bVpr26BBgzRgwABlzpzZ2ta7d+9ETvUuAh1wHBTSAcfxuIV0sht4OlBIBxzH4xbSyW7g6UAhHXAcSV5I9/f3t1kXM87BLBadPPloZyoS6IDjoJAOOI7HKaST3cDTg0I64Dgep5BOdgNPDwrpgONIaCE9wWuknzp16lHnAgAAUgDZDQBA6kJ2AwDguJwe3gUAAAAAAAAAgKcXhXQAAAAAAAAAAOygkA4AAAAAAAAAgB0U0gEAAAAAAAAAsINCOgAAAAAAAAAAdiS6kO7s7KzLly/Hag8ODpazs3OSTAoAACQdshsAgNSF7AYAwPEkupBujImzPTIyUm5ubo89IQAAkLTIbgAAUheyGwAAx+OS0I6TJk2SJFksFs2ePVtp06a1bouOjtaWLVtUuHDhpJ8hAAB4JGQ3AACpC9kNAIDjSnAhffz48ZLufjM+Y8YMm5+Tubm5yd/fXzNmzEj6GQIAgEdCdgMAkLqQ3QAAOK4EF9IDAgIkSbVr19Z3332nDBkyJNukAADA4yO7AQBIXchuAAAcV4IL6ff9/PPPyTEPAACQTMhuAABSF7IbAADHk+iLjb700kv65JNPYrWPHTtWLVu2TJJJAQCApEN2AwCQupDdAAA4nkQX0rds2aKGDRvGam/QoIG2bNmSJJMCAABJh+wGACB1IbsBAHA8iS6k37x5U25ubrHaXV1dFRISkiSTAgAASYfsBgAgdSG7AQBwPIkupJcoUUKLFy+O1f7NN9+oaNGiSTIpAACQdMhuAABSF7IbAADHk+iLjQ4bNkwtWrTQiRMnVKdOHUnSxo0btWjRIi1dujTJJwgAAB4P2Q0AQOpCdgMA4HgSXUhv3Lixli9frlGjRmnZsmXy9PRUyZIltWHDBtWsWTM55ggAAB4D2Q0AQOpCdgMA4HgsxhiTVIP9+eefKl68+CPtG3EnqWYB4HFlqPtBSk8BwD3hvwxP1vHJbuDJkKF28r5WAEi48K3J+16a7AaeDBnqj0npKQC4J3zD4AT1S/Qa6f8WGhqqmTNnqmLFiipVqtTjDgcAAJIZ2Q0AQOpCdgMAkPIeuZC+ZcsWdejQQb6+vvrss89Up04d7dy5MynnBgAAkhDZDQBA6kJ2AwDgOBK1RvrFixf11Vdfac6cOQoJCVGrVq0UGRmp5cuXc+VwAAAcENkNAEDqQnYDAOCYEnxGeuPGjVWoUCEdOHBAEyZM0Pnz5zV58uTknBsAAHgMZDcAAKkL2Q0AgONK8Bnpq1evVu/evdW9e3cVLFgwOecEAACSANkNAEDqQnYDAOC4EnxG+q+//qrQ0FCVK1dOlSpV0pQpUxQUFJSccwMAAI+B7AYAIHUhuwEAcFwJLqRXrlxZs2bN0oULF9S1a1d988038vPzU0xMjNavX6/Q0NDknCcAAEgkshsAgNSF7AYAwHFZjDHmUXc+evSo5syZo/nz5+v69euqV6+efvzxx0caK+LOo84CQFLLUPeDlJ4CgHvCfxmepOOR3cCTKUPtpH2tAPDowrcm7Xtpsht4MmWoPyalpwDgnvANgxPUL8FnpMelUKFCGjt2rM6ePatFixY9zlAAAOA/QHYDAJC6kN0AADiGxzojPSnxzTjgODgjHXAcSX1GelIiuwHHwRnpgONI6jPSkxLZDTgOzkgHHMd/ckY6AAAAAAAAAABPOgrpAAAAAAAAAADYQSEdAAAAAAAAAAA7KKQDAAAAAAAAAGAHhXQAAAAAAAAAAOygkA4AAAAAAAAAgB0U0gEAAAAAAAAAsINCOgAAAAAAAAAAdlBIBwAAAAAAAADADgrpAAAAAAAAAADYQSEdAAAAAAAAAAA7KKQDAAAAAAAAAGAHhXQAAAAAAAAAAOygkA4AAAAAAAAAgB0U0gEAAAAAAAAAsINCOgAAAAAAAAAAdlBIBwAAAAAAAADADgrpAAAAAAAAAADYQSEdAAAAAAAAAAA7KKQDAAAAAAAAAGAHhXQAAAAAAAAAAOygkA4AAAAAAAAAgB0U0gEAAAAAAAAAsINCOgAAAAAAAAAAdlBIBwAAAAAAAADADgrpAAAAAAAAAADYQSEdAAAAAAAAAAA7KKQDAAAAAAAAAGAHhXQAAAAAAAAAAOygkA4AAAAAAAAAgB0U0gEAAAAAAAAAsINCOgAAAAAAAAAAdlBIBwAAAAAAAADADgrpAAAAAAAAAADYQSEdAAAAAAAAAAA7KKQDAAAAAAAAAGAHhXQAAAAAAAAAAOygkA4AAAAAAAAAgB0U0gEAAAAAAAAAsINCOgAAAAAAAAAAdlBIBwAAAAAAAADADgrpAAAAAAAAAADYYTHGmJSeBJ4MkZGRGj16tIYMGSJ3d/eUng7wVOP5CCAheK0AHAfPRwAJwWsF4Dh4Pj59KKQjyYSEhMjHx0c3btxQunTpUno6wFON5yOAhOC1AnAcPB8BJASvFYDj4Pn49GFpFwAAAAAAAAAA7KCQDgAAAAAAAACAHRTSAQAAAAAAAACwg0I6koy7u7tGjBjBBRYAB8DzEUBC8FoBOA6ejwASgtcKwHHwfHz6cLFRAAAAAAAAAADs4Ix0AAAAAAAAAADsoJAOAAAAAAAAAIAdFNIBAAAAAAAAALCDQjoAAAAAAAAAAHZQSEecOnXqpGbNmqX0NCRJI0eOVOnSpVN6GoBDSannKM9HwHGR3YBjI7sB/BvZDTg2shv/RiE9FenUqZMsFossFovc3NxUoEABffDBB7pz585/PpfNmzdb52KxWOTp6alixYpp5syZSX5f/fv318aNG5N8XCCpOfJzNFu2bHrppZd08uTJBI9hsVi0fPny5Jsk8BRw5NcFshtw7Oco2Q2kDEd+XSC7Acd+jpLdTz6XlJ4AEqd+/fqaO3euIiMj9dNPP6lHjx5ydXXVkCFDYvW9ffu23NzcknU+R48eVbp06RQeHq4VK1aoe/fuyp8/v+rWrZtk95E2bVqlTZs2ycYDkpMjPke9vb117Ngxvfnmm2rcuLEOHDggZ2fnZL1fAP9wxNcFshv4hyM+R8luIGU54usC2Q38wxGfo2T304Ez0lMZd3d3Zc+eXXny5FH37t313HPP6ccff5T0z09OPv74Y/n5+alQoUKSpDNnzqhVq1ZKnz69MmbMqKZNm+rUqVPWMaOjo/XOO+8offr0ypQpkwYOHChjTILmkzVrVmXPnl158+ZV7969lTdvXu3bt8+6PSYmRqNHj1bevHnl6empUqVKadmyZdbt97+927hxo8qXL680adKoSpUqOnr0qLXPv3/ScufOHfXu3ds630GDBqljx442P7epVauWevfurYEDBypjxozKnj27Ro4cmYhHGng0jvgc9fX1VY0aNTR8+HAdOnRIx48f1549e1SvXj1lzpxZPj4+qlmzps1z19/fX5LUvHlzWSwW6+375s+fL39/f/n4+Kh169YKDQ21bouMjFTv3r2VNWtWeXh4qFq1atqzZ88jPJrAk8ERXxfIbuAfjvgcJbuBlOWIrwtkN/APR3yOkt1PBwrpqZynp6du375tvb1x40YdPXpU69ev18qVKxUVFaUXXnhB3t7e2rp1q7Zt26a0adOqfv361v3GjRunr776Sl9++aV+/fVXXb16Vd9//32i5mGM0Zo1axQYGKhKlSpZ20ePHq3//e9/mjFjhv766y/17dtX7du31y+//GKz/9ChQzVu3Djt3btXLi4ueu211+K9r08++UQLFizQ3LlztW3bNoWEhMT5M5h58+bJy8tLu3bt0tixY/XBBx9o/fr1iTou4HE5ynP0/lyku9/Ih4aGqmPHjvr111+1c+dOFSxYUA0bNrQG8/0Anjt3ri5cuGATyCdOnNDy5cu1cuVKrVy5Ur/88ovGjBlj3T5w4EB9++23mjdvnvbt26cCBQrohRde0NWrVxP/AAJPIEd5XSC7gbg5ynP0/lwkshtIaY7yukB2A3FzlOfo/blIZPcTyyDV6Nixo2natKkxxpiYmBizfv164+7ubvr372/dni1bNhMZGWndZ/78+aZQoUImJibG2hYZGWk8PT3N2rVrjTHG+Pr6mrFjx1q3R0VFmZw5c1rvKy4///yzkWS8vLyMl5eXcXFxMU5OTuajjz6y9omIiDBp0qQx27dvt9n39ddfN23atLEZZ8OGDdbtq1atMpJMeHi4McaYESNGmFKlSlm3Z8uWzXz66afW23fu3DG5c+e2mW/NmjVNtWrVbO63QoUKZtCgQfEeE/C4HPE5eu3aNWOMMefPnzdVqlQxOXLksLn/+6Kjo423t7dZsWKFtU2S+f777236jRgxwqRJk8aEhIRY2wYMGGAqVapkjDHm5s2bxtXV1SxYsMC6/fbt28bPz8/mGICnhSO+LpDdwD8c8TlKdgMpyxFfF8hu4B+O+Bwlu58erJGeyqxcuVJp06ZVVFSUYmJi1LZtW5ufTpUoUcJm7ac//vhDx48fl7e3t804EREROnHihG7cuKELFy7YfJvt4uKi8uXLJ+gnLFu3bpW3t7ciIyO1e/du9ezZUxkzZlT37t11/PhxhYWFqV69ejb73L59W2XKlLFpK1mypPXfvr6+kqTLly8rd+7cNv1u3LihS5cuqWLFitY2Z2dnlStXTjExMfGOeX/cy5cvP/SYgMfhaM/RnDlzyhijsLAwlSpVSt9++63c3Nx06dIlvffee9q8ebMuX76s6OhohYWFKTAw8KFj+vv728z3wefWiRMnFBUVpapVq1q3u7q6qmLFijp8+PBDxwaeRI72ukB2A7Yc7TlKdgMpz9FeF8huwJajPUfJ7qcHhfRUpnbt2po+fbrc3Nzk5+cnFxfb/4ReXl42t2/evKly5cppwYIFscbKkiXLY88nb968Sp8+vSSpWLFi2rVrlz7++GN1795dN2/elCStWrVKOXLksNnP3d3d5rarq6v13xaLRZJiBXRiPTjm/XEfd0zgYRztObp161alS5dOWbNmtQnhjh07Kjg4WBMnTlSePHnk7u6uZ5991ubncPHhuQUkjqO9LpDdgC1He46S3UDKc7TXBbIbsOVoz1Gy++lBIT2V8fLyUoECBRLcv2zZslq8eLGyZs2qdOnSxdnH19dXu3btUo0aNSTdvajIb7/9prJlyyZ6fs7OzgoPD5ckFS1aVO7u7goMDFTNmjUTPVZcfHx8lC1bNu3Zs8c63+joaO3bt8/mwihASnG05+iDb7oftG3bNk2bNk0NGzaUdPfCK0FBQTZ9XF1dFR0dneBjkaT8+fPLzc1N27ZtU548eSRJUVFR2rNnj95+++1EjQU8KRztdeHfyG487RztOUp2AynP0V4X/o3sxtPO0Z6jZPfTg4uNPuHatWunzJkzq2nTptq6dasCAgK0efNm9e7dW2fPnpUk9enTR2PGjNHy5ct15MgRvfXWW7p+/XqCxr98+bIuXryo06dPa+nSpZo/f76aNm0qSfL29lb//v3Vt29fzZs3TydOnNC+ffs0efJkzZs375GPqVevXho9erR++OEHHT16VH369NG1a9es36gDqUlyP0fjU7BgQc2fP1+HDx/Wrl271K5dO+tFUe7z9/fXxo0bdfHiRV27di1B43p5eal79+4aMGCA1qxZo0OHDqlLly4KCwvT66+//lhzBp4WZDfg2MhuAP9GdgOOjexGUuGM9CdcmjRptGXLFg0aNEgtWrRQaGiocuTIobp161q/hevXr58uXLjw/3bumCXVMAwD8H30D9TgJM7iKjU1NDq3OAQqONgeQdAqaEtbWwT9DcN/4V66tUpTi+cMZ+rU+QrpYHKua/54ed7h44ablye9Xi+lUin9fj9HR0dZLpcfnl+v15P83h1Vq9VycnLyai/VcDhMpVLJeDzOw8NDdnZ20mw2c3Fxsfadzs/P8/T0lG63m3K5nMFgkFarlXK5vPaZsCn/+h/9m9vb2wwGgzSbzdRqtYxGo5ydnb365urqKqenp7m5uUm1Ws18Pv/U2ZeXl1mtVul0Onl+fs7e3l4mk0l2d3fXnhf+J7IbvjfZDfxJdsP3Jrv5Kj9+fmZrPnxjq9UqjUYj7XY7w+Fw0+MAAB+Q3QCwXWQ3gBfpbKHFYpH7+/scHh7m5eUl19fXeXx8zPHx8aZHAwDeIbsBYLvIboC37Ehn65RKpdzd3WV/fz8HBweZzWaZTqdpNBqbHg0AeIfsBoDtIrsB3rLaBQAAAAAACniRDgAAAAAABRTpAAAAAABQQJEOAAAAAAAFFOkAAAAAAFBAkQ4AAAAAAAUU6QAAAAAAUECRDgAAAAAABRTpAAAAAABQ4BcmwNP5cKhLQAAAAABJRU5ErkJggg==", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "#same test on train dataset\n", + "# ============================================================\n", + "# CELL: BENCHMARK ACCURACY & CONFUSION MATRIX (robust)\n", + "# ============================================================\n", + "\n", + "import pandas as pd\n", + "import numpy as np\n", + "import matplotlib.pyplot as plt\n", + "import seaborn as sns\n", + "from sklearn.metrics import (\n", + " confusion_matrix,\n", + " accuracy_score,\n", + " precision_recall_curve,\n", + " f1_score,\n", + " average_precision_score,\n", + ")\n", + "\n", + "# -------------- config --------------\n", + "TEST_FILE = \"train_modified.parquet\"\n", + "\n", + "benchmarks = [\n", + " \"CADD_raw_rankscore\", \"REVEL_rankscore\", \"AlphaMissense_rankscore\",\n", + " \"PrimateAI_rankscore\", \"MPC_rankscore\", \"ClinPred_rankscore\",\n", + " \"BayesDel_addAF_rankscore\", \"EVE_rankscore\", \"ESM1b_rankscore\"\n", + "]\n", + "\n", + "# -------------- load --------------\n", + "df = pd.read_parquet(TEST_FILE)\n", + "print(\"Loaded test rows:\", len(df))\n", + "\n", + "# -------------- clean & map labels --------------\n", + "# find label column\n", + "label_col = None\n", + "for c in (\"clean_label\", \"label\", \"labels\"):\n", + " if c in df.columns:\n", + " label_col = c\n", + " break\n", + "if label_col is None:\n", + " raise KeyError(\"Could not find 'label', 'clean_label', or 'labels' column in test file.\")\n", + "\n", + "def map_label_val(x):\n", + " if pd.isna(x):\n", + " return np.nan\n", + " s = str(x).strip().lower()\n", + " if s in (\"pathogenic\", \"path\", \"p\", \"1\", \"true\", \"t\", \"yes\", \"y\"):\n", + " return 1\n", + " if s in (\"benign\", \"benign \", \"b\", \"0\", \"false\", \"f\", \"no\", \"n\"):\n", + " return 0\n", + " # try numeric\n", + " try:\n", + " nv = float(s)\n", + " if nv == 1.0: return 1\n", + " if nv == 0.0: return 0\n", + " except:\n", + " pass\n", + " return np.nan\n", + "\n", + "raw_labels = df[label_col]\n", + "y_true = raw_labels.map(map_label_val)\n", + "\n", + "# drop unknowns\n", + "if y_true.isnull().any():\n", + " drop_count = int(y_true.isnull().sum())\n", + " print(f\"⚠️ Dropping {drop_count} rows with unmappable labels.\")\n", + " valid_mask = ~y_true.isnull()\n", + " df = df.loc[valid_mask].reset_index(drop=True)\n", + " y_true = y_true.loc[valid_mask].reset_index(drop=True)\n", + "else:\n", + " y_true = y_true.reset_index(drop=True)\n", + "\n", + "y_true = y_true.astype(int)\n", + "print(\"Final eval rows:\", len(y_true))\n", + "if len(np.unique(y_true)) < 2:\n", + " raise RuntimeError(\"Need at least two label classes in test set for evaluation.\")\n", + "\n", + "# -------------- evaluate each benchmark --------------\n", + "results = []\n", + "confusion_matrices = {}\n", + "\n", + "for tool in benchmarks:\n", + " if tool not in df.columns:\n", + " print(f\"⚠️ Skipping {tool} (not present in parquet).\")\n", + " continue\n", + "\n", + " scores = df[tool].astype(float).copy()\n", + " # fill NaN with median (stable)\n", + " if scores.isna().any():\n", + " median_val = float(scores.median())\n", + " scores = scores.fillna(median_val)\n", + "\n", + " # handle constant scores (no thresholds)\n", + " unique_vals = np.unique(scores)\n", + " if len(unique_vals) == 1:\n", + " # fallback threshold = median (or 0.5) and compute metrics\n", + " best_thresh = unique_vals[0]\n", + " y_pred = (scores >= best_thresh).astype(int)\n", + " acc = accuracy_score(y_true, y_pred)\n", + " f1_best = f1_score(y_true, y_pred, zero_division=0)\n", + " results.append({\n", + " \"Model\": tool.replace(\"_rankscore\", \"\").replace(\"_raw\", \"\"),\n", + " \"Accuracy\": acc,\n", + " \"Best_Threshold\": float(best_thresh),\n", + " \"F1_Score\": float(f1_best),\n", + " \"AP\": float(average_precision_score(y_true, scores))\n", + " })\n", + " confusion_matrices[tool.replace(\"_rankscore\", \"\").replace(\"_raw\", \"\")] = confusion_matrix(y_true, y_pred)\n", + " print(f\" {tool}: constant scores — using threshold {best_thresh:.4f}\")\n", + " continue\n", + "\n", + " # compute PR curve then choose threshold maximizing F1 (properly index thresholds)\n", + " precision, recall, thresholds = precision_recall_curve(y_true, scores)\n", + " # precision,recall length = len(thresholds)+1\n", + " f1_all = (2 * precision * recall) / (precision + recall + 1e-12)\n", + " # only consider indices that have associated thresholds => 1..end\n", + " if len(thresholds) > 0:\n", + " idx_rel = np.argmax(f1_all[1:]) # best among entries that map to thresholds\n", + " best_idx = idx_rel + 1 # absolute index into precision/recall arrays\n", + " best_thresh = float(thresholds[best_idx - 1])\n", + " else:\n", + " # very rare: fallback\n", + " best_thresh = float(np.median(scores))\n", + "\n", + " y_pred = (scores >= best_thresh).astype(int)\n", + " acc = accuracy_score(y_true, y_pred)\n", + " f1_best = f1_score(y_true, y_pred, zero_division=0)\n", + " ap = average_precision_score(y_true, scores)\n", + "\n", + " results.append({\n", + " \"Model\": tool.replace(\"_rankscore\", \"\").replace(\"_raw\", \"\"),\n", + " \"Accuracy\": acc,\n", + " \"Best_Threshold\": best_thresh,\n", + " \"F1_Score\": float(f1_best),\n", + " \"AP\": float(ap)\n", + " })\n", + " confusion_matrices[tool.replace(\"_rankscore\", \"\").replace(\"_raw\", \"\")] = confusion_matrix(y_true, y_pred)\n", + "\n", + " print(f\" {tool}: Acc={acc:.4f} F1={f1_best:.4f} AP={ap:.4f} thresh={best_thresh:.4f}\")\n", + "\n", + "# -------------- report --------------\n", + "if not results:\n", + " print(\"❌ No benchmarks found.\")\n", + "else:\n", + " res_df = pd.DataFrame(results).sort_values(\"Accuracy\", ascending=False).reset_index(drop=True)\n", + " print(\"\\n\" + \"=\" * 60)\n", + " print(\"🏆 BENCHMARK LEADERBOARD (thresholds optimized for F1)\")\n", + " print(\"=\" * 60)\n", + " display_cols = [\"Model\", \"Accuracy\", \"Best_Threshold\", \"F1_Score\", \"AP\"]\n", + " print(res_df[display_cols].to_string(index=False, formatters={\n", + " 'Accuracy': '{:.2%}'.format,\n", + " 'Best_Threshold': '{:.4f}'.format,\n", + " 'F1_Score': '{:.4f}'.format,\n", + " 'AP': '{:.4f}'.format\n", + " }))\n", + "\n", + " # plot confusion matrices in grid\n", + " n_tools = len(res_df)\n", + " cols = 3\n", + " rows = (n_tools + cols - 1) // cols\n", + " fig, axes = plt.subplots(rows, cols, figsize=(5 * cols, 4 * rows))\n", + " axes = axes.flatten()\n", + "\n", + " for i, row in res_df.iterrows():\n", + " name = row[\"Model\"]\n", + " cm = confusion_matrices.get(name)\n", + " ax = axes[i]\n", + " sns.heatmap(cm, annot=True, fmt=\"d\", cmap=\"Blues\", cbar=False, ax=ax,\n", + " xticklabels=[\"Pred Benign\", \"Pred Patho\"], yticklabels=[\"Act Benign\", \"Act Patho\"])\n", + " ax.set_title(f\"{name}\\nAcc: {row['Accuracy']:.1%} F1: {row['F1_Score']:.3f}\")\n", + "\n", + " # hide unused axes\n", + " for j in range(i + 1, len(axes)):\n", + " axes[j].axis(\"off\")\n", + "\n", + " plt.tight_layout()\n", + " plt.show()\n" + ] + }, + { + "cell_type": "code", + "execution_count": 22, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Loaded test rows: 1200890\n", + "Final eval rows: 1200890\n", + " CADD_raw_rankscore: Acc=0.2024 F1=0.2184 AP=0.7528 thresh=0.6457\n", + " REVEL_rankscore: Acc=0.9008 F1=0.4104 AP=0.3705 thresh=0.6705\n", + " AlphaMissense_rankscore: Acc=0.8988 F1=0.4034 AP=0.3643 thresh=0.5102\n", + " PrimateAI_rankscore: Acc=0.8826 F1=0.3624 AP=0.2936 thresh=0.3776\n", + " MPC_rankscore: Acc=0.8811 F1=0.3335 AP=0.2826 thresh=0.3977\n", + " ClinPred_rankscore: Acc=0.9094 F1=0.4621 AP=0.4051 thresh=0.3591\n", + " BayesDel_addAF_rankscore: Acc=0.9670 F1=0.8504 AP=0.7997 thresh=0.6334\n", + " EVE_rankscore: Acc=0.8960 F1=0.2945 AP=0.2636 thresh=0.5810\n", + " ESM1b_rankscore: Acc=0.8939 F1=0.3758 AP=0.3339 thresh=0.5267\n", + "\n", + "============================================================\n", + "🏆 BENCHMARK LEADERBOARD (thresholds optimized for F1)\n", + "============================================================\n", + " Model Accuracy Best_Threshold F1_Score AP\n", + "BayesDel_addAF 96.70% 0.6334 0.8504 0.7997\n", + " ClinPred 90.94% 0.3591 0.4621 0.4051\n", + " REVEL 90.08% 0.6705 0.4104 0.3705\n", + " AlphaMissense 89.88% 0.5102 0.4034 0.3643\n", + " EVE 89.60% 0.5810 0.2945 0.2636\n", + " ESM1b 89.39% 0.5267 0.3758 0.3339\n", + " PrimateAI 88.26% 0.3776 0.3624 0.2936\n", + " MPC 88.11% 0.3977 0.3335 0.2826\n", + " CADD 20.24% 0.6457 0.2184 0.7528\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "#same test on train dataset\n", + "# ============================================================\n", + "# CELL: BENCHMARK ACCURACY & CONFUSION MATRIX (robust)\n", + "# ============================================================\n", + "\n", + "import pandas as pd\n", + "import numpy as np\n", + "import matplotlib.pyplot as plt\n", + "import seaborn as sns\n", + "from sklearn.metrics import (\n", + " confusion_matrix,\n", + " accuracy_score,\n", + " precision_recall_curve,\n", + " f1_score,\n", + " average_precision_score,\n", + ")\n", + "\n", + "# -------------- config --------------\n", + "TEST_FILE = \"train_enriched.parquet\"\n", + "\n", + "benchmarks = [\n", + " \"CADD_raw_rankscore\", \"REVEL_rankscore\", \"AlphaMissense_rankscore\",\n", + " \"PrimateAI_rankscore\", \"MPC_rankscore\", \"ClinPred_rankscore\",\n", + " \"BayesDel_addAF_rankscore\", \"EVE_rankscore\", \"ESM1b_rankscore\"\n", + "]\n", + "\n", + "# -------------- load --------------\n", + "df = pd.read_parquet(TEST_FILE)\n", + "print(\"Loaded test rows:\", len(df))\n", + "\n", + "# -------------- clean & map labels --------------\n", + "# find label column\n", + "label_col = None\n", + "for c in (\"clean_label\", \"label\", \"labels\"):\n", + " if c in df.columns:\n", + " label_col = c\n", + " break\n", + "if label_col is None:\n", + " raise KeyError(\"Could not find 'label', 'clean_label', or 'labels' column in test file.\")\n", + "\n", + "def map_label_val(x):\n", + " if pd.isna(x):\n", + " return np.nan\n", + " s = str(x).strip().lower()\n", + " if s in (\"pathogenic\", \"path\", \"p\", \"1\", \"true\", \"t\", \"yes\", \"y\"):\n", + " return 1\n", + " if s in (\"benign\", \"benign \", \"b\", \"0\", \"false\", \"f\", \"no\", \"n\"):\n", + " return 0\n", + " # try numeric\n", + " try:\n", + " nv = float(s)\n", + " if nv == 1.0: return 1\n", + " if nv == 0.0: return 0\n", + " except:\n", + " pass\n", + " return np.nan\n", + "\n", + "raw_labels = df[label_col]\n", + "y_true = raw_labels.map(map_label_val)\n", + "\n", + "# drop unknowns\n", + "if y_true.isnull().any():\n", + " drop_count = int(y_true.isnull().sum())\n", + " print(f\"⚠️ Dropping {drop_count} rows with unmappable labels.\")\n", + " valid_mask = ~y_true.isnull()\n", + " df = df.loc[valid_mask].reset_index(drop=True)\n", + " y_true = y_true.loc[valid_mask].reset_index(drop=True)\n", + "else:\n", + " y_true = y_true.reset_index(drop=True)\n", + "\n", + "y_true = y_true.astype(int)\n", + "print(\"Final eval rows:\", len(y_true))\n", + "if len(np.unique(y_true)) < 2:\n", + " raise RuntimeError(\"Need at least two label classes in test set for evaluation.\")\n", + "\n", + "# -------------- evaluate each benchmark --------------\n", + "results = []\n", + "confusion_matrices = {}\n", + "\n", + "for tool in benchmarks:\n", + " if tool not in df.columns:\n", + " print(f\"⚠️ Skipping {tool} (not present in parquet).\")\n", + " continue\n", + "\n", + " scores = df[tool].astype(float).copy()\n", + " # fill NaN with median (stable)\n", + " if scores.isna().any():\n", + " median_val = float(scores.median())\n", + " scores = scores.fillna(median_val)\n", + "\n", + " # handle constant scores (no thresholds)\n", + " unique_vals = np.unique(scores)\n", + " if len(unique_vals) == 1:\n", + " # fallback threshold = median (or 0.5) and compute metrics\n", + " best_thresh = unique_vals[0]\n", + " y_pred = (scores >= best_thresh).astype(int)\n", + " acc = accuracy_score(y_true, y_pred)\n", + " f1_best = f1_score(y_true, y_pred, zero_division=0)\n", + " results.append({\n", + " \"Model\": tool.replace(\"_rankscore\", \"\").replace(\"_raw\", \"\"),\n", + " \"Accuracy\": acc,\n", + " \"Best_Threshold\": float(best_thresh),\n", + " \"F1_Score\": float(f1_best),\n", + " \"AP\": float(average_precision_score(y_true, scores))\n", + " })\n", + " confusion_matrices[tool.replace(\"_rankscore\", \"\").replace(\"_raw\", \"\")] = confusion_matrix(y_true, y_pred)\n", + " print(f\" {tool}: constant scores — using threshold {best_thresh:.4f}\")\n", + " continue\n", + "\n", + " # compute PR curve then choose threshold maximizing F1 (properly index thresholds)\n", + " precision, recall, thresholds = precision_recall_curve(y_true, scores)\n", + " # precision,recall length = len(thresholds)+1\n", + " f1_all = (2 * precision * recall) / (precision + recall + 1e-12)\n", + " # only consider indices that have associated thresholds => 1..end\n", + " if len(thresholds) > 0:\n", + " idx_rel = np.argmax(f1_all[1:]) # best among entries that map to thresholds\n", + " best_idx = idx_rel + 1 # absolute index into precision/recall arrays\n", + " best_thresh = float(thresholds[best_idx - 1])\n", + " else:\n", + " # very rare: fallback\n", + " best_thresh = float(np.median(scores))\n", + "\n", + " y_pred = (scores >= best_thresh).astype(int)\n", + " acc = accuracy_score(y_true, y_pred)\n", + " f1_best = f1_score(y_true, y_pred, zero_division=0)\n", + " ap = average_precision_score(y_true, scores)\n", + "\n", + " results.append({\n", + " \"Model\": tool.replace(\"_rankscore\", \"\").replace(\"_raw\", \"\"),\n", + " \"Accuracy\": acc,\n", + " \"Best_Threshold\": best_thresh,\n", + " \"F1_Score\": float(f1_best),\n", + " \"AP\": float(ap)\n", + " })\n", + " confusion_matrices[tool.replace(\"_rankscore\", \"\").replace(\"_raw\", \"\")] = confusion_matrix(y_true, y_pred)\n", + "\n", + " print(f\" {tool}: Acc={acc:.4f} F1={f1_best:.4f} AP={ap:.4f} thresh={best_thresh:.4f}\")\n", + "\n", + "# -------------- report --------------\n", + "if not results:\n", + " print(\"❌ No benchmarks found.\")\n", + "else:\n", + " res_df = pd.DataFrame(results).sort_values(\"Accuracy\", ascending=False).reset_index(drop=True)\n", + " print(\"\\n\" + \"=\" * 60)\n", + " print(\"🏆 BENCHMARK LEADERBOARD (thresholds optimized for F1)\")\n", + " print(\"=\" * 60)\n", + " display_cols = [\"Model\", \"Accuracy\", \"Best_Threshold\", \"F1_Score\", \"AP\"]\n", + " print(res_df[display_cols].to_string(index=False, formatters={\n", + " 'Accuracy': '{:.2%}'.format,\n", + " 'Best_Threshold': '{:.4f}'.format,\n", + " 'F1_Score': '{:.4f}'.format,\n", + " 'AP': '{:.4f}'.format\n", + " }))\n", + "\n", + " # plot confusion matrices in grid\n", + " n_tools = len(res_df)\n", + " cols = 3\n", + " rows = (n_tools + cols - 1) // cols\n", + " fig, axes = plt.subplots(rows, cols, figsize=(5 * cols, 4 * rows))\n", + " axes = axes.flatten()\n", + "\n", + " for i, row in res_df.iterrows():\n", + " name = row[\"Model\"]\n", + " cm = confusion_matrices.get(name)\n", + " ax = axes[i]\n", + " sns.heatmap(cm, annot=True, fmt=\"d\", cmap=\"Blues\", cbar=False, ax=ax,\n", + " xticklabels=[\"Pred Benign\", \"Pred Patho\"], yticklabels=[\"Act Benign\", \"Act Patho\"])\n", + " ax.set_title(f\"{name}\\nAcc: {row['Accuracy']:.1%} F1: {row['F1_Score']:.3f}\")\n", + "\n", + " # hide unused axes\n", + " for j in range(i + 1, len(axes)):\n", + " axes[j].axis(\"off\")\n", + "\n", + " plt.tight_layout()\n", + " plt.show()\n" + ] + }, + { + "cell_type": "code", + "execution_count": 20, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "🧠 INITIALIZING SHAP ANALYSIS...\n", + "📉 Extracting embeddings for 150 samples...\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Encoding Data: 1%| | 5/440 [00:03<04:30, 1.61it/s]\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "✅ Data Prepared. Background shape: torch.Size([100, 1289])\n", + "⚡ Calculating SHAP values (GradientExplainer)...\n", + "✅ SHAP Calculation Complete.\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/tmp/ipykernel_10337/4129238603.py:111: FutureWarning: The NumPy global RNG was seeded by calling `np.random.seed`. In a future version this function will no longer use the global RNG. Pass `rng` explicitly to opt-in to the new behaviour and silence this warning.\n", + " shap.summary_plot(\n", + "/home/zeus/miniconda3/envs/cloudspace/lib/python3.10/site-packages/shap/plots/_beeswarm.py:723: FutureWarning: The NumPy global RNG was seeded by calling `np.random.seed`. In a future version this function will no longer use the global RNG. Pass `rng` explicitly to opt-in to the new behaviour and silence this warning.\n", + " summary_legacy(\n", + "/home/zeus/miniconda3/envs/cloudspace/lib/python3.10/site-packages/shap/plots/_beeswarm.py:743: FutureWarning: The NumPy global RNG was seeded by calling `np.random.seed`. In a future version this function will no longer use the global RNG. Pass `rng` explicitly to opt-in to the new behaviour and silence this warning.\n", + " summary_legacy(\n" + ] + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAA0UAAAIQCAYAAABOs8VFAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjgsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvwVt1zgAAAAlwSFlzAAAPYQAAD2EBqD+naQAANl1JREFUeJzt3Xl0VeW9+P9PGJIwBbBAGBpBcMA6QAsFUZHaRul16KVqweECUudSq6ZeBQeQOuC11ku/inKhTu2tF+rU+lW+KEX5eVuxVIZWrVAVEGplcgAKlQDZvz9cnBoTlBMJFJ/Xa62zltnZe5/nnDwH884+e5+CLMuyAAAASFSDPT0AAACAPUkUAQAASRNFAABA0kQRAACQNFEEAAAkTRQBAABJE0UAAEDSRBEAAJA0UQQAACRNFAF8Cj/72c+ie/fu0bhx42jVqtUu229BQUFcd911ua/vu+++KCgoiGXLluW9r9mzZ0dBQUHMnj17l43vwz7N2CCi9jn0la98Jb7yla/ssvu47rrroqCgYJftD/hsEUVArbb/kvLCCy/s6aF8atOnT68WGLvKokWL4uyzz45u3brFlClTYvLkyZ+4zcKFC+Pf/u3foqysLIqKimKfffaJ8vLyuPfee2Pbtm27fIz/TLb/UlrbbdKkSfVyn/X1s/+s6dKlS7WfR7t27aJ///7x6KOP7umh5WXTpk1x3XXX1dsfAIDPrkZ7egAA9W369OkxceLEXf7L8ezZs6Oqqip+/OMfx/777/+J6//kJz+JCy+8MEpLS2Po0KFxwAEHxIYNG2LWrFlxzjnnxFtvvRVXXXVVrdsOHTo0Tj/99CgqKsp7nMccc0z8/e9/j8LCwry3rQ933XVXNG/evNqyvn371st91dfP/rOoZ8+e8f3vfz8iIv7617/Gf/3Xf8Upp5wSd911V1x44YW7fTxPPfVU3tts2rQpxo0bFxFR4yjTNddcE6NGjdoVQwM+g0QRQB2tXr06ImKn3jb3/PPPx4UXXhj9+vWL6dOnR4sWLXLfu/TSS+OFF16Il156aYfbN2zYMBo2bFincTZo0CCKi4vrtG19OO2006JNmzZ7ehifysaNG6NZs2Z7ehi7VKdOneLf/u3fcl8PGzYs9t9///jP//zPHUbR1q1bo6qqql6Ce1fvs1GjRtGokV97gNp5+xyw084+++xo3rx5LF++PE466aRo3rx5dOrUKSZOnBgRES+++GJ89atfjWbNmkXnzp3jgQceqLb99rfkPfvss3HBBRfE5z73uSgpKYlhw4bFu+++W23dX/3qV3HiiSdGx44do6ioKLp16xbXX399rW8x+93vfhcnnHBCtG7dOpo1axaHH354/PjHP86Nefv4Pvz2oE9y5513xiGHHBJFRUXRsWPHGDlyZLz33nu573fp0iXGjh0bERFt27atcQ7QR40bNy4KCgri5z//ebUg2q53795x9tln73D72s656NKlS5x00knxm9/8Jvr06RPFxcXRtWvX+OlPf1pt2x2dU/Rxz1tExB//+Mc4++yzo2vXrlFcXBzt27ePb3/72/H222/vcJy7wn//939Hr169okmTJrHPPvvE6aefHitWrKi2zv/+7//Gt771rdh3332jqKgoysrK4rLLLou///3vuXU+7me/o+dk2bJlUVBQEPfdd1+1/TRv3jxef/31OOGEE6JFixZx1llnRUREVVVVTJgwIQ455JAoLi6O0tLSuOCCC2rM5xdeeCEGDhwYbdq0iSZNmsR+++0X3/72t3fq+fikuRjxwVGRQw89NP70pz/FscceG02bNo1OnTrFLbfcslP3UZv27dvHwQcfHEuXLq323Nx6660xYcKE6NatWxQVFcWf/vSniPjg7aSnnXZa7LPPPlFcXBy9e/eOxx57rMZ+X3755fjqV78aTZo0ic9//vNxww03RFVVVY31ajun6P3334/rrrsuDjzwwCguLo4OHTrEKaecEq+//nosW7Ys2rZtGxH/eL19+HVZ2zlFW7dujeuvvz73WLp06RJXXXVVbN68udp6O/taA/Ze/mQC5GXbtm3xL//yL3HMMcfELbfcEj//+c/ju9/9bjRr1iyuvvrqOOuss+KUU06JSZMmxbBhw6Jfv36x3377VdvHd7/73WjVqlVcd911sXjx4rjrrrvijTfeyP2iGvFBBDRv3jwqKiqiefPm8fTTT8eYMWNi/fr18cMf/jC3r5kzZ8ZJJ50UHTp0iEsuuSTat28fr7zySjz++ONxySWXxAUXXBB//etfY+bMmfGzn/1spx7jddddF+PGjYvy8vK46KKLcmP8/e9/H7/97W+jcePGMWHChPjpT38ajz76aO7tYIcffnit+9u0aVPMmjUrjjnmmNh3333r+MzX7rXXXovTTjstzjnnnBg+fHjcc889cfbZZ0evXr3ikEMO2eF2n/S8bV9nyZIlMWLEiGjfvn28/PLLMXny5Hj55Zfj+eefr/NJ6++88061rxs2bBitW7eOiIgbb7wxrr322hg8eHCce+65sWbNmrj99tvjmGOOiQULFuSOyj344IOxadOmuOiii+Jzn/tczJ07N26//fb4y1/+Eg8++GBERJ1+9juydevWGDhwYBx99NFx6623RtOmTXP3cd9998WIESPie9/7XixdujTuuOOOWLBgQW6urF69Oo4//vho27ZtjBo1Klq1ahXLli2LRx555BPvd2fm4nbvvvtufP3rX49TTjklBg8eHA899FBceeWVcdhhh8W//Mu/5P2Yt2zZEitWrIjPfe5z1Zbfe++98f7778f555+fOy/u5ZdfjqOOOio6deoUo0aNimbNmsUvfvGLGDRoUDz88MPxzW9+MyIiVq5cGccee2xs3bo1t97kyZOjSZMmnziebdu2xUknnRSzZs2K008/PS655JLYsGFDzJw5M1566aUoLy+Pu+66Ky666KL45je/GaecckpExA5flxER5557btx///1x2mmnxfe///343e9+F+PHj49XXnmlxvlUdX2tAXuJDKAW9957bxYR2e9///vcsuHDh2cRkd100025Ze+++27WpEmTrKCgIJs6dWpu+aJFi7KIyMaOHVtjn7169coqKytzy2+55ZYsIrJf/epXuWWbNm2qMaYLLrgga9q0afb+++9nWZZlW7duzfbbb7+sc+fO2bvvvltt3aqqqtx/jxw5MtvZf+5Wr16dFRYWZscff3y2bdu23PI77rgji4jsnnvuyS0bO3ZsFhHZmjVrPnaff/jDH7KIyC655JKdGkOWZTt87pYuXZpb1rlz5ywismeffbba+IuKirLvf//7uWXPPPNMFhHZM888k2XZzj9vtf0M/ud//qfGfdY2ttpsf74+euvcuXOWZVm2bNmyrGHDhtmNN95YbbsXX3wxa9SoUbXltY1t/PjxWUFBQfbGG2/klu3oZ//R52S7pUuXZhGR3Xvvvbll2+f9qFGjqq37v//7v1lEZD//+c+rLZ8xY0a15Y8++miN19LOyGcuDhgwIIuI7Kc//Wlu2ebNm7P27dtnp5566ifeV+fOnbPjjz8+W7NmTbZmzZrsD3/4Q3b66adnEZFdfPHFWZb947kpKSnJVq9eXW37r33ta9lhhx2We21m2Qdz6cgjj8wOOOCA3LJLL700i4jsd7/7XbXH2bJlyxpzaMCAAdmAAQNyX99zzz1ZRGS33XZbjfFvn7dr1qyp8drZbvv8227hwoVZRGTnnntutfUuv/zyLCKyp59+utrzszOvNWDv5e1zQN7OPffc3H+3atUqDjrooGjWrFkMHjw4t/yggw6KVq1axZIlS2psf/7551f7C/dFF10UjRo1iunTp+eWffgvxxs2bIi1a9dG//79Y9OmTbFo0aKIiFiwYEEsXbo0Lr300hrn9dT1KMavf/3rqKysjEsvvTQaNPjHP5HnnXdelJSUxBNPPJH3PtevXx8RUevb5j6tL3zhC9G/f//c123bto2DDjqo1ud9u5193j78M3j//fdj7dq1ccQRR0RExPz58+s85ocffjhmzpyZu/385z+PiIhHHnkkqqqqYvDgwbF27drcrX379nHAAQfEM888U+vYNm7cGGvXro0jjzwysiyLBQsW1HlsH+eiiy6q9vWDDz4YLVu2jOOOO67aeHv16hXNmzfPjXf7c/z444/Hli1bdvr+8p2LzZs3r3ZOUGFhYfTp0+dj58KHPfXUU9G2bdto27Zt9OjRIx588MEYOnRo/Md//Ee19U499dTc29QiPjjy9/TTT8fgwYNzr9W1a9fG22+/HQMHDoxXX3013nzzzYj44MIXRxxxRPTp0ye3fdu2bXNvR/w4Dz/8cLRp0yYuvvjiGt+ry+t9+783FRUV1ZZvv9jER5/furzWgL2Ht88BeSkuLq72C1FERMuWLePzn/98jV9MWrZsWePcioiIAw44oNrXzZs3jw4dOlQ7X+bll1+Oa665Jp5++ulcVGy3bt26iIh4/fXXIyLi0EMPrfPj+ag33ngjIj6Iug8rLCyMrl275r6fj5KSkoj4IO52tdrejte6detan/ftdvZ5e+edd2LcuHExderU3EUlttv+M6iLY445ptYLLbz66quRZVmN+bHdh0N6+fLlMWbMmHjsscdqPNZPM7YdadSoUXz+85+vMd5169ZFu3btat1m+3M2YMCAOPXUU2PcuHHxn//5n/GVr3wlBg0aFGeeeebHXk0w37lY22uwdevW8cc//nGnHmPfvn3jhhtuiIKCgmjatGkcfPDBtV5E5KNvh33ttdciy7K49tpr49prr61136tXr45OnTrFG2+8UeuVBj/6GGvz+uuvx0EHHbTLLpbwxhtvRIMGDWpcObJ9+/bRqlWrGs9vXV5rwN5DFAF52dEV0Ha0PMuyvO/jvffeiwEDBkRJSUn84Ac/iG7dukVxcXHMnz8/rrzyylpPyv5ntv/++0ejRo3ixRdf3OX73pXP+0cNHjw4nnvuufj3f//36NmzZzRv3jyqqqri61//er38DKqqqqKgoCD+3//7f7U+ru2X8d62bVscd9xx8c4778SVV14Z3bt3j2bNmsWbb74ZZ5999k6NbUdHFnb0WVFFRUXVjtZsH2+7du1yR7o+avsfDwoKCuKhhx6K559/Pv7v//2/8eSTT8a3v/3t+NGPfhTPP/98jcuT19WnnQtt2rSJ8vLyT1zvo+f/bH++L7/88hg4cGCt2+zMJev3lJ09ylSfrzVgzxNFwG736quvxrHHHpv7+m9/+1u89dZbccIJJ0TEB1cGe/vtt+ORRx6JY445Jrfe9qtgbdetW7eIiNxJ1juSz1trOnfuHBERixcvjq5du+aWV1ZWxtKlS3fql8aPatq0aXz1q1+Np59+OlasWBFlZWV572NX2pnn7d13341Zs2bFuHHjYsyYMbnlr776ar2OK8uy2G+//eLAAw/c4Xovvvhi/PnPf477778/hg0blls+c+bMGuvu6Ge//cIOH72KWz5HArt16xa//vWv46ijjtqpCwUcccQRccQRR8SNN94YDzzwQJx11lkxderUam9H/bD6mIv1YfvYGjdu/Ilj6ty5c61zaPHixZ94P926dYvf/e53sWXLlmpHDT8s39d6VVVVvPrqq3HwwQfnlq9atSree++93PMPpME5RcBuN3ny5GrnVtx1112xdevW3BWytv9F9sN/ga2srIw777yz2n6+9KUvxX777RcTJkyo8cvth7fd/nkyH12nNuXl5VFYWBj/5//8n2r7uPvuu2PdunVx4okn7tyD/IixY8dGlmUxdOjQ+Nvf/lbj+/PmzYv777+/TvvO1848b7X9DCIiJkyYUG/jOuWUU6Jhw4Yxbty4GvebZVnuUuC1jS3LsmqXE99uRz/7zp07R8OGDePZZ5+ttvyjc+zjDB48OLZt2xbXX399je9t3bo1d5/vvvtujcfTs2fPiIgal37+sPqai7tau3bt4itf+Ur813/9V7z11ls1vr9mzZrcf59wwgnx/PPPx9y5c6t9f0dH2z7s1FNPjbVr18Ydd9xR43vbn5/tVwXcmdf69j/CfHRO33bbbRER/zTPL7B7OFIE7HaVlZXxta99LQYPHhyLFy+OO++8M44++uj4xje+ERERRx55ZLRu3TqGDx8e3/ve96KgoCB+9rOf1fjFskGDBnHXXXfFySefHD179owRI0ZEhw4dYtGiRfHyyy/Hk08+GRERvXr1ioiI733vezFw4MBo2LBhnH766bWOrW3btjF69OgYN25cfP3rX49vfOMbuTF++ctfrnYiez6OPPLImDhxYnznO9+J7t27x9ChQ+OAAw6IDRs2xOzZs+Oxxx6LG264oU77ztfOPG8lJSW5y65v2bIlOnXqFE899VSNo3W7Urdu3eKGG26I0aNHx7Jly2LQoEHRokWLWLp0aTz66KNx/vnnx+WXXx7du3ePbt26xeWXXx5vvvlmlJSUxMMPP1zruR07+tm3bNkyvvWtb8Xtt98eBQUF0a1bt3j88cdrnDv1cQYMGBAXXHBBjB8/PhYuXBjHH398NG7cOF599dV48MEH48c//nGcdtppcf/998edd94Z3/zmN6Nbt26xYcOGmDJlSpSUlOR+Ma9Nfc3F+jBx4sQ4+uij47DDDovzzjsvunbtGqtWrYo5c+bEX/7yl/jDH/4QERFXXHFF/OxnP4uvf/3rcckll+Quyd25c+dPPPdp2LBh8dOf/jQqKipi7ty50b9//9i4cWP8+te/ju985zvxr//6r9GkSZP4whe+ENOmTYsDDzww9tlnnzj00ENrPX+uR48eMXz48Jg8eXLuLbtz586N+++/PwYNGlTtaDaQgN15qTtg77GjS3I3a9asxroDBgzIDjnkkBrLO3funJ144ok19vn//X//X3b++ednrVu3zpo3b56dddZZ2dtvv11t29/+9rfZEUcckTVp0iTr2LFjdsUVV2RPPvlkrZdR/s1vfpMdd9xxWYsWLbJmzZplhx9+eHb77bfnvr9169bs4osvztq2bZsVFBTs1OW577jjjqx79+5Z48aNs9LS0uyiiy6qcfnqnb0k94fNmzcvO/PMM7OOHTtmjRs3zlq3bp197Wtfy+6///5ql12Onbwk94ef3+0+einjHV1++pOet7/85S/ZN7/5zaxVq1ZZy5Yts29961vZX//6150aW2129vl6+OGHs6OPPjpr1qxZ1qxZs6x79+7ZyJEjs8WLF+fW+dOf/pSVl5dnzZs3z9q0aZOdd955uUuff/hy2h/3s1+zZk126qmnZk2bNs1at26dXXDBBdlLL71U6yW5a5v3202ePDnr1atX1qRJk6xFixbZYYcdll1xxRXZX//61yzLsmz+/PnZGWecke27775ZUVFR1q5du+ykk07KXnjhhY99Hrbbmbm4o9fg8OHDc5c8/zg7mksftv2S3D/84Q9r/f7rr7+eDRs2LGvfvn3WuHHjrFOnTtlJJ52UPfTQQ9XW++Mf/5gNGDAgKy4uzjp16pRdf/312d133/2Jl+TOsg8uxX711Vdn++23X9a4ceOsffv22WmnnZa9/vrruXWee+65rFevXllhYWG1ufrRS3JnWZZt2bIlGzduXG5/ZWVl2ejRo6tdWvzjnp/axgjsnQqyzBmCwO6x/UMuf//730fv3r339HAAACLCOUUAAEDiRBEAAJA0UQQAACQt7yh69tln4+STT46OHTtGQUFB/PKXv/zEbWbPnh1f+tKXoqioKPbff/+477776jBUYG939tlnR5ZlzicCAP6p5B1FGzdujB49esTEiRN3av2lS5fGiSeeGMcee2wsXLgwLr300jj33HNzl8oFAADYkz7V1ecKCgri0UcfjUGDBu1wnSuvvDKeeOKJeOmll3LLTj/99HjvvfdixowZdb1rAACAXaLeP7x1zpw5UV5eXm3ZwIED49JLL93hNps3b672Kd9VVVXxzjvvxOc+97koKCior6ECAAD/5LIsiw0bNkTHjh2jQYNdc4mEeo+ilStXRmlpabVlpaWlsX79+vj73/8eTZo0qbHN+PHjY9y4cfU9NAAAYC+1YsWK+PznP79L9lXvUVQXo0ePjoqKitzX69ati3333TdWrFgRJSUle3BkAADAnrR+/fooKyuLFi1a7LJ91nsUtW/fPlatWlVt2apVq6KkpKTWo0QREUVFRVFUVFRjeUlJiSgCAAB26Wk19f45Rf369YtZs2ZVWzZz5szo169ffd81AADAJ8o7iv72t7/FwoULY+HChRHxwSW3Fy5cGMuXL4+ID976NmzYsNz6F154YSxZsiSuuOKKWLRoUdx5553xi1/8Ii677LJd8wgAAAA+hbyj6IUXXogvfvGL8cUvfjEiIioqKuKLX/xijBkzJiIi3nrrrVwgRUTst99+8cQTT8TMmTOjR48e8aMf/Sh+8pOfxMCBA3fRQwAAAKi7T/U5RbvL+vXro2XLlrFu3TrnFAEAQMLqow3q/ZwiAACAf2aiCAAASJooAgAAkiaKAACApIkiAAAgaaIIAABImigCAACSJooAAICkiSIAACBpoggAAEiaKAIAAJImigAAgKSJIgAAIGmiCAAASJooAgAAkiaKAACApIkiAAAgaaIIAABImigCAACSJooAAICkiSIAACBpoggAAEiaKAIAAJImigAAgKSJIgAAIGmiCAAASJooAgAAkiaKAACApIkiAAAgaaIIAABImigCAACSJooAAICkiSIAACBpoggAAEiaKAIAAJImigAAgKSJIgAAIGmiCAAASJooAgAAkiaKAACApIkiAAAgaaIIAABImigCAACSJooAAICkiSIAACBpoggAAEiaKAIAAJImigAAgKSJIgAAIGmiCAAASJooAgAAkiaKAACApIkiAAAgaaIIAABImigCAACSJooAAICkiSIAACBpoggAAEiaKAIAAJImigAAgKSJIgAAIGmiCAAASJooAgAAkiaKAACApIkiAAAgaaIIAABImigCAACSJooAAICkiSIAACBpoggAAEiaKAIAAJImigAAgKSJIgAAIGmiCAAASJooAgAAkiaKAACApIkiAAAgaaIIAABImigCAACSJooAAICkiSIAACBpdYqiiRMnRpcuXaK4uDj69u0bc+fO/dj1J0yYEAcddFA0adIkysrK4rLLLov333+/TgMGAADYlfKOomnTpkVFRUWMHTs25s+fHz169IiBAwfG6tWra13/gQceiFGjRsXYsWPjlVdeibvvvjumTZsWV1111acePAAAwKeVdxTddtttcd5558WIESPiC1/4QkyaNCmaNm0a99xzT63rP/fcc3HUUUfFmWeeGV26dInjjz8+zjjjjE88ugQAALA75BVFlZWVMW/evCgvL//HDho0iPLy8pgzZ06t2xx55JExb968XAQtWbIkpk+fHieccMKnGDYAAMCu0SifldeuXRvbtm2L0tLSastLS0tj0aJFtW5z5plnxtq1a+Poo4+OLMti69atceGFF37s2+c2b94cmzdvzn29fv36fIYJAACw0+r96nOzZ8+Om266Ke68886YP39+PPLII/HEE0/E9ddfv8Ntxo8fHy1btszdysrK6nuYAABAogqyLMt2duXKyspo2rRpPPTQQzFo0KDc8uHDh8d7770Xv/rVr2ps079//zjiiCPihz/8YW7Zf//3f8f5558ff/vb36JBg5pdVtuRorKysli3bl2UlJTs7HABAIDPmPXr10fLli13aRvkdaSosLAwevXqFbNmzcotq6qqilmzZkW/fv1q3WbTpk01wqdhw4YREbGjHisqKoqSkpJqNwAAgPqQ1zlFEREVFRUxfPjw6N27d/Tp0ycmTJgQGzdujBEjRkRExLBhw6JTp04xfvz4iIg4+eST47bbbosvfvGL0bdv33jttdfi2muvjZNPPjkXRwAAAHtK3lE0ZMiQWLNmTYwZMyZWrlwZPXv2jBkzZuQuvrB8+fJqR4auueaaKCgoiGuuuSbefPPNaNu2bZx88slx44037rpHAQAAUEd5nVO0p9TH+wYBAIC9zx4/pwgAAOCzRhQBAABJE0UAAEDSRBEAAJA0UQQAACRNFAEAAEkTRQAAQNJEEQAAkDRRBAAAJE0UAQAASRNFAABA0kQRAACQNFEEAAAkTRQBAABJE0UAAEDSRBEAAJA0UQQAACRNFAEAAEkTRQAAQNJEEQAAkDRRBAAAJE0UAQAASRNFAABA0kQRAACQNFEEAAAkTRQBAABJE0UAAEDSRBEAAJA0UQQAACRNFAEAAEkTRQAAQNJEEQAAkDRRBAAAJE0UAQAASRNFAABA0kQRAACQNFEEAAAkTRQBAABJE0UAAEDSRBEAAJA0UQQAACRNFAEAAEkTRQAAQNJEEQAAkDRRBAAAJE0UAQAASRNFAABA0kQRAACQNFEEAAAkTRQBAABJE0UAAEDSRBEAAJA0UQQAACRNFAEAAEkTRQAAQNJEEQAAkDRRBAAAJE0UAQAASRNFAABA0kQRAACQNFEEAAAkTRQBAABJE0UAAEDSRBEAAJA0UQQAACRNFAEAAEkTRQAAQNJEEQAAkDRRBAAAJE0UAQAASRNFAABA0kQRAACQNFEEAAAkTRQBAABJE0UAAEDSRBEAAJA0UQQAACRNFAEAAEkTRQAAQNJEEQAAkDRRBAAAJE0UAQAASatTFE2cODG6dOkSxcXF0bdv35g7d+7Hrv/ee+/FyJEjo0OHDlFUVBQHHnhgTJ8+vU4DBgAA2JUa5bvBtGnToqKiIiZNmhR9+/aNCRMmxMCBA2Px4sXRrl27GutXVlbGcccdF+3atYuHHnooOnXqFG+88Ua0atVqV4wfAADgUynIsizLZ4O+ffvGl7/85bjjjjsiIqKqqirKysri4osvjlGjRtVYf9KkSfHDH/4wFi1aFI0bN67TINevXx8tW7aMdevWRUlJSZ32AQAA7P3qow3yevtcZWVlzJs3L8rLy/+xgwYNory8PObMmVPrNo899lj069cvRo4cGaWlpXHooYfGTTfdFNu2bdvh/WzevDnWr19f7QYAAFAf8oqitWvXxrZt26K0tLTa8tLS0li5cmWt2yxZsiQeeuih2LZtW0yfPj2uvfba+NGPfhQ33HDDDu9n/Pjx0bJly9ytrKwsn2ECAADstHq/+lxVVVW0a9cuJk+eHL169YohQ4bE1VdfHZMmTdrhNqNHj45169blbitWrKjvYQIAAInK60ILbdq0iYYNG8aqVauqLV+1alW0b9++1m06dOgQjRs3joYNG+aWHXzwwbFy5cqorKyMwsLCGtsUFRVFUVFRPkMDAACok7yOFBUWFkavXr1i1qxZuWVVVVUxa9as6NevX63bHHXUUfHaa69FVVVVbtmf//zn6NChQ61BBAAAsDvl/fa5ioqKmDJlStx///3xyiuvxEUXXRQbN26MESNGRETEsGHDYvTo0bn1L7roonjnnXfikksuiT//+c/xxBNPxE033RQjR47cdY8CAACgjvL+nKIhQ4bEmjVrYsyYMbFy5cro2bNnzJgxI3fxheXLl0eDBv9orbKysnjyySfjsssui8MPPzw6deoUl1xySVx55ZW77lEAAADUUd6fU7Qn+JwiAAAg4p/gc4oAAAA+a0QRAACQNFEEAAAkTRQBAABJE0UAAEDSRBEAAJA0UQQAACRNFAEAAEkTRQAAQNJEEQAAkDRRBAAAJE0UAQAASRNFAABA0kQRAACQNFEEAAAkTRQBAABJE0UAAEDSRBEAAJA0UQQAACRNFAEAAEkTRQAAQNJEEQAAkDRRBAAAJE0UAQAASRNFAABA0kQRAACQNFEEAAAkTRQBAABJE0UAAEDSRBEAAJA0UQQAACRNFAEAAEkTRQAAQNJEEQAAkDRRBAAAJE0UAQAASRNFAABA0kQRAACQNFEEAAAkTRQBAABJE0UAAEDSRBEAAJA0UQQAACRNFAEAAEkTRQAAQNJEEQAAkDRRBAAAJE0UAQAASRNFAABA0kQRAACQNFEEAAAkTRQBAABJE0UAAEDSRBEAAJA0UQQAACRNFAEAAEkTRQAAQNJEEQAAkDRRBAAAJE0UAQAASRNFAABA0kQRAACQNFEEAAAkTRQBAABJE0UAAEDSRBEAAJA0UQQAACRNFAEAAEkTRQAAQNJEEQAAkDRRBAAAJE0UAQAASRNFAABA0kQRAACQNFEEAAAkTRQBAABJE0UAAEDSRBEAAJA0UQQAACRNFAEAAEkTRQAAQNLqFEUTJ06MLl26RHFxcfTt2zfmzp27U9tNnTo1CgoKYtCgQXW5WwAAgF0u7yiaNm1aVFRUxNixY2P+/PnRo0ePGDhwYKxevfpjt1u2bFlcfvnl0b9//zoPFgAAYFfLO4puu+22OO+882LEiBHxhS98ISZNmhRNmzaNe+65Z4fbbNu2Lc4666wYN25cdO3a9VMNGAAAYFfKK4oqKytj3rx5UV5e/o8dNGgQ5eXlMWfOnB1u94Mf/CDatWsX55xzzk7dz+bNm2P9+vXVbgAAAPUhryhau3ZtbNu2LUpLS6stLy0tjZUrV9a6zW9+85u4++67Y8qUKTt9P+PHj4+WLVvmbmVlZfkMEwAAYKfV69XnNmzYEEOHDo0pU6ZEmzZtdnq70aNHx7p163K3FStW1OMoAQCAlDXKZ+U2bdpEw4YNY9WqVdWWr1q1Ktq3b19j/ddffz2WLVsWJ598cm5ZVVXVB3fcqFEsXrw4unXrVmO7oqKiKCoqymdoAAAAdZLXkaLCwsLo1atXzJo1K7esqqoqZs2aFf369auxfvfu3ePFF1+MhQsX5m7f+MY34thjj42FCxd6WxwAALDH5XWkKCKioqIihg8fHr17944+ffrEhAkTYuPGjTFixIiIiBg2bFh06tQpxo8fH8XFxXHooYdW275Vq1YRETWWAwAA7Al5R9GQIUNizZo1MWbMmFi5cmX07NkzZsyYkbv4wvLly6NBg3o9VQkAAGCXKciyLNvTg/gk69evj5YtW8a6deuipKRkTw8HAADYQ+qjDRzSAQAAkiaKAACApIkiAAAgaaIIAABImigCAACSJooAAICkiSIAACBpoggAAEiaKAIAAJImigAAgKSJIgAAIGmiCAAASJooAgAAkiaKAACApIkiAAAgaaIIAABImigCAACSJooAAICkiSIAACBpoggAAEiaKAIAAJImigAAgKSJIgAAIGmiCAAASJooAgAAkiaKAACApIkiAAAgaaIIAABImigCAACSJooAAICkiSIAACBpoggAAEiaKAIAAJImigAAgKSJIgAAIGmiCAAASJooAgAAkiaKAACApIkiAAAgaaIIAABImigCAACSJooAAICkiSIAACBpoggAAEiaKAIAAJImigAAgKSJIgAAIGmiCAAASJooAgAAkiaKAACApIkiAAAgaaIIAABImigCAACSJooAAICkiSIAACBpoggAAEiaKAIAAJImigAAgKSJIgAAIGmiCAAASJooAgAAkiaKAACApIkiAAAgaaIIAABImigCAACSJooAAICkiSIAACBpoggAAEiaKAIAAJImigAAgKSJIgAAIGmiCAAASJooAgAAkiaKAACApIkiAAAgaaIIAABImigCAACSJooAAICkiSIAACBpoggAAEiaKAIAAJJWpyiaOHFidOnSJYqLi6Nv374xd+7cHa47ZcqU6N+/f7Ru3Tpat24d5eXlH7s+AADA7pR3FE2bNi0qKipi7NixMX/+/OjRo0cMHDgwVq9eXev6s2fPjjPOOCOeeeaZmDNnTpSVlcXxxx8fb7755qcePAAAwKdVkGVZls8Gffv2jS9/+ctxxx13REREVVVVlJWVxcUXXxyjRo36xO23bdsWrVu3jjvuuCOGDRu2U/e5fv36aNmyZaxbty5KSkryGS4AAPAZUh9tkNeRosrKypg3b16Ul5f/YwcNGkR5eXnMmTNnp/axadOm2LJlS+yzzz75jRQAAKAeNMpn5bVr18a2bduitLS02vLS0tJYtGjRTu3jyiuvjI4dO1YLq4/avHlzbN68Off1+vXr8xkmAADATtutV5+7+eabY+rUqfHoo49GcXHxDtcbP358tGzZMncrKyvbjaMEAABSklcUtWnTJho2bBirVq2qtnzVqlXRvn37j9321ltvjZtvvjmeeuqpOPzwwz923dGjR8e6detytxUrVuQzTAAAgJ2WVxQVFhZGr169YtasWbllVVVVMWvWrOjXr98Ot7vlllvi+uuvjxkzZkTv3r0/8X6KioqipKSk2g0AAKA+5HVOUURERUVFDB8+PHr37h19+vSJCRMmxMaNG2PEiBERETFs2LDo1KlTjB8/PiIi/uM//iPGjBkTDzzwQHTp0iVWrlwZERHNmzeP5s2b78KHAgAAkL+8o2jIkCGxZs2aGDNmTKxcuTJ69uwZM2bMyF18Yfny5dGgwT8OQN11111RWVkZp512WrX9jB07Nq677rpPN3oAAIBPKe/PKdoTfE4RAAAQ8U/wOUUAAACfNaIIAABImigCAACSJooAAICkiSIAACBpoggAAEiaKAIAAJImigAAgKSJIgAAIGmiCAAASJooAgAAkiaKAACApIkiAAAgaaIIAABImigCAACSJooAAICkiSIAACBpoggAAEiaKAIAAJImigAAgKSJIgAAIGmiCAAASJooAgAAkiaKAACApIkiAAAgaaIIAABImigCAACSJooAAICkiSIAACBpoggAAEiaKAIAAJImigAAgKSJIgAAIGmiCAAASJooAgAAkiaKAACApIkiAAAgaaIIAABImigCAACSJooAAICkiSIAACBpoggAAEiaKAIAAJImigAAgKSJIgAAIGmiCAAASJooAgAAkiaKAACApIkiAAAgaaIIAABImigCAACSJooAAICkiSIAACBpoggAAEiaKAIAAJImigAAgKSJIgAAIGmiCAAASJooAgAAkiaKAACApIkiAAAgaaIIAABImigCAACSJooAAICkiSIAACBpoggAAEiaKAIAAJImigAAgKSJIgAAIGmiCAAASJooAgAAkiaKAACApIkiAAAgaaIIAABImigCAACSJooAAICkiSIAACBpoggAAEiaKAIAAJImigAAgKSJIgAAIGmiCAAASFqdomjixInRpUuXKC4ujr59+8bcuXM/dv0HH3wwunfvHsXFxXHYYYfF9OnT6zRYAACAXS3vKJo2bVpUVFTE2LFjY/78+dGjR48YOHBgrF69utb1n3vuuTjjjDPinHPOiQULFsSgQYNi0KBB8dJLL33qwQMAAHxaBVmWZfls0Ldv3/jyl78cd9xxR0REVFVVRVlZWVx88cUxatSoGusPGTIkNm7cGI8//nhu2RFHHBE9e/aMSZMm7dR9rl+/Plq2bBnr1q2LkpKSfIYLAAB8htRHGzTKZ+XKysqYN29ejB49OresQYMGUV5eHnPmzKl1mzlz5kRFRUW1ZQMHDoxf/vKXO7yfzZs3x+bNm3Nfr1u3LiI+eAIAAIB0bW+CPI/tfKy8omjt2rWxbdu2KC0trba8tLQ0Fi1aVOs2K1eurHX9lStX7vB+xo8fH+PGjauxvKysLJ/hAgAAn1Fvv/12tGzZcpfsK68o2l1Gjx5d7ejSe++9F507d47ly5fvsgcOtVm/fn2UlZXFihUrvFWTemWusbuYa+wu5hq7y7p162LfffeNffbZZ5ftM68oatOmTTRs2DBWrVpVbfmqVauiffv2tW7Tvn37vNaPiCgqKoqioqIay1u2bOlFxm5RUlJirrFbmGvsLuYau4u5xu7SoMGu+3ShvPZUWFgYvXr1ilmzZuWWVVVVxaxZs6Jfv361btOvX79q60dEzJw5c4frAwAA7E55v32uoqIihg8fHr17944+ffrEhAkTYuPGjTFixIiIiBg2bFh06tQpxo8fHxERl1xySQwYMCB+9KMfxYknnhhTp06NF154ISZPnrxrHwkAAEAd5B1FQ4YMiTVr1sSYMWNi5cqV0bNnz5gxY0buYgrLly+vdijryCOPjAceeCCuueaauOqqq+KAAw6IX/7yl3HooYfu9H0WFRXF2LFja31LHexK5hq7i7nG7mKusbuYa+wu9THX8v6cIgAAgM+SXXd2EgAAwF5IFAEAAEkTRQAAQNJEEQAAkLR/miiaOHFidOnSJYqLi6Nv374xd+7cj13/wQcfjO7du0dxcXEcdthhMX369N00UvZ2+cy1KVOmRP/+/aN169bRunXrKC8v/8S5Cdvl++/adlOnTo2CgoIYNGhQ/Q6Qz4x859p7770XI0eOjA4dOkRRUVEceOCB/j/KTsl3rk2YMCEOOuigaNKkSZSVlcVll10W77///m4aLXujZ599Nk4++eTo2LFjFBQUxC9/+ctP3Gb27NnxpS99KYqKimL//feP++67L+/7/aeIomnTpkVFRUWMHTs25s+fHz169IiBAwfG6tWra13/ueeeizPOOCPOOeecWLBgQQwaNCgGDRoUL7300m4eOXubfOfa7Nmz44wzzohnnnkm5syZE2VlZXH88cfHm2++uZtHzt4m37m23bJly+Lyyy+P/v3776aRsrfLd65VVlbGcccdF8uWLYuHHnooFi9eHFOmTIlOnTrt5pGzt8l3rj3wwAMxatSoGDt2bLzyyitx9913x7Rp0+Kqq67azSNnb7Jx48bo0aNHTJw4cafWX7p0aZx44olx7LHHxsKFC+PSSy+Nc889N5588sn87jj7J9CnT59s5MiRua+3bduWdezYMRs/fnyt6w8ePDg78cQTqy3r27dvdsEFF9TrONn75TvXPmrr1q1ZixYtsvvvv7++hshnRF3m2tatW7Mjjzwy+8lPfpINHz48+9d//dfdMFL2dvnOtbvuuivr2rVrVllZubuGyGdEvnNt5MiR2Ve/+tVqyyoqKrKjjjqqXsfJZ0dEZI8++ujHrnPFFVdkhxxySLVlQ4YMyQYOHJjXfe3xI0WVlZUxb968KC8vzy1r0KBBlJeXx5w5c2rdZs6cOdXWj4gYOHDgDteHiLrNtY/atGlTbNmyJfbZZ5/6GiafAXWdaz/4wQ+iXbt2cc455+yOYfIZUJe59thjj0W/fv1i5MiRUVpaGoceemjcdNNNsW3btt01bPZCdZlrRx55ZMybNy/3FrslS5bE9OnT44QTTtgtYyYNu6oLGu3KQdXF2rVrY9u2bVFaWlpteWlpaSxatKjWbVauXFnr+itXrqy3cbL3q8tc+6grr7wyOnbsWOPFBx9Wl7n2m9/8Ju6+++5YuHDhbhghnxV1mWtLliyJp59+Os4666yYPn16vPbaa/Gd73wntmzZEmPHjt0dw2YvVJe5duaZZ8batWvj6KOPjizLYuvWrXHhhRd6+xy71I66YP369fH3v/89mjRpslP72eNHimBvcfPNN8fUqVPj0UcfjeLi4j09HD5DNmzYEEOHDo0pU6ZEmzZt9vRw+IyrqqqKdu3axeTJk6NXr14xZMiQuPrqq2PSpEl7emh8xsyePTtuuummuPPOO2P+/PnxyCOPxBNPPBHXX3/9nh4a1LDHjxS1adMmGjZsGKtWraq2fNWqVdG+fftat2nfvn1e60NE3ebadrfeemvcfPPN8etf/zoOP/zw+hwmnwH5zrXXX389li1bFieffHJuWVVVVURENGrUKBYvXhzdunWr30GzV6rLv2sdOnSIxo0bR8OGDXPLDj744Fi5cmVUVlZGYWFhvY6ZvVNd5tq1114bQ4cOjXPPPTciIg477LDYuHFjnH/++XH11VdHgwb+Ns+nt6MuKCkp2emjRBH/BEeKCgsLo1evXjFr1qzcsqqqqpg1a1b069ev1m369etXbf2IiJkzZ+5wfYio21yLiLjlllvi+uuvjxkzZkTv3r13x1DZy+U717p37x4vvvhiLFy4MHf7xje+kbuSTllZ2e4cPnuRuvy7dtRRR8Vrr72WC++IiD//+c/RoUMHQcQO1WWubdq0qUb4bI/xD86hh09vl3VBfteAqB9Tp07NioqKsvvuuy/705/+lJ1//vlZq1atspUrV2ZZlmVDhw7NRo0alVv/t7/9bdaoUaPs1ltvzV555ZVs7NixWePGjbMXX3xxTz0E9hL5zrWbb745KywszB566KHsrbfeyt02bNiwpx4Ce4l859pHufocOyvfubZ8+fKsRYsW2Xe/+91s8eLF2eOPP561a9cuu+GGG/bUQ2Avke9cGzt2bNaiRYvsf/7nf7IlS5ZkTz31VNatW7ds8ODBe+ohsBfYsGFDtmDBgmzBggVZRGS33XZbtmDBguyNN97IsizLRo0alQ0dOjS3/pIlS7KmTZtm//7v/5698sor2cSJE7OGDRtmM2bMyOt+/ymiKMuy7Pbbb8/23XffrLCwMOvTp0/2/PPP5743YMCAbPjw4dXW/8UvfpEdeOCBWWFhYXbIIYdkTzzxxG4eMXurfOZa586ds4iocRs7duzuHzh7nXz/XfswUUQ+8p1rzz33XNa3b9+sqKgo69q1a3bjjTdmW7du3c2jZm+Uz1zbsmVLdt1112XdunXLiouLs7Kysuw73/lO9u677+7+gbPXeOaZZ2r93Wv73Bo+fHg2YMCAGtv07NkzKywszLp27Zrde++9ed9vQZY5fgkAAKRrj59TBAAAsCeJIgAAIGmiCAAASJooAgAAkiaKAACApIkiAAAgaaIIAABImigCAACSJooAAICkiSIAACBpoggAAEiaKAIAAJL2/wPMLL4JXEh39wAAAABJRU5ErkJggg==", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/tmp/ipykernel_10337/4129238603.py:131: FutureWarning: \n", + "\n", + "Passing `palette` without assigning `hue` is deprecated and will be removed in v0.14.0. Assign the `x` variable to `hue` and set `legend=False` for the same effect.\n", + "\n", + " sns.barplot(\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "🏆 FINAL VERDICT:\n", + "✅ SUPERHUMAN: The model is primarily looking at the DNA (64.9%)!\n" + ] + } + ], + "source": [ + "# ==========================================\n", + "# 📊 CELL 5: SHAP EXPLAINABILITY (DNA vs Tabular) — FIXED\n", + "# ==========================================\n", + "!pip install -q shap matplotlib seaborn\n", + "\n", + "import shap\n", + "import torch\n", + "import numpy as np\n", + "import pandas as pd\n", + "import matplotlib.pyplot as plt\n", + "import seaborn as sns\n", + "from tqdm import tqdm\n", + "\n", + "print(\"🧠 INITIALIZING SHAP ANALYSIS...\")\n", + "\n", + "# 1. CONFIGURATION\n", + "# SHAP is expensive. We use a representative subset.\n", + "N_BACKGROUND = 100\n", + "N_TEST = 50\n", + "TOTAL_SAMPLES = N_BACKGROUND + N_TEST\n", + "\n", + "# 2. DEFINE SHAP WRAPPER (Focus on Classifier Head)\n", + "class HeadWrapper(torch.nn.Module):\n", + " def __init__(self, classifier):\n", + " super().__init__()\n", + " self.classifier = classifier\n", + "\n", + " def forward(self, combined_input):\n", + " return self.classifier(combined_input)\n", + "\n", + "# Handle compiled models\n", + "raw_model = model if not hasattr(model, \"_orig_mod\") else model._orig_mod\n", + "shap_model = HeadWrapper(raw_model.classifier).to(\"cpu\").eval()\n", + "\n", + "# 3. GENERATE LATENT EMBEDDINGS\n", + "print(f\"📉 Extracting embeddings for {TOTAL_SAMPLES} samples...\")\n", + "\n", + "dataloader = torch.utils.data.DataLoader(\n", + " test_fast, \n", + " batch_size=32, \n", + " shuffle=True, \n", + " collate_fn=data_collator\n", + ")\n", + "\n", + "latent_vectors = []\n", + "collected = 0\n", + "\n", + "model.eval()\n", + "\n", + "# --- FIX IS HERE: Get device from the first parameter ---\n", + "device = next(model.parameters()).device \n", + "# --------------------------------------------------------\n", + "\n", + "for batch in tqdm(dataloader, desc=\"Encoding Data\"):\n", + " if collected >= TOTAL_SAMPLES: break\n", + " \n", + " with torch.no_grad():\n", + " input_ids = batch[\"input_ids\"].to(device)\n", + " mask = batch[\"attention_mask\"].to(device)\n", + " # Ensure features are float32 before moving to device\n", + " feats = batch[\"features\"].to(dtype=torch.float32).to(device)\n", + " \n", + " # A. Get DNA Embedding\n", + " enc_module = raw_model.encoder\n", + " out = enc_module(input_ids=input_ids, attention_mask=mask, return_dict=True)\n", + " \n", + " # Mean Pooling\n", + " if hasattr(out, \"pooler_output\") and out.pooler_output is not None:\n", + " pooled = out.pooler_output\n", + " else:\n", + " hidden = out.last_hidden_state\n", + " mask_exp = mask.unsqueeze(-1)\n", + " pooled = (hidden * mask_exp).sum(1) / mask_exp.sum(1).clamp(min=1e-9)\n", + " \n", + " # B. Concatenate\n", + " combined = torch.cat([pooled, feats], dim=1).cpu()\n", + " latent_vectors.append(combined)\n", + " collected += len(batch[\"input_ids\"])\n", + "\n", + "# Prepare Tensors\n", + "full_data = torch.cat(latent_vectors, dim=0)\n", + "background_data = full_data[:N_BACKGROUND]\n", + "test_data = full_data[N_BACKGROUND : N_BACKGROUND+N_TEST]\n", + "\n", + "print(f\"✅ Data Prepared. Background shape: {background_data.shape}\")\n", + "\n", + "# 4. RUN SHAP (GradientExplainer)\n", + "print(\"⚡ Calculating SHAP values (GradientExplainer)...\")\n", + "explainer = shap.GradientExplainer(shap_model, background_data)\n", + "shap_values = explainer.shap_values(test_data)\n", + "\n", + "# Handle list output\n", + "if isinstance(shap_values, list):\n", + " vals = shap_values[1]\n", + "else:\n", + " vals = shap_values\n", + "\n", + "print(\"✅ SHAP Calculation Complete.\")\n", + "\n", + "# ==========================================\n", + "# 📈 PLOTTING\n", + "# ==========================================\n", + "\n", + "# A. TABULAR FEATURE IMPORTANCE\n", + "# We look only at the last columns corresponding to features\n", + "tabular_shap = vals[:, -len(feature_cols):]\n", + "tabular_data = test_data[:, -len(feature_cols):].numpy()\n", + "\n", + "plt.figure(figsize=(10, 6))\n", + "plt.title(\"Impact of Clinical Features on Prediction\")\n", + "shap.summary_plot(\n", + " tabular_shap, \n", + " tabular_data, \n", + " feature_names=feature_cols, \n", + " show=False,\n", + " plot_type=\"dot\"\n", + ")\n", + "plt.tight_layout()\n", + "plt.savefig(\"shap_tabular_beeswarm.png\", dpi=300)\n", + "plt.show()\n", + "\n", + "# B. DNA vs TABULAR BATTLE\n", + "dna_impact = np.abs(vals[:, :-len(feature_cols)]).sum(axis=1).mean()\n", + "tab_impact = np.abs(vals[:, -len(feature_cols):]).sum(axis=1).mean()\n", + "\n", + "total = dna_impact + tab_impact\n", + "dna_pct = (dna_impact / total) * 100\n", + "tab_pct = (tab_impact / total) * 100\n", + "\n", + "plt.figure(figsize=(8, 5))\n", + "sns.barplot(\n", + " x=[\"DNA Sequence\", \"Clinical Features\"], \n", + " y=[dna_impact, tab_impact], \n", + " palette=[\"#2ecc71\", \"#3498db\"]\n", + ")\n", + "plt.title(f\"Model Intelligence Source\\n(DNA: {dna_pct:.1f}% vs Features: {tab_pct:.1f}%)\")\n", + "plt.ylabel(\"Mean Absolute SHAP Value\")\n", + "plt.savefig(\"shap_modality_comparison.png\", dpi=300)\n", + "plt.show()\n", + "\n", + "print(\"\\n🏆 FINAL VERDICT:\")\n", + "if dna_pct > 50:\n", + " print(f\"✅ SUPERHUMAN: The model is primarily looking at the DNA ({dna_pct:.1f}%)!\")\n", + "else:\n", + " print(f\"⚖️ HYBRID: The model relies heavily on conservation scores ({tab_pct:.1f}%)!\")" + ] + }, + { + "cell_type": "code", + "execution_count": 28, + "metadata": {}, + "outputs": [], + "source": [ + "# ======================================================\n", + "# 🔌 Bridge: DataFrame → HF-style Torch batch for Ablation\n", + "# ======================================================\n", + "\n", + "def prepare_batch(df_batch):\n", + " # 1) Tokenize DNA\n", + " enc = tokenizer(\n", + " df_batch[\"raw_sequence\"].tolist(),\n", + " truncation=True,\n", + " padding=\"max_length\",\n", + " max_length=MAX_LEN,\n", + " return_tensors=\"pt\"\n", + " )\n", + "\n", + " # 2) Pack features in correct order\n", + " feats = []\n", + " for c in feature_cols:\n", + " feats.append(df_batch[c].astype(np.float32).values)\n", + "\n", + " feats = np.stack(feats, axis=1) # (batch, n_features)\n", + "\n", + " # 3) Labels\n", + " labels = (df_batch[\"clean_label\"] == \"Pathogenic\").astype(np.int64).values\n", + "\n", + " return {\n", + " \"input_ids\": enc[\"input_ids\"],\n", + " \"attention_mask\": enc[\"attention_mask\"],\n", + " \"features\": torch.tensor(feats, dtype=torch.float32),\n", + " \"labels\": torch.tensor(labels, dtype=torch.long),\n", + " }\n" + ] + }, + { + "cell_type": "code", + "execution_count": 32, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "🧪 STARTING FULL ABLATION SUITE (Tensor-Native)...\n", + "🔥 Model locked to: cuda\n", + "📊 Establishing Baseline Performance...\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "Testing: Baseline Model: 1%| | 1/110 [00:01<02:15, 1.25s/it]" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + " \r" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "✅ Baseline AUC established: 0.9182\n", + "\n", + "🔥 RUNNING TENSOR ABLATION TESTS...\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + " \r" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "🏆 ABLATION IMPACT RANKING (What matters most?):\n", + " name auc drop\n", + " No DNA + No Conservation 0.517988 0.400191\n", + " No DNA + No gnomAD 0.557964 0.360215\n", + " No DNA (Sequence Blind) 0.558319 0.359860\n", + " No PhastCons 0.901000 0.017178\n", + " No Tabular (Pure DNA) 0.908187 0.009992\n", + " No gnomAD + No Conservation 0.908187 0.009992\n", + "No Conservation (All Scores) 0.911717 0.006462\n", + " No gnomAD (Freq Blind) 0.915300 0.002879\n", + " No PhyloP 0.917488 0.000691\n", + " No GERP (Evo Blind) 0.922640 -0.004461\n" + ] + } + ], + "source": [ + "# ==========================================\n", + "# 📉 CELL 2: FULL ABLATION STUDY (Tensor-Native) — FIXED\n", + "# ==========================================\n", + "import matplotlib.pyplot as plt\n", + "import pandas as pd\n", + "import numpy as np\n", + "import torch\n", + "from tqdm import tqdm\n", + "from sklearn.metrics import roc_curve, auc\n", + "from torch.utils.data import DataLoader\n", + "\n", + "print(\"🧪 STARTING FULL ABLATION SUITE (Tensor-Native)...\")\n", + "\n", + "# 1. SETUP DEVICE & MODEL\n", + "# Explicitly force GPU to fix \"Device Mismatch\" error\n", + "DEVICE = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n", + "model.to(DEVICE)\n", + "print(f\"🔥 Model locked to: {DEVICE}\")\n", + "\n", + "# 2. DEFINE ABLATION FUNCTION\n", + "def evaluate_tensor_ablation(test_config, baseline_auc=None):\n", + " removals = test_config.get(\"remove\", [])\n", + " name = test_config[\"name\"]\n", + " \n", + " all_preds = []\n", + " all_labels = []\n", + " BATCH_SIZE = 128\n", + " \n", + " # Create DataLoader from 'test_fast'\n", + " loader = DataLoader(test_fast, batch_size=BATCH_SIZE, shuffle=False, collate_fn=data_collator)\n", + " \n", + " model.eval()\n", + " \n", + " for batch in tqdm(loader, desc=f\"Testing: {name}\", leave=False):\n", + " # Move inputs to GPU\n", + " input_ids = batch[\"input_ids\"].to(DEVICE)\n", + " mask = batch[\"attention_mask\"].to(DEVICE)\n", + " features = batch[\"features\"].to(DEVICE)\n", + " labels = batch[\"labels\"].to(\"cpu\") # Keep labels on CPU for sklearn metrics\n", + " \n", + " # --- DESTRUCTION LOGIC ---\n", + " \n", + " # 1. Kill DNA? (Shuffle input_ids within batch)\n", + " if \"DNA\" in removals:\n", + " perm = torch.randperm(input_ids.size(0))\n", + " input_ids = input_ids[perm]\n", + " mask = mask[perm]\n", + " \n", + " # 2. Kill Tabular Features? (Zero out columns)\n", + " if \"gnomAD\" in removals:\n", + " indices = [i for i, c in enumerate(feature_cols) if \"gnomad\" in c.lower()]\n", + " for idx in indices:\n", + " features[:, idx] = 0.0\n", + " \n", + " if \"GERP\" in removals:\n", + " indices = [i for i, c in enumerate(feature_cols) if \"gerp\" in c.lower()]\n", + " for idx in indices:\n", + " features[:, idx] = 0.0\n", + "\n", + " if \"PhyloP\" in removals:\n", + " indices = [i for i, c in enumerate(feature_cols) if \"phylop\" in c.lower()]\n", + " for idx in indices:\n", + " features[:, idx] = 0.0\n", + " \n", + " if \"PhastCons\" in removals:\n", + " indices = [i for i, c in enumerate(feature_cols) if \"phastcons\" in c.lower()]\n", + " for idx in indices:\n", + " features[:, idx] = 0.0\n", + " \n", + " if \"All_Conservation\" in removals:\n", + " gnomad_idxs = [i for i, c in enumerate(feature_cols) if \"gnomad\" in c.lower()]\n", + " # Kill everything that is NOT gnomAD\n", + " for i in range(features.size(1)):\n", + " if i not in gnomad_idxs:\n", + " features[:, i] = 0.0\n", + " \n", + " if \"All_Tabular\" in removals:\n", + " features.fill_(0.0)\n", + "\n", + " # --- INFERENCE ---\n", + " with torch.no_grad():\n", + " # Ensure float32 for features\n", + " features = features.to(dtype=torch.float32)\n", + " \n", + " out = model(input_ids=input_ids, attention_mask=mask, features=features)\n", + " logits = out.logits\n", + " probs = torch.softmax(logits, dim=1)[:, 1]\n", + " \n", + " all_preds.extend(probs.cpu().numpy())\n", + " all_labels.extend(labels.numpy())\n", + " \n", + " # Metrics\n", + " fpr, tpr, _ = roc_curve(all_labels, all_preds)\n", + " roc_auc = auc(fpr, tpr)\n", + " \n", + " drop = 0.0\n", + " if baseline_auc is not None:\n", + " drop = baseline_auc - roc_auc\n", + " \n", + " return {\"name\": name, \"auc\": roc_auc, \"drop\": drop, \"fpr\": fpr, \"tpr\": tpr}\n", + "\n", + "# 3. CALCULATE BASELINE\n", + "print(\"📊 Establishing Baseline Performance...\")\n", + "baseline_res = evaluate_tensor_ablation({\"name\": \"Baseline Model\", \"remove\": []})\n", + "baseline_auc = baseline_res[\"auc\"]\n", + "print(f\"✅ Baseline AUC established: {baseline_auc:.4f}\")\n", + "\n", + "# 4. DEFINE ABLATION GROUPS\n", + "ablation_tests = [\n", + " {\"name\": \"No DNA (Sequence Blind)\", \"remove\": [\"DNA\"]},\n", + " {\"name\": \"No gnomAD (Freq Blind)\", \"remove\": [\"gnomAD\"]},\n", + " {\"name\": \"No GERP (Evo Blind)\", \"remove\": [\"GERP\"]},\n", + " {\"name\": \"No PhyloP\", \"remove\": [\"PhyloP\"]},\n", + " {\"name\": \"No PhastCons\", \"remove\": [\"PhastCons\"]},\n", + " {\"name\": \"No Conservation (All Scores)\", \"remove\": [\"All_Conservation\"]},\n", + " {\"name\": \"No DNA + No gnomAD\", \"remove\": [\"DNA\", \"gnomAD\"]},\n", + " {\"name\": \"No DNA + No Conservation\", \"remove\": [\"DNA\", \"All_Conservation\"]},\n", + " {\"name\": \"No gnomAD + No Conservation\", \"remove\": [\"gnomAD\", \"All_Conservation\"]},\n", + " {\"name\": \"No Tabular (Pure DNA)\", \"remove\": [\"All_Tabular\"]}\n", + "]\n", + "\n", + "# 5. RUN ALL TESTS\n", + "results = [baseline_res] \n", + "plt.figure(figsize=(12, 8))\n", + "plt.plot([0, 1], [0, 1], color='navy', lw=2, linestyle='--')\n", + "\n", + "# Plot Baseline\n", + "plt.plot(baseline_res[\"fpr\"], baseline_res[\"tpr\"], lw=3, color='black', \n", + " label=f'Baseline (AUC={baseline_auc:.3f})')\n", + "\n", + "print(\"\\n🔥 RUNNING TENSOR ABLATION TESTS...\")\n", + "for test in ablation_tests:\n", + " res = evaluate_tensor_ablation(test, baseline_auc=baseline_auc)\n", + " results.append(res)\n", + " \n", + " plt.plot(res[\"fpr\"], res[\"tpr\"], lw=1.5, alpha=0.7, label=f'{res[\"name\"]} (AUC={res[\"auc\"]:.3f})')\n", + "\n", + "# 6. FINALIZE PLOT\n", + "plt.xlim([0.0, 1.0])\n", + "plt.ylim([0.0, 1.05])\n", + "plt.xlabel('False Positive Rate')\n", + "plt.ylabel('True Positive Rate')\n", + "plt.title('PathoPreter Ablation: Feature Importance Analysis')\n", + "plt.legend(loc=\"lower right\", fontsize='small')\n", + "plt.grid(True, alpha=0.3)\n", + "plt.savefig(\"ablation_study_10_tests.png\", dpi=300)\n", + "plt.show()\n", + "\n", + "# 7. PRINT LEADERBOARD\n", + "print(\"\\n🏆 ABLATION IMPACT RANKING (What matters most?):\")\n", + "res_df = pd.DataFrame([r for r in results if r[\"name\"] != \"Baseline Model\"])\n", + "res_df = res_df[[\"name\", \"auc\", \"drop\"]]\n", + "res_df = res_df.sort_values(\"drop\", ascending=False)\n", + "print(res_df.to_string(index=False))" + ] + }, + { + "cell_type": "code", + "execution_count": 24, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "📦 PACKAGING MODEL FROM: ./PathoPreter_Final_A100\n", + "✅ Saved model_metadata.json (feature columns preserved)\n", + "✅ Generated modeling_pathopreter.py (Auto-load script)\n", + "🚀 Zipping to PathoPreter_A100_Complete.zip...\n", + " - Storing (Fast): model.safetensors\n", + "\n", + "✅ SUCCESS! Download 'PathoPreter_A100_Complete.zip' now.\n", + " It contains the Model, Tokenizer, Scalers, and Python Class Code.\n" + ] + } + ], + "source": [ + "# ==========================================\n", + "# 📦 CELL 6: THE \"TIME CAPSULE\" PACKAGER\n", + "# ==========================================\n", + "import json\n", + "import zipfile\n", + "import os\n", + "from tqdm import tqdm\n", + "\n", + "# Configuration\n", + "SOURCE_DIR = \"./PathoPreter_Final_A100\"\n", + "OUTPUT_ZIP = \"PathoPreter_A100_Complete.zip\"\n", + "\n", + "print(f\"📦 PACKAGING MODEL FROM: {SOURCE_DIR}\")\n", + "\n", + "# 1. SAVE CRITICAL METADATA (Column Names)\n", + "# This is the #1 thing people forget 2 months later!\n", + "metadata = {\n", + " \"feature_cols\": feature_cols,\n", + " \"num_labels\": NUM_LABELS,\n", + " \"feature_dim\": FEATURE_DIM,\n", + " \"model_type\": \"nt-500m-hybrid\",\n", + " \"base_model\": \"InstaDeepAI/nucleotide-transformer-500m-human-ref\"\n", + "}\n", + "\n", + "with open(os.path.join(SOURCE_DIR, \"model_metadata.json\"), \"w\") as f:\n", + " json.dump(metadata, f, indent=4)\n", + "print(\"✅ Saved model_metadata.json (feature columns preserved)\")\n", + "\n", + "# 2. GENERATE PYTHON MODEL CLASS FILE\n", + "# We write the class definition to a .py file so you can import it easily later\n", + "model_code = \"\"\"\n", + "import torch\n", + "import torch.nn as nn\n", + "from transformers import AutoModel\n", + "from transformers.modeling_outputs import SequenceClassifierOutput\n", + "\n", + "class SequenceClassificationWithFeatures(nn.Module):\n", + " def __init__(self, encoder, hidden_size, feature_dim, num_labels):\n", + " super().__init__()\n", + " self.encoder = encoder\n", + " self.feature_dim = feature_dim\n", + " self.classifier = nn.Sequential(\n", + " nn.Linear(hidden_size + feature_dim, hidden_size),\n", + " nn.GELU(),\n", + " nn.Dropout(0.1),\n", + " nn.Linear(hidden_size, num_labels)\n", + " )\n", + "\n", + " @classmethod\n", + " def from_pretrained(cls, path, device=\"cpu\"):\n", + " import json\n", + " import os\n", + " from safetensors.torch import load_file\n", + " \n", + " # Load Config\n", + " with open(os.path.join(path, \"model_metadata.json\"), \"r\") as f:\n", + " meta = json.load(f)\n", + " \n", + " # Load Encoder\n", + " encoder = AutoModel.from_pretrained(path, trust_remote_code=True)\n", + " \n", + " # Init Model\n", + " model = cls(\n", + " encoder=encoder, \n", + " hidden_size=encoder.config.hidden_size,\n", + " feature_dim=meta[\"feature_dim\"],\n", + " num_labels=meta[\"num_labels\"]\n", + " )\n", + " \n", + " # Load Weights\n", + " state_dict = load_file(os.path.join(path, \"model.safetensors\"))\n", + " model.load_state_dict(state_dict)\n", + " \n", + " return model.to(device)\n", + "\n", + " def forward(self, input_ids=None, attention_mask=None, features=None, labels=None):\n", + " out = self.encoder(input_ids=input_ids, attention_mask=attention_mask, return_dict=True)\n", + "\n", + " if hasattr(out, \"pooler_output\") and out.pooler_output is not None:\n", + " pooled = out.pooler_output\n", + " else:\n", + " last = out.last_hidden_state\n", + " mask = attention_mask.unsqueeze(-1)\n", + " pooled = (last * mask).sum(1) / mask.sum(1).clamp(min=1e-9)\n", + "\n", + " # Cast features to match model dtype/device\n", + " features = features.to(pooled.device).to(pooled.dtype)\n", + " x = torch.cat([pooled, features], dim=1)\n", + " logits = self.classifier(x)\n", + "\n", + " loss = None\n", + " if labels is not None:\n", + " loss = nn.CrossEntropyLoss()(logits, labels)\n", + "\n", + " return SequenceClassifierOutput(loss=loss, logits=logits)\n", + "\"\"\"\n", + "\n", + "with open(os.path.join(SOURCE_DIR, \"modeling_pathopreter.py\"), \"w\") as f:\n", + " f.write(model_code)\n", + "print(\"✅ Generated modeling_pathopreter.py (Auto-load script)\")\n", + "\n", + "# 3. CREATE README INSTRUCTIONS\n", + "readme_text = \"\"\"\n", + "PATHO-PRETER v1 (A100 Fine-Tune)\n", + "================================\n", + "Trained by: Rohit\n", + "Base Model: Nucleotide Transformer 500M\n", + "Architecture: Hybrid (DNA Sequence + 9 Conservation Scores)\n", + "\n", + "HOW TO LOAD:\n", + "1. Unzip this folder.\n", + "2. Place 'modeling_pathopreter.py' in your python path.\n", + "3. Run:\n", + " \n", + " from modeling_pathopreter import SequenceClassificationWithFeatures\n", + " import numpy as np\n", + "\n", + " # Load Model\n", + " model = SequenceClassificationWithFeatures.from_pretrained(\"./PathoPreter_Final_A100\")\n", + " \n", + " # Load Scalers (Crucial for inference!)\n", + " means = np.load(\"./PathoPreter_Final_A100/feat_means.npy\")\n", + " stds = np.load(\"./PathoPreter_Final_A100/feat_stds.npy\")\n", + "\n", + " # Preprocessing\n", + " # (features - means) / stds\n", + "\"\"\"\n", + "with open(os.path.join(SOURCE_DIR, \"README.txt\"), \"w\") as f:\n", + " f.write(readme_text)\n", + "\n", + "# 4. SMART ZIP (No Compression for Weights)\n", + "print(f\"🚀 Zipping to {OUTPUT_ZIP}...\")\n", + "\n", + "with zipfile.ZipFile(OUTPUT_ZIP, 'w') as zf:\n", + " for root, dirs, files in os.walk(SOURCE_DIR):\n", + " for file in files:\n", + " file_path = os.path.join(root, file)\n", + " arcname = os.path.relpath(file_path, start=SOURCE_DIR)\n", + " \n", + " # DECISION: Compress text, Store binary weights (for speed)\n", + " if file.endswith(\".safetensors\") or file.endswith(\".bin\"):\n", + " compression = zipfile.ZIP_STORED\n", + " print(f\" - Storing (Fast): {arcname}\")\n", + " else:\n", + " compression = zipfile.ZIP_DEFLATED\n", + " \n", + " zf.write(file_path, arcname, compress_type=compression)\n", + "\n", + "print(f\"\\n✅ SUCCESS! Download '{OUTPUT_ZIP}' now.\")\n", + "print(\" It contains the Model, Tokenizer, Scalers, and Python Class Code.\")" + ] + }, + { + "cell_type": "code", + "execution_count": 35, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Samples: 14073 14073\n", + "Real AUC: 0.9182\n", + "Permutation AUC:0.5044\n", + "✅ No leakage or memorization detected\n" + ] + } + ], + "source": [ + "from sklearn.metrics import roc_auc_score\n", + "import numpy as np\n", + "\n", + "# Use the same objects you created in calibration cell\n", + "# These came from: preds = trainer.predict(test_fast)\n", + "\n", + "y_test = preds.label_ids # shape = (N_test,)\n", + "logits = preds.predictions # shape = (N_test, 2)\n", + "probs = torch.softmax(torch.tensor(logits), dim=-1)[:,1].numpy()\n", + "\n", + "print(\"Samples:\", len(y_test), len(probs))\n", + "\n", + "# --- real AUC ---\n", + "real_auc = roc_auc_score(y_test, probs)\n", + "\n", + "# --- permutation ---\n", + "perm = np.random.permutation(y_test)\n", + "perm_auc = roc_auc_score(perm, probs)\n", + "\n", + "print(f\"Real AUC: {real_auc:.4f}\")\n", + "print(f\"Permutation AUC:{perm_auc:.4f}\")\n", + "\n", + "if perm_auc < 0.55:\n", + " print(\"✅ No leakage or memorization detected\")\n", + "else:\n", + " print(\"❌ Possible leakage — investigate splits\")\n" + ] + }, + { + "cell_type": "code", + "execution_count": 36, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "🧬 PathoPreter Performance by Allele Frequency\n", + "============================================================\n", + "Ultra-rare (<1e-6) | N= 8027 | Pathogenic%= 67.1% | AUC=0.9238\n", + "Rare (1e-6–1e-4) | N= 4665 | Pathogenic%= 34.7% | AUC=0.9069\n", + "Low-freq (1e-4–1e-2) | N= 967 | Pathogenic%= 5.5% | AUC=0.8219\n", + "Common (>1e-2) | N= 414 | Pathogenic%= 0.5% | AUC=0.9472\n", + "\n", + "📊 Summary\n" + ] + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
AF binNPathogenic %ROC-AUC
0Ultra-rare (<1e-6)80270.6713590.923828
1Rare (1e-6–1e-4)46650.3466240.906936
2Low-freq (1e-4–1e-2)9670.0548090.821890
3Common (>1e-2)4140.0048310.947209
\n", + "
" + ], + "text/plain": [ + " AF bin N Pathogenic % ROC-AUC\n", + "0 Ultra-rare (<1e-6) 8027 0.671359 0.923828\n", + "1 Rare (1e-6–1e-4) 4665 0.346624 0.906936\n", + "2 Low-freq (1e-4–1e-2) 967 0.054809 0.821890\n", + "3 Common (>1e-2) 414 0.004831 0.947209" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "# ======================================================\n", + "# 🧬 CELL: Performance by Allele Frequency (gnomAD bins)\n", + "# ======================================================\n", + "\n", + "import pandas as pd\n", + "import numpy as np\n", + "from sklearn.metrics import roc_auc_score\n", + "\n", + "# Reload test dataframe (same order as test_fast)\n", + "df = pd.read_parquet(\"test_enriched_SEQUENCES.parquet\").reset_index(drop=True)\n", + "\n", + "# Safety check\n", + "assert len(df) == len(probs), \"❌ Test set mismatch — check alignment\"\n", + "\n", + "# Attach predictions\n", + "df[\"PathoPreter\"] = probs\n", + "df[\"y_true\"] = y_test\n", + "\n", + "# Define frequency bins used in clinical genomics\n", + "bins = [0, 1e-6, 1e-4, 1e-2, 1]\n", + "labels = [\"Ultra-rare (<1e-6)\", \"Rare (1e-6–1e-4)\", \"Low-freq (1e-4–1e-2)\", \"Common (>1e-2)\"]\n", + "\n", + "df[\"AF_bin\"] = pd.cut(df[\"gnomad_af\"], bins=bins, labels=labels, include_lowest=True)\n", + "\n", + "print(\"\\n🧬 PathoPreter Performance by Allele Frequency\")\n", + "print(\"=\" * 60)\n", + "\n", + "rows = []\n", + "for bin in labels:\n", + " sub = df[df[\"AF_bin\"] == bin]\n", + " if len(sub) < 100:\n", + " print(f\"{bin:18s} | too few samples ({len(sub)})\")\n", + " continue\n", + " \n", + " if len(np.unique(sub[\"y_true\"])) < 2:\n", + " print(f\"{bin:18s} | only one class present\")\n", + " continue\n", + " \n", + " auc = roc_auc_score(sub[\"y_true\"], sub[\"PathoPreter\"])\n", + " pos_rate = sub[\"y_true\"].mean()\n", + " \n", + " rows.append((bin, len(sub), pos_rate, auc))\n", + " print(f\"{bin:18s} | N={len(sub):6d} | Pathogenic%={pos_rate*100:5.1f}% | AUC={auc:.4f}\")\n", + "\n", + "# Summary table\n", + "af_df = pd.DataFrame(rows, columns=[\"AF bin\", \"N\", \"Pathogenic %\", \"ROC-AUC\"])\n", + "\n", + "print(\"\\n📊 Summary\")\n", + "display(af_df)\n" + ] + }, + { + "cell_type": "code", + "execution_count": 37, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "🏥 CASE 1: Testing potential pathogenic variant...\n", + "\n", + "========================================\n", + "🧬 VARIANT DIAGNOSIS REPORT\n", + "========================================\n", + "Input Sequence: AAAAAAAAAA...AAAAAAAAAA\n", + "Prediction: 🟢 BENIGN\n", + "Probability: 0.0124\n", + "Confidence: 98.8%\n", + "========================================\n", + "\n", + "😊 CASE 2: Testing common benign variant...\n", + "\n", + "========================================\n", + "🧬 VARIANT DIAGNOSIS REPORT\n", + "========================================\n", + "Input Sequence: TTTTTTTTTT...TTTTTTTTTT\n", + "Prediction: 🟢 BENIGN\n", + "Probability: 0.0012\n", + "Confidence: 99.9%\n", + "========================================\n", + "\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/tmp/ipykernel_10337/1800919678.py:47: UserWarning: Creating a tensor from a list of numpy.ndarrays is extremely slow. Please consider converting the list to a single numpy.ndarray with numpy.array() before converting to a tensor. (Triggered internally at /pytorch/torch/csrc/utils/tensor_new.cpp:253.)\n", + " features_tensor = torch.tensor([feats], dtype=torch.float32).to(DEVICE)\n" + ] + }, + { + "data": { + "text/plain": [ + "0.001151399570517242" + ] + }, + "execution_count": 37, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "# ==========================================\n", + "# 🚀 CELL 7: THE \"SINGLE VARIANT PREDICTOR\" (The App)\n", + "# ==========================================\n", + "import torch\n", + "import numpy as np\n", + "\n", + "def predict_single_variant(dna_sequence, gnomad_af=0.0, conservation_scores=None):\n", + " \"\"\"\n", + " Simulates a real-world clinical query.\n", + " dna_sequence: 60bp string (center is the mutation)\n", + " gnomad_af: Allele Frequency (0.0 to 1.0)\n", + " conservation_scores: list of [GERP, PhyloP, etc...] (Defaults to averages if None)\n", + " \"\"\"\n", + " \n", + " # 1. Validation\n", + " dna_sequence = dna_sequence.upper().strip()\n", + " if len(dna_sequence) < 10:\n", + " return \"❌ Error: Sequence too short. Needs ~60bp context.\"\n", + " \n", + " # 2. Tokenize DNA (Using the k-mer tokenizer from training)\n", + " # Note: We manually simulate tokenization for this demo or use the loaded tokenizer\n", + " inputs = tokenizer(dna_sequence, return_tensors=\"pt\", padding=\"max_length\", max_length=tokenizer.model_max_length, truncation=True)\n", + " \n", + " # 3. Prepare Tabular Features\n", + " # If user doesn't provide detailed scores, we use the DATASET MEANS (Imputation)\n", + " # This is exactly what web-servers like CADD do for missing data.\n", + " if conservation_scores is None:\n", + " # Create a vector of zeros (representing mean values after normalization)\n", + " # We assume 0.0 because we standardized features (mean=0, std=1) during training\n", + " feats = np.zeros(len(feature_cols)) \n", + " else:\n", + " feats = np.array(conservation_scores)\n", + "\n", + " # 4. Inject Specific inputs (like gnomAD) if known\n", + " # Find gnomAD column index\n", + " try:\n", + " g_idx = [i for i, c in enumerate(feature_cols) if \"gnomad\" in c.lower()][0]\n", + " # Normalize gnomAD (Assuming approx mean=0.005, std=0.05 from training data stats)\n", + " # For the demo, we just set it raw, but in prod you load the scaler.\n", + " feats[g_idx] = (gnomad_af - feat_means[g_idx]) / feat_stds[g_idx]\n", + " except:\n", + " pass\n", + "\n", + " # 5. Convert to Tensor\n", + " input_ids = inputs[\"input_ids\"].to(DEVICE)\n", + " mask = inputs[\"attention_mask\"].to(DEVICE)\n", + " features_tensor = torch.tensor([feats], dtype=torch.float32).to(DEVICE)\n", + " \n", + " # 6. Predict\n", + " model.eval()\n", + " with torch.no_grad():\n", + " out = model(input_ids, mask, features_tensor)\n", + " prob = torch.softmax(out.logits, dim=1)[0, 1].item()\n", + " \n", + " # 7. Format Output\n", + " risk_level = \"🔴 PATHOGENIC\" if prob > 0.5 else \"🟢 BENIGN\"\n", + " confidence = prob if prob > 0.5 else 1 - prob\n", + " \n", + " print(\"\\n\" + \"=\"*40)\n", + " print(f\"🧬 VARIANT DIAGNOSIS REPORT\")\n", + " print(\"=\"*40)\n", + " print(f\"Input Sequence: {dna_sequence[:10]}...{dna_sequence[-10:]}\")\n", + " print(f\"Prediction: {risk_level}\")\n", + " print(f\"Probability: {prob:.4f}\")\n", + " print(f\"Confidence: {confidence:.1%}\")\n", + " print(\"=\"*40 + \"\\n\")\n", + " \n", + " return prob\n", + "\n", + "# --- TEST CASE 1: A Real Pathogenic Mutation (CFTR DeltaF508 equivalent context) ---\n", + "# High conservation (implicit), low frequency\n", + "seq_patho = \"A\" * 60 # Placeholder sequence for demo\n", + "print(\"🏥 CASE 1: Testing potential pathogenic variant...\")\n", + "predict_single_variant(seq_patho, gnomad_af=0.00001)\n", + "\n", + "# --- TEST CASE 2: A Common Benign Variant ---\n", + "# High frequency (50%)\n", + "seq_benign = \"T\" * 60 # Placeholder\n", + "print(\"😊 CASE 2: Testing common benign variant...\")\n", + "predict_single_variant(seq_benign, gnomad_af=0.50)" + ] + }, + { + "cell_type": "code", + "execution_count": 38, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\n", + "==================================================\n", + "🧬 PATHOPRETER — VARIANT REPORT\n", + "==================================================\n", + "Sequence (snippet): AAAAAAAAAAAA...AAAAAAAAAAAA\n", + "gnomAD AF: 4.500e-01\n", + "Prediction: 🟢 BENIGN\n", + "Pathogenic prob: 0.0000\n", + "==================================================\n", + "\n", + "\n", + "==================================================\n", + "🧬 PATHOPRETER — VARIANT REPORT\n", + "==================================================\n", + "Sequence (snippet): TTGCTGACCTGC...CTGAGGCTGACT\n", + "gnomAD AF: 1.000e-05\n", + "Prediction: 🟢 BENIGN\n", + "Pathogenic prob: 0.0018\n", + "==================================================\n", + "\n" + ] + }, + { + "data": { + "text/plain": [ + "0.0018386653391644359" + ] + }, + "execution_count": 38, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "import torch\n", + "import numpy as np\n", + "\n", + "# Make sure model is on correct device\n", + "DEVICE = next(model.parameters()).device\n", + "\n", + "def predict_single_variant(\n", + " dna_sequence,\n", + " gnomad_af=0.0,\n", + " conservation_scores=None,\n", + "):\n", + " \"\"\"\n", + " Clinical-style single variant inference.\n", + "\n", + " dna_sequence: ~60–1000bp centered on variant\n", + " gnomad_af: allele frequency (0–1)\n", + " conservation_scores: list in same order as feature_cols\n", + " [GERP++, GERP91, phyloP100, phyloP470, phyloP17,\n", + " phastCons100, phastCons470, phastCons17]\n", + " (optional)\n", + " \"\"\"\n", + "\n", + " # --------------------------\n", + " # 1. Validate DNA\n", + " # --------------------------\n", + " dna_sequence = dna_sequence.upper().strip()\n", + " if len(dna_sequence) < 30:\n", + " raise ValueError(\"DNA sequence too short — need ≥30bp context\")\n", + "\n", + " # --------------------------\n", + " # 2. Tokenize DNA\n", + " # --------------------------\n", + " inputs = tokenizer(\n", + " dna_sequence,\n", + " return_tensors=\"pt\",\n", + " padding=\"max_length\",\n", + " truncation=True,\n", + " max_length=tokenizer.model_max_length,\n", + " )\n", + "\n", + " input_ids = inputs[\"input_ids\"].to(DEVICE)\n", + " mask = inputs[\"attention_mask\"].to(DEVICE)\n", + "\n", + " # --------------------------\n", + " # 3. Prepare tabular features\n", + " # --------------------------\n", + " # If user does not provide scores → assume dataset mean (0 after normalization)\n", + " if conservation_scores is None:\n", + " feats = np.zeros(len(feature_cols), dtype=np.float32)\n", + " else:\n", + " feats = np.array(conservation_scores, dtype=np.float32)\n", + "\n", + " # Inject gnomAD if available\n", + " for i, c in enumerate(feature_cols):\n", + " if \"gnomad\" in c.lower():\n", + " feats[i] = (gnomad_af - feat_means[i]) / feat_stds[i]\n", + "\n", + " # Normalize all other features\n", + " feats = (feats - feat_means) / feat_stds\n", + "\n", + " # Convert safely to tensor\n", + " features_tensor = torch.from_numpy(feats.astype(np.float32)).unsqueeze(0).to(DEVICE)\n", + "\n", + " # --------------------------\n", + " # 4. Run model\n", + " # --------------------------\n", + " model.eval()\n", + " with torch.no_grad():\n", + " out = model(\n", + " input_ids=input_ids,\n", + " attention_mask=mask,\n", + " features=features_tensor,\n", + " )\n", + " prob = torch.softmax(out.logits, dim=1)[0, 1].item()\n", + "\n", + " # --------------------------\n", + " # 5. Format clinical output\n", + " # --------------------------\n", + " if prob >= 0.9:\n", + " risk = \"🔴 HIGHLY PATHOGENIC\"\n", + " elif prob >= 0.7:\n", + " risk = \"🟠 LIKELY PATHOGENIC\"\n", + " elif prob >= 0.3:\n", + " risk = \"🟡 UNCERTAIN\"\n", + " else:\n", + " risk = \"🟢 BENIGN\"\n", + "\n", + " print(\"\\n\" + \"=\"*50)\n", + " print(\"🧬 PATHOPRETER — VARIANT REPORT\")\n", + " print(\"=\"*50)\n", + " print(f\"Sequence (snippet): {dna_sequence[:12]}...{dna_sequence[-12:]}\")\n", + " print(f\"gnomAD AF: {gnomad_af:.3e}\")\n", + " print(f\"Prediction: {risk}\")\n", + " print(f\"Pathogenic prob: {prob:.4f}\")\n", + " print(\"=\"*50 + \"\\n\")\n", + "\n", + " return prob\n", + "\n", + "\n", + "# ============================\n", + "# 🧪 Example tests\n", + "# ============================\n", + "\n", + "# Fake benign DNA (nonsense)\n", + "seq_benign = \"A\" * 80\n", + "predict_single_variant(seq_benign, gnomad_af=0.45)\n", + "\n", + "# Realistic coding DNA-like context (toy example)\n", + "seq_realistic = \"TTGCTGACCTGCTGCTGAGGCTGACTTCAGGTGCTGATGGCTGAGGCTGACT\"\n", + "predict_single_variant(seq_realistic, gnomad_af=0.00001)\n", + "\n" + ] + }, + { + "cell_type": "code", + "execution_count": 44, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "⚠️ 'gene' column missing. Extracting from variant_id...\n", + "\n", + "================================================================================\n", + "From 14000 balanced (1:1 path:benign) test samples\n", + "================================================================================\n", + "\n", + "================================================================================\n", + "💀 WORST FALSE NEGATIVES (Dangerous Misses): 260 cases\n", + "The patient is SICK, but model said SAFE with high confidence.\n", + "================================================================================\n", + " gene variant_id PathoPreter_Prob gnomad_af\n", + "Unknown 19_3532086_C_A 0.001082 0.0\n", + "Unknown 6_116120454_A_T 0.001389 0.0\n", + "Unknown 2_47800542_G_T 0.001410 0.0\n", + "Unknown 19_3532086_C_G 0.001598 0.0\n", + "Unknown 17_80110949_C_G 0.002051 0.0\n", + "\n", + "================================================================================\n", + "🚨 WORST FALSE POSITIVES (False Alarms): 769 cases\n", + "The patient is HEALTHY, but model said SICK with high confidence.\n", + "================================================================================\n", + " gene variant_id PathoPreter_Prob gnomad_af\n", + "Unknown 10_110780974_A_T 0.999030 0.000000\n", + "Unknown 18_45038505_A_G 0.998675 0.000010\n", + "Unknown 18_2923825_G_A 0.997155 0.000003\n", + "Unknown 10_106706548_G_A 0.997110 0.000205\n", + "Unknown 5_88728563_G_A 0.996972 0.000000\n", + "\n", + "💡 INSIGHT:\n", + "- If FN are mostly in 'Low-Freq' range, it's likely Recessive Carrier issues.\n", + "- If FP have high gnomAD, check if they are risk factors rather than direct causes.\n" + ] + } + ], + "source": [ + "# ==========================================\n", + "# 🕵️‍♂️ CELL 8: THE AUTOPSY (Error Analysis) — FIXED\n", + "# ==========================================\n", + "import pandas as pd\n", + "\n", + "# 1. Reload Data & Predictions\n", + "df = pd.read_parquet(\"test_enriched_SEQUENCES.parquet\").reset_index(drop=True)\n", + "df[\"PathoPreter_Prob\"] = probs # From previous prediction cell\n", + "df[\"y_true\"] = y_test # From previous prediction cell\n", + "\n", + "# 2. Smart Column Detection\n", + "# If 'gene' is missing, try to find a similar column or extract from variant_id\n", + "available_cols = df.columns.tolist()\n", + "gene_col = \"gene\"\n", + "\n", + "if \"gene\" not in available_cols:\n", + " # Try common alternatives\n", + " for alt in [\"gene_symbol\", \"symbol\", \"gene_name\", \"genename\"]:\n", + " if alt in available_cols:\n", + " gene_col = alt\n", + " break\n", + " else:\n", + " # Fallback: Extract gene from variant_id if it looks like \"NM_...(GENE):...\"\n", + " print(\"⚠️ 'gene' column missing. Extracting from variant_id...\")\n", + " df[\"gene\"] = df[\"variant_id\"].apply(lambda x: x.split(\"(\")[1].split(\")\")[0] if \"(\" in x and \")\" in x else \"Unknown\")\n", + " gene_col = \"gene\"\n", + "\n", + "# 3. Define Columns to Display (Only use what actually exists)\n", + "cols_to_show = [gene_col, \"variant_id\", \"PathoPreter_Prob\"]\n", + "if \"gnomad_af\" in available_cols:\n", + " cols_to_show.append(\"gnomad_af\")\n", + "\n", + "# 4. Identify Failures\n", + "# False Negatives: Actual=1, Model said 0 (Prob < 0.2 for \"Confident Miss\")\n", + "fn_mask = (df[\"y_true\"] == 1) & (df[\"PathoPreter_Prob\"] < 0.2)\n", + "\n", + "# False Positives: Actual=0, Model said 1 (Prob > 0.8 for \"Confident Alarm\")\n", + "fp_mask = (df[\"y_true\"] == 0) & (df[\"PathoPreter_Prob\"] > 0.8)\n", + "\n", + "print(\"\\n\" + \"=\"*80)\n", + "print(f\"From 14000 balanced (1:1 path:benign) test samples\")\n", + "print(\"=\"*80)\n", + "# 5. Display WORST False Negatives (Dangerous Misses)\n", + "print(\"\\n\" + \"=\"*80)\n", + "print(f\"💀 WORST FALSE NEGATIVES (Dangerous Misses): {fn_mask.sum()} cases\")\n", + "print(\"The patient is SICK, but model said SAFE with high confidence.\")\n", + "print(\"=\"*80)\n", + "\n", + "if fn_mask.sum() > 0:\n", + " # Sort by lowest probability (most confident wrong answer)\n", + " worst_fn = df[fn_mask].sort_values(\"PathoPreter_Prob\").head(5)\n", + " print(worst_fn[cols_to_show].to_string(index=False))\n", + "else:\n", + " print(\"✅ Amazing! No confident False Negatives found.\")\n", + "\n", + "# 6. Display WORST False Positives (False Alarms)\n", + "print(\"\\n\" + \"=\"*80)\n", + "print(f\"🚨 WORST FALSE POSITIVES (False Alarms): {fp_mask.sum()} cases\")\n", + "print(\"The patient is HEALTHY, but model said SICK with high confidence.\")\n", + "print(\"=\"*80)\n", + "\n", + "if fp_mask.sum() > 0:\n", + " # Sort by highest probability (most confident wrong answer)\n", + " worst_fp = df[fp_mask].sort_values(\"PathoPreter_Prob\", ascending=False).head(5)\n", + " print(worst_fp[cols_to_show].to_string(index=False))\n", + "else:\n", + " print(\"✅ Amazing! No confident False Positives found.\")\n", + "\n", + "print(\"\\n💡 INSIGHT:\")\n", + "print(\"- If FN are mostly in 'Low-Freq' range, it's likely Recessive Carrier issues.\")\n", + "print(\"- If FP have high gnomAD, check if they are risk factors rather than direct causes.\")" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.10.11" + } + }, + "nbformat": 4, + "nbformat_minor": 2 +}