diff --git "a/train/binding_affinity_model_clean.ipynb" "b/train/binding_affinity_model_clean.ipynb"
new file mode 100644--- /dev/null
+++ "b/train/binding_affinity_model_clean.ipynb"
@@ -0,0 +1,1494 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 20,
+   "metadata": {
+    "metadata": {}
+   },
+   "outputs": [],
+   "source": [
+    "import numpy as np\n",
+    "from torch.utils.data import Dataset, DataLoader\n",
+    "from sklearn.model_selection import train_test_split\n",
+    "import torch\n",
+    "import pandas as pd\n",
+    "import torch.nn as nn\n",
+    "\n",
+    "%matplotlib inline\n",
+    "\n",
+    "class BindingDataset(Dataset):\n",
+    "    def __init__(self, dataset):\n",
+    "        self.dataset = dataset\n",
+    "        \n",
+    "    def __len__(self):\n",
+    "        return len(self.dataset)\n",
+    "    \n",
+    "    def __getitem__(self, idx):\n",
+    "        item = self.dataset[idx]\n",
+    "        return {\n",
+    "            'esm_embedding': torch.tensor(item['esm_embedding'], dtype=torch.float32),\n",
+    "            'smiles_embedding': torch.tensor(item['smiles_embedding'], dtype=torch.float32),\n",
+    "            'affinity': torch.tensor(item['affinity'], dtype=torch.float32)\n",
+    "        }\n",
+    "class AffinityDataSplitter:\n",
+    "    def __init__(self, dataset, val_size=0.15, random_state=42):\n",
+    "        self.dataset = dataset\n",
+    "        self.val_size = val_size\n",
+    "        self.random_state = random_state\n",
+    "        \n",
+    "    def calculate_affinity_bins(self, n_bins=10):\n",
+    "        \"\"\"Create balanced bins based on affinity distribution\"\"\"\n",
+    "        affinities = np.array(self.dataset['affinity'])\n",
+    "        return pd.qcut(affinities, q=n_bins, labels=False)\n",
+    "    \n",
+    "    def stratified_split(self):\n",
+    "        \"\"\"Perform stratified split considering affinity distribution\"\"\"\n",
+    "        # Get affinity bins for stratification\n",
+    "        affinity_bins = self.calculate_affinity_bins()\n",
+    "        \n",
+    "        # Perform stratified split\n",
+    "        train_idx, val_idx = train_test_split(\n",
+    "            np.arange(len(self.dataset)),\n",
+    "            test_size=self.val_size,\n",
+    "            stratify=affinity_bins,\n",
+    "            random_state=self.random_state\n",
+    "        )\n",
+    "        \n",
+    "        train_dataset = self.dataset.select(train_idx.tolist())\n",
+    "        val_dataset = self.dataset.select(val_idx.tolist())\n",
+    "        \n",
+    "        # Verify distribution\n",
+    "        self._verify_split_distribution(train_dataset, val_dataset)\n",
+    "        \n",
+    "        return train_dataset, val_dataset\n",
+    "    \n",
+    "    def _verify_split_distribution(self, train, val):\n",
+    "        \"\"\"Verify that the splits maintain similar distributions\"\"\"\n",
+    "        print(\"Distribution Statistics:\")\n",
+    "        \n",
+    "        # Affinity distribution - convert to numpy arrays first\n",
+    "        train_affinities = np.array(list(train['affinity']))\n",
+    "        val_affinities = np.array(list(val['affinity']))\n",
+    "        \n",
+    "        print(\"\\nAffinity Distribution:\")\n",
+    "        print(f\"Train: mean={np.mean(train_affinities):.2f}, std={np.std(train_affinities):.2f}\")\n",
+    "        print(f\"Val:   mean={np.mean(val_affinities):.2f}, std={np.std(val_affinities):.2f}\")\n",
+    "        \n",
+    "        # Sequence length distribution\n",
+    "        print(\"\\nSequence Length Distribution:\")\n",
+    "        train_lengths = [len(seq) for seq in train['sequence']]\n",
+    "        val_lengths = [len(seq) for seq in val['sequence']]\n",
+    "        \n",
+    "        print(f\"Train: mean={np.mean(train_lengths):.1f}, std={np.std(train_lengths):.1f}\")\n",
+    "        print(f\"Val:   mean={np.mean(val_lengths):.1f}, std={np.std(val_lengths):.1f}\")\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 21,
+   "metadata": {
+    "metadata": {}
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Distribution Statistics:\n",
+      "\n",
+      "Affinity Distribution:\n",
+      "Train: mean=5.97, std=1.57\n",
+      "Val:   mean=5.96, std=1.54\n",
+      "\n",
+      "Sequence Length Distribution:\n",
+      "Train: mean=232.9, std=154.8\n",
+      "Val:   mean=218.4, std=132.4\n"
+     ]
+    }
+   ],
+   "source": [
+    "from datasets import load_from_disk\n",
+    "dataset = load_from_disk('/scratch/pranamlab/sophtang/home/scoring/data/binding')\n",
+    "splitter = AffinityDataSplitter(\n",
+    "    dataset=dataset,\n",
+    "    val_size=0.2,\n",
+    "    random_state=42\n",
+    ")\n",
+    "\n",
+    "# Perform the split\n",
+    "train_dataset, val_dataset = splitter.stratified_split()\n",
+    "\n",
+    "# Create DataLoader for each split\n",
+    "train_loader = DataLoader(BindingDataset(train_dataset), batch_size=32, shuffle=True)\n",
+    "val_loader = DataLoader(BindingDataset(val_dataset), batch_size=32)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 22,
+   "metadata": {
+    "metadata": {}
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Affinity Statistics:\n",
+      "Min: 2.04\n",
+      "Max: 13.77\n",
+      "Mean: 5.97\n",
+      "Median: 5.76\n",
+      "\n",
+      "Unique affinity measures: Counter({'Kd=5uM': 14, 'Kd=1.5uM': 12, 'Kd=10uM': 12, 'Kd=1.9uM': 12, 'Kd=1.8uM': 12, 'Kd=1.3uM': 10, 'Kd=15uM': 10, 'Kd=2.2uM': 10, 'Kd=8uM': 9, 'Kd=0.3uM': 9, 'Kd=1.1uM': 9, 'Kd=0.7uM': 9, 'Kd=0.2uM': 8, 'Kd=1.4uM': 8, 'Kd=20uM': 8, 'Kd=1.2uM': 8, 'Kd=3.7uM': 8, 'Kd=0.5uM': 7, 'Kd=0.32uM': 7, 'Kd=29.6uM': 7, 'Kd=0.55uM': 7, 'IC50>500uM': 7, 'Kd=10nM': 6, 'Kd=0.4uM': 6, 'Kd=0.27uM': 6, 'Kd=40uM': 6, 'Kd=1uM': 6, 'Kd=3.4uM': 6, 'Kd=5.2uM': 6, 'Kd=22uM': 6, 'Kd=7uM': 6, 'Kd=100uM': 6, 'Kd=2.3uM': 5, 'Kd=2.7uM': 5, 'Kd=12nM': 5, 'Kd=20nM': 5, 'Kd=50uM': 5, 'Kd=40nM': 5, 'Kd=2.9uM': 5, 'Kd=2.4uM': 5, 'Kd=1.6uM': 5, 'Kd=0.54uM': 5, 'Kd=2.0uM': 5, 'Kd=29uM': 5, 'Kd=1.7uM': 5, 'Kd=11uM': 5, 'Kd=0.8uM': 5, 'Kd=0.1uM': 5, 'Kd=6uM': 5, 'Kd=70uM': 5, 'Kd=17uM': 5, 'Kd=13uM': 4, 'Kd=2nM': 4, 'Kd=13nM': 4, 'Kd=23uM': 4, 'Kd=150nM': 4, 'Kd=0.34uM': 4, 'Kd=2.3nM': 4, 'Kd=5.8uM': 4, 'Kd=7.5uM': 4, 'Kd=0.18uM': 4, 'Kd=4.9uM': 4, 'Kd=46uM': 4, 'Kd=19uM': 4, 'Kd=2.1uM': 4, 'Kd=3uM': 4, 'Kd=2uM': 4, 'Kd=280nM': 4, 'Kd=50nM': 4, 'Kd=44uM': 4, 'Kd=6.8uM': 4, 'Ki=60nM': 4, 'Kd=9uM': 4, 'Ki=1nM': 4, 'Kd=45uM': 4, 'Kd=3.3uM': 4, 'Kd=18uM': 4, 'Kd=0.24uM': 4, 'Kd=0.19uM': 4, 'Kd=0.15uM': 4, 'Kd=5.6uM': 4, 'Ki=50nM': 4, 'Kd=5.7uM': 4, 'Kd=24uM': 4, 'Kd=0.16uM': 4, 'Kd=6.3uM': 4, 'Kd=0.9uM': 4, 'Kd=70nM': 4, 'Kd=39nM': 3, 'Kd=65nM': 3, 'Kd=90uM': 3, 'Kd=3.1uM': 3, 'Kd=35uM': 3, 'Kd=1nM': 3, 'Kd=200nM': 3, 'Kd=113nM': 3, 'Kd=25uM': 3, 'Kd=30uM': 3, 'Ki=0.1nM': 3, 'Kd=31uM': 3, 'Kd=39uM': 3, 'Kd=120nM': 3, 'Kd=42nM': 3, 'Kd=0.36uM': 3, 'Kd=28uM': 3, 'Kd=14uM': 3, 'Kd=0.93uM': 3, 'IC50=500nM': 3, 'Kd=0.38uM': 3, 'Kd=0.07uM': 3, 'Kd=8.8uM': 3, 'Kd=170nM': 3, 'Kd=3nM': 3, 'Kd=12uM': 3, 'Kd=48uM': 3, 'Kd=0.47uM': 3, 'Kd=10.2uM': 3, 'Kd=4uM': 3, 'Kd=4.32uM': 3, 'Kd=2.5uM': 3, 'Kd=94uM': 3, 'Kd=9.3uM': 3, 'Ki=40nM': 3, 'Kd=0.75uM': 3, 'Kd=7.3uM': 3, 'Kd=60uM': 3, 'Kd=16uM': 3, 'Kd=0.43uM': 3, 'Ki=1.3uM': 3, 'Kd=36uM': 3, 'Kd=55nM': 3, 'Kd=3.0uM': 3, 'Kd=7.9uM': 3, 'Kd=4.7uM': 3, 'Kd=7.1uM': 3, 'Kd=7.4uM': 3, 'Kd=4.8uM': 3, 'Kd=0.94uM': 3, 'Kd=3.9uM': 3, 'Kd=16.5uM': 3, 'Kd=0.23uM': 3, 'Kd=6.4uM': 3, 'Kd>500uM': 3, 'Kd=1.0uM': 3, 'Kd=15.6nM': 2, 'Ki=2.3nM': 2, 'Ki=60uM': 2, 'Kd=21uM': 2, 'Kd=0.12uM': 2, 'Kd=0.35uM': 2, 'Kd=34nM': 2, 'Kd=33uM': 2, 'Kd=0.11nM': 2, 'Kd=1.1nM': 2, 'IC50=5nM': 2, 'Kd=0.33uM': 2, 'Kd=6.6uM': 2, 'Kd=47uM': 2, 'Kd=6.5uM': 2, 'Kd=0.99uM': 2, 'Kd=2.4nM': 2, 'Ki=11nM': 2, 'Kd=1pM': 2, 'Kd=0.9nM': 2, 'Kd=0.49uM': 2, 'Kd=5.9uM': 2, 'Kd=4nM': 2, 'Kd=27uM': 2, 'Kd=2.1nM': 2, 'Kd=12.9uM': 2, 'IC50=50nM': 2, 'Kd=52nM': 2, 'Kd=7nM': 2, 'Kd=3.3nM': 2, 'Kd=0.59uM': 2, 'IC50=36nM': 2, 'Kd=5nM': 2, 'Kd=49uM': 2, 'Kd=2.75uM': 2, 'Kd=3.8uM': 2, 'Kd=1.4nM': 2, 'Kd=5.4uM': 2, 'IC50=1.6uM': 2, 'IC50=58nM': 2, 'Kd=66nM': 2, 'Kd=27nM': 2, 'IC50=25nM': 2, 'Kd=8.9nM': 2, 'Kd=3.7nM': 2, 'Kd=0.67uM': 2, 'Kd=59uM': 2, 'Kd=8.0uM': 2, 'Kd=5.78uM': 2, 'Kd=0.44uM': 2, 'IC50=49uM': 2, 'Kd=0.29uM': 2, 'Kd=0.45uM': 2, 'Kd=32nM': 2, 'Kd=3.5uM': 2, 'Ki=6.4nM': 2, 'Kd=2.67uM': 2, 'Kd=290uM': 2, 'Kd=110uM': 2, 'Kd=23.9uM': 2, 'Kd=20.4uM': 2, 'Kd=7.8uM': 2, 'Kd=5.0uM': 2, 'Kd=4.3uM': 2, 'Kd=8.7uM': 2, 'Kd=19nM': 2, 'Kd=75nM': 2, 'Kd=0.6uM': 2, 'Kd=0.62uM': 2, 'Kd=4.0uM': 2, 'Ki=1.6nM': 2, 'Kd=220nM': 2, 'Kd=131nM': 2, 'Kd=16.4nM': 2, 'Kd=84uM': 2, 'Ki=0.3nM': 2, 'Kd=260nM': 2, 'Kd=0.40uM': 2, 'Ki=1100nM': 2, 'Kd=11.4uM': 2, 'Kd=53nM': 2, 'Kd=0.03uM': 2, 'IC50=0.21uM': 2, 'Kd=0.25uM': 2, 'Kd=28.2uM': 2, 'Kd=27.7uM': 2, 'Kd=9.4uM': 2, 'Kd=45nM': 2, 'Ki=7nM': 2, 'Kd=0.8mM': 2, 'Kd=0.68uM': 2, 'Kd=181nM': 2, 'Kd=9.2uM': 2, 'Kd=8.9uM': 2, 'IC50=0.54uM': 2, 'Kd=14.4uM': 2, 'Kd=1.55uM': 2, 'Kd=36nM': 2, 'IC50=26nM': 2, 'Kd=0.39uM': 2, 'Kd=4.1uM': 2, 'Kd=3.6uM': 2, 'Kd=113uM': 2, 'Kd=85nM': 2, 'Kd=0.61uM': 2, 'Kd=8.2uM': 2, 'Kd=230uM': 2, 'Kd=0.054uM': 2, 'Kd=0.31uM': 2, 'Kd=104uM': 2, 'Kd=23.6uM': 2, 'Kd=9.1uM': 2, 'Kd=0.14uM': 2, 'Kd=0.28uM': 2, 'Kd=5.1uM': 2, 'Kd=46.1uM': 2, 'Kd=370nM': 2, 'Kd=20.6uM': 2, 'Kd=5.5uM': 2, 'Kd=90nM': 2, 'Kd=5.57uM': 2, 'Kd=55uM': 2, 'Kd=2.8uM': 2, 'Kd=9.6uM': 2, 'Kd=3.2uM': 2, 'Kd=4.5uM': 2, 'Kd=10.5uM': 2, 'Kd=402uM': 2, 'Kd=4.2uM': 2, 'Kd=25.8uM': 2, 'Kd=130uM': 2, 'Kd=1.05uM': 2, 'Kd=0.22uM': 2, 'Kd=0.26uM': 2, 'Kd=10.0uM': 2, 'IC50<0.004uM': 2, 'IC50=1.5nM': 2, 'Kd=20pM': 1, 'Kd=15.4pM': 1, 'Kd=8.4pM': 1, 'Kd=10.8uM': 1, 'Ki=0.9nM': 1, 'Ki=3nM': 1, 'Ki=0.8nM': 1, 'Ki=0.13nM': 1, 'Ki=1.8nM': 1, 'Kd=155uM': 1, 'Kd=180nM': 1, 'Kd=22nM': 1, 'Kd=3pM': 1, 'Kd=277uM': 1, 'Kd=126uM': 1, 'Ki=0.12uM': 1, 'IC50=8.2uM': 1, 'Kd=6.61nM': 1, 'Kd=46.9uM': 1, 'Kd=51uM': 1, 'Kd=172uM': 1, 'Kd=150uM': 1, 'Kd=0.877uM': 1, 'Kd=10.7nM': 1, 'Kd=24.21uM': 1, 'Kd=210nM': 1, 'Kd=18nM': 1, 'Ki=20pM': 1, 'IC50=13nM': 1, 'Kd=0.98uM': 1, 'IC50=0.6nM': 1, 'Kd=4.32nM': 1, 'Ki=1.9nM': 1, 'Kd=9.2nM': 1, 'Ki=273nM': 1, 'Kd=0.03nM': 1, 'Kd=2.7nM': 1, 'Kd=201.8nM': 1, 'Kd=4.34uM': 1, 'Ki=0.2nM': 1, 'Kd=0.18nM': 1, 'Kd=5.07nM': 1, 'Kd=1.38uM': 1, 'Kd=8.7nM': 1, 'Kd=0.05uM': 1, 'Kd=350uM': 1, 'IC50=0.38nM': 1, 'Kd=123nM': 1, 'Kd=193nM': 1, 'Kd=153nM': 1, 'Kd=0.79uM': 1, 'Kd=3.75uM': 1, 'Kd=30nM': 1, 'Kd=16nM': 1, 'Ki=4.8nM': 1, 'Kd=71.4nM': 1, 'Kd=3.21nM': 1, 'Kd=13.98nM': 1, 'Kd=2115nM': 1, 'Kd=126nM': 1, 'Kd=2.0nM': 1, 'Kd=0.055uM': 1, 'IC50=6uM': 1, 'Kd=0.6nM': 1, 'Ki=98nM': 1, 'Kd=1.97uM': 1, 'Kd=14.7nM': 1, 'Kd=6.8nM': 1, 'Kd=9.58uM': 1, 'Kd=0.64uM': 1, 'Kd=1.78nM': 1, 'Kd=0.96nM': 1, 'Kd=45.7nM': 1, 'Kd=900nM': 1, 'Kd=2.63nM': 1, 'Kd=102nM': 1, 'Ki=350nM': 1, 'Kd=360nM': 1, 'Kd=8.4nM': 1, 'Kd=35nM': 1, 'Kd=55.3uM': 1, 'Kd=22.2uM': 1, 'Kd=10.1uM': 1, 'Kd=310uM': 1, 'Kd=230nM': 1, 'Kd=9.26uM': 1, 'Kd=1.56nM': 1, 'Kd=17nM': 1, 'Kd=7.9nM': 1, 'Kd=810nM': 1, 'Kd=0.989nM': 1, 'Kd=10.2nM': 1, 'Kd=40.8nM': 1, 'Kd=300nM': 1, 'IC50=6.6uM': 1, 'Kd=21nM': 1, 'IC50=0.22uM': 1, 'Kd=2.49uM': 1, 'IC50=68nM': 1, 'Kd=41.3uM': 1, 'Kd=0.046uM': 1, 'Kd=165nM': 1, 'Kd=0.141uM': 1, 'Kd=7.31nM': 1, 'Kd=0.030nM': 1, 'Kd=1540nM': 1, 'Kd=160nM': 1, 'Kd=20.3nM': 1, 'Kd=1.03uM': 1, 'Kd=99nM': 1, 'Ki=54nM': 1, 'Kd=0.741uM': 1, 'Kd=346nM': 1, 'Kd=57.5uM': 1, 'Kd=0.298uM': 1, 'Kd=1.53nM': 1, 'Kd=0.21uM': 1, 'Kd=108nM': 1, 'Kd=6.98uM': 1, 'Kd=6.9uM': 1, 'Kd=8nM': 1, 'Kd=9.9nM': 1, 'IC50=0.49uM': 1, 'IC50=66nM': 1, 'Kd=4.9nM': 1, 'Kd=3220nM': 1, 'Kd=2640nM': 1, 'Kd=1.88uM': 1, 'Kd=8.05uM': 1, 'Kd=107nM': 1, 'Kd=80nM': 1, 'Kd=20.7pM': 1, 'Kd=125nM': 1, 'Kd=15.2uM': 1, 'Kd=0.25nM': 1, 'Kd=169nM': 1, 'Kd=0.04uM': 1, 'Kd=58nM': 1, 'IC50=187nM': 1, 'Kd=1.9nM': 1, 'Kd=26nM': 1, 'Kd=2.5nM': 1, 'Kd=353nM': 1, 'Kd=422nM': 1, 'Kd=28.1nM': 1, 'Kd=7.35nM': 1, 'IC50=8.1nM': 1, 'Kd=161nM': 1, 'Kd=500pM': 1, 'Kd=3.8nM': 1, 'Kd=0.758pM': 1, 'Kd=85.7nM': 1, 'Kd=8.2nM': 1, 'Kd=0.3nM': 1, 'Kd=3.36uM': 1, 'Kd=367nM': 1, 'Kd=343nM': 1, 'Kd=4.39nM': 1, 'Kd=2.48uM': 1, 'Kd=887nM': 1, 'Ki=12nM': 1, 'IC50=39nM': 1, 'Kd=29nM': 1, 'Kd=3.5mM': 1, 'Kd=0.02uM': 1, 'Kd=0.08uM': 1, 'Kd=154uM': 1, 'Kd=0.70uM': 1, 'Kd=8.20uM': 1, 'Kd=0.83uM': 1, 'Ki=17.2uM': 1, 'IC50=90.1uM': 1, 'Kd=13.4uM': 1, 'Kd=0.69uM': 1, 'Kd=0.006uM': 1, 'Kd=91.7uM': 1, 'Kd=6.0nM': 1, 'Kd=92nM': 1, 'Kd=2000uM': 1, 'Ki=9.02mM': 1, 'Ki=871uM': 1, 'Ki=754uM': 1, 'Kd=17.0uM': 1, 'IC50=0.0876uM': 1, 'IC50=39uM': 1, 'Kd=46.5uM': 1, 'Ki=13nM': 1, 'IC50=16.0nM': 1, 'Kd=650nM': 1, 'IC50=450nM': 1, 'Kd=450uM': 1, 'Kd=19.3uM': 1, 'IC50=142nM': 1, 'IC50=4.4nM': 1, 'Kd=2.63uM': 1, 'Kd=2.22uM': 1, 'Kd=36.7uM': 1, 'Ki=2.5uM': 1, 'Ki=0.20uM': 1, 'Ki=13.6uM': 1, 'Ki=23.7uM': 1, 'Ki=0.36uM': 1, 'Ki=24uM': 1, 'IC50=0.08uM': 1, 'Kd=62.8uM': 1, 'Kd=240uM': 1, 'Kd=0.1nM': 1, 'Kd=680uM': 1, 'Kd=0.029uM': 1, 'Kd=9.5uM': 1, 'Kd=15.8uM': 1, 'Kd=137uM': 1, 'Kd=98uM': 1, 'Kd=14.2uM': 1, 'Kd=12.1uM': 1, 'Kd=7715nM': 1, 'Kd=4.4uM': 1, 'Kd=0.42uM': 1, 'Ki=420nM': 1, 'Kd=7.27uM': 1, 'Kd=71uM': 1, 'Kd=77uM': 1, 'Kd=330uM': 1, 'Kd=20.7uM': 1, 'Kd>70uM': 1, 'Kd=89nM': 1, 'Kd=21.0uM': 1, 'IC50=48.8nM': 1, 'IC50=4.6uM': 1, 'Kd=250uM': 1, 'IC50=189nM': 1, 'Kd=4.56uM': 1, 'Ki=21nM': 1, 'IC50=0.11uM': 1, 'IC50=0.7uM': 1, 'Ki=0.17uM': 1, 'Ki<21nM': 1, 'Ki=27nM': 1, 'Ki=17fM': 1, 'Ki=38fM': 1, 'Kd=602uM': 1, 'Kd=21.6uM': 1, 'Ki=260nM': 1, 'Kd=530nM': 1, 'Ki=750nM': 1, 'Kd=76nM': 1, 'Kd=250nM': 1, 'IC50=2nM': 1, 'Ki=80nM': 1, 'Ki=13uM': 1, 'Kd=8.8nM': 1, 'IC50=0.12uM': 1, 'Kd=0.041uM': 1, 'IC50=0.06mM': 1, 'Ki=0.6nM': 1, 'Ki=1.01uM': 1, 'Kd=3.57nM': 1, 'Kd=56uM': 1, 'Kd=0.046mM': 1, 'IC50=100uM': 1, 'Kd=0.60uM': 1, 'IC50=1.0uM': 1, 'Kd=1.37uM': 1, 'Kd=1.02uM': 1, 'Kd=203uM': 1, 'Ki=10.2nM': 1, 'Kd=0.0933mM': 1, 'IC50=0.15uM': 1, 'IC50=15uM': 1, 'Kd=0.118uM': 1, 'Ki=0.027nM': 1, 'Ki=2.1uM': 1, 'Ki=0.5uM': 1, 'Ki=4.4uM': 1, 'Ki=67nM': 1, 'Ki=540nM': 1, 'Ki=200nM': 1, 'Ki=25nM': 1, 'Ki<10uM': 1, 'IC50=7.2uM': 1, 'Kd=600uM': 1, 'Kd=416nM': 1, 'Ki=416nM': 1, 'Kd=872nM': 1, 'Kd=210uM': 1, 'IC50=4.9nM': 1, 'Kd=37uM': 1, 'Kd=80uM': 1, 'IC50=0.67uM': 1, 'Kd=270nM': 1, 'Kd=670nM': 1, 'Kd=125uM': 1, 'Ki=6pM': 1, 'Ki=7.7uM': 1, 'Kd=3.4mM': 1, 'IC50=53.4uM': 1, 'IC50=6.76uM': 1, 'Kd=120.8uM': 1, 'Kd=13.5uM': 1, 'IC50=0.48nM': 1, 'IC50=4.7nM': 1, 'IC50=2uM': 1, 'IC50=0.006uM': 1, 'IC50=0.1uM': 1, 'Ki=23.8nM': 1, 'Kd=3.34uM': 1, 'Ki=3.64uM': 1, 'Kd=22.5uM': 1, 'Kd>800uM': 1, 'Kd=31.9uM': 1, 'Kd=44.2uM': 1, 'Kd=16.1uM': 1, 'Kd~2uM': 1, 'Ki=0.028uM': 1, 'Ki=0.064uM': 1, 'Kd=1.5nM': 1, 'Kd=6.3nM': 1, 'Ki=3.59nM': 1, 'Kd=281.2uM': 1, 'Kd=115.1uM': 1, 'Kd=259.1uM': 1, 'Ki=56.2uM': 1, 'Ki=16.2uM': 1, 'Kd=3.86uM': 1, 'Kd=0.685uM': 1, 'IC50=0.82uM': 1, 'IC50=0.46uM': 1, 'Kd=248uM': 1, 'Kd=214uM': 1, 'IC50=0.62uM': 1, 'Kd=40.6uM': 1, 'Kd=1.77uM': 1, 'Ki=1.72uM': 1, 'Ki=17.73uM': 1, 'Ki=1.96uM': 1, 'IC50=4.9uM': 1, 'Ki=315uM': 1, 'Kd=0.51uM': 1, 'IC50=5uM': 1, 'Kd=49.5uM': 1, 'Kd=436nM': 1, 'Kd=694nM': 1, 'Kd=9.9uM': 1, 'Kd=69.91nM': 1, 'Kd=8.3uM': 1, 'Kd=0.183uM': 1, 'Kd=17.06uM': 1, 'Kd=34.25uM': 1, 'Ki=38.7uM': 1, 'Kd=2400nM': 1, 'Kd=30.3uM': 1, 'Ki=92000nM': 1, 'Ki=220nM': 1, 'Kd=121nM': 1, 'Kd=1.19uM': 1, 'Kd=10.35uM': 1, 'Kd=1.14uM': 1, 'Kd=127.3nM': 1, 'Kd=1.21uM': 1, 'Ki=3.6uM': 1, 'Ki=1000nM': 1, 'Kd=49.3uM': 1, 'Kd=288nM': 1, 'Kd=0.71uM': 1, 'Kd=324nM': 1, 'Kd=25.1uM': 1, 'Kd=4.6uM': 1, 'IC50=1.97uM': 1, 'Kd=5.04uM': 1, 'Kd=0.134uM': 1, 'Kd=0.76uM': 1, 'Kd=59.4uM': 1, 'Ki=671uM': 1, 'IC50=301.3uM': 1, 'Kd=33.6uM': 1, 'Kd=64uM': 1, 'IC50=7.7nM': 1, 'Kd=17.3uM': 1, 'Ki=0.009uM': 1, 'IC50=0.69uM': 1, 'Ki=298nM': 1, 'Kd=0.083nM': 1, 'Kd=30.2uM': 1, 'Kd=22.4uM': 1, 'IC50=120uM': 1, 'Kd=3.5nM': 1, 'Kd=221uM': 1, 'IC50=16.0uM': 1, 'IC50=10.0uM': 1, 'IC50=105.6uM': 1, 'Kd=0.290uM': 1, 'IC50=0.004uM': 1, 'Kd=4.93uM': 1, 'Kd=150.6uM': 1, 'Kd=29.58uM': 1, 'IC50=390nM': 1, 'IC50=1200nM': 1, 'Kd=189nM': 1, 'Kd=1.64uM': 1, 'Kd=320uM': 1, 'Kd=0.094uM': 1, 'Kd=68.9uM': 1, 'Kd=161.5uM': 1, 'Ki=113uM': 1, 'Kd=1590uM': 1, 'Kd=93uM': 1, 'Kd=675uM': 1, 'Kd=72uM': 1, 'Kd=280uM': 1, 'Kd=0.5mM': 1, 'IC50=12uM': 1, 'IC50=8.4uM': 1, 'Kd=35.3uM': 1, 'Kd=33nM': 1, 'Kd=67nM': 1, 'Ki=19.4uM': 1, 'Kd=0.57nM': 1, 'Kd=26.9uM': 1, 'Kd=8.1uM': 1, 'Ki=20nM': 1, 'Kd=0.024uM': 1, 'Kd=324uM': 1, 'IC50=15.44uM': 1, 'Kd=1.59uM': 1, 'IC50=9.786uM': 1, 'Kd=1.72uM': 1, 'Kd=27.5uM': 1, 'Kd=15.55uM': 1, 'Kd=135uM': 1, 'Kd=104.5nM': 1, 'IC50=1.4uM': 1, 'IC50=2.3uM': 1, 'IC50=0.933nM': 1, 'Kd=175nM': 1, 'Kd=0.043nM': 1, 'Ki=270nM': 1, 'Kd=29.8nM': 1, 'Kd=31nM': 1, 'Kd=2.6uM': 1, 'Kd=89uM': 1, 'Kd=9.24uM': 1, 'Kd=26.3uM': 1, 'Kd=0.043uM': 1, 'Ki=75nM': 1, 'Kd=19.5uM': 1, 'Ki=12.5uM': 1, 'Ki=3.2nM': 1, 'IC50=9nM': 1, 'Kd=2.9nM': 1, 'Kd=121.5uM': 1, 'Kd=3.21uM': 1, 'Kd~1mM': 1, 'Ki=1.10uM': 1, 'Kd=5.1nM': 1, 'Kd=5.2nM': 1, 'Ki=5uM': 1, 'Ki=3.39nM': 1, 'Kd=14.5uM': 1, 'Kd=11.2uM': 1, 'Ki=67uM': 1, 'IC50=0.33uM': 1, 'Kd=75.1nM': 1, 'Kd=664nM': 1, 'Ki=33uM': 1, 'Ki=207uM': 1, 'Kd=131.6nM': 1, 'Kd=671nM': 1, 'Kd=6.337nM': 1, 'Kd=7.6uM': 1, 'Kd=30.4uM': 1, 'Kd=4.6nM': 1, 'Kd=16.81uM': 1, 'Kd=0.58uM': 1, 'Kd=1.29uM': 1, 'Kd=66.2uM': 1, 'Kd=21.3uM': 1, 'Kd=2.23uM': 1, 'Kd=0.393uM': 1, 'Kd=95uM': 1, 'Kd=174uM': 1, 'Kd=55.6uM': 1, 'Kd=158uM': 1, 'Kd=798nM': 1, 'IC50=3.2uM': 1, 'Ki=3.63uM': 1, 'IC50=0.0094uM': 1, 'IC50=0.021uM': 1, 'IC50=0.0695uM': 1, 'Kd=10.7uM': 1, 'Kd=0.91uM': 1, 'Kd=19.7uM': 1, 'Kd=29.2uM': 1, 'Kd=1.07uM': 1, 'Kd=7.78uM': 1, 'Kd=68.6uM': 1, 'IC50=79.4uM': 1, 'Kd=50.5uM': 1, 'IC50=0.3uM': 1, 'Ki=320nM': 1, 'Ki=30.1nM': 1, 'Ki=2265nM': 1, 'Kd=24.6uM': 1, 'Kd=32.2uM': 1, 'Kd=318uM': 1, 'Kd=2.12uM': 1, 'Kd=124uM': 1, 'Ki=0.24nM': 1, 'Ki=66nM': 1, 'Ki<1nM': 1, 'Ki=240nM': 1, 'Ki=2000nM': 1, 'IC50=1.1nM': 1, 'Kd=74uM': 1, 'Kd=30.0nM': 1, 'Kd=190uM': 1, 'Ki=2430nM': 1, 'Kd=7200nM': 1, 'Kd=57nM': 1, 'Kd=420nM': 1, 'Kd=25nM': 1, 'Kd=0.335uM': 1, 'Kd=3.58uM': 1, 'Kd=38.4uM': 1, 'Kd=0.56uM': 1, 'Ki=3.48nM': 1, 'Kd=229nM': 1, 'Kd=15.8nM': 1, 'IC50=410nM': 1, 'Kd=7.0uM': 1, 'Kd=18.9uM': 1, 'Kd=29.1uM': 1, 'Kd=16.3uM': 1, 'Kd=58.2uM': 1, 'Kd=53.5uM': 1, 'Ki=0.23uM': 1, 'Kd=61uM': 1, 'Ki=2.7uM': 1, 'IC50=52uM': 1, 'Kd=53.8nM': 1, 'Kd=7.95uM': 1, 'Kd=53uM': 1, 'Kd=13.7uM': 1, 'IC50=19uM': 1, 'IC50=61.2uM': 1, 'IC50=10uM': 1, 'IC50=35.4uM': 1, 'IC50=17.2uM': 1, 'Kd=6.71uM': 1, 'Kd=13.3uM': 1, 'Kd=2.84uM': 1, 'Kd=4.99uM': 1, 'Kd=193uM': 1, 'IC50=158nM': 1, 'Ki=0.14nM': 1, 'Ki=1.2nM': 1, 'Ki=2.0nM': 1, 'Ki=0.051nM': 1, 'Kd=390uM': 1, 'Kd=3.15uM': 1, 'Kd=2.96uM': 1, 'Ki=0.7nM': 1, 'Ki=15.3nM': 1, 'Ki=246nM': 1, 'Kd=18.8uM': 1, 'Kd=7.8nM': 1, 'Kd=45.2nM': 1, 'Kd=4.3nM': 1, 'Kd=42.5nM': 1, 'Kd=6.6nM': 1, 'Kd=9.1nM': 1, 'Kd=3.29uM': 1, 'Ki=110nM': 1, 'IC50=71uM': 1, 'Kd=8.86uM': 1, 'Kd=300uM': 1, 'Kd=168uM': 1, 'Kd=8.6uM': 1, 'Ki=3.12uM': 1, 'Ki=3.80uM': 1, 'Kd=1.8mM': 1, 'Kd=920nM': 1, 'IC50=0.148uM': 1, 'Kd=400nM': 1, 'Kd=310nM': 1, 'Kd=2.33uM': 1, 'Kd=0.46uM': 1, 'IC50=4.5uM': 1, 'Kd=45.8nM': 1, 'Kd=0.16mM': 1, 'IC50=0.01uM': 1, 'IC50=0.039uM': 1, 'Kd=43nM': 1, 'Kd=0.86uM': 1, 'IC50=0.5uM': 1, 'IC50=0.9uM': 1, 'Kd=36.3uM': 1, 'Kd=96.7uM': 1, 'Kd=22.1uM': 1, 'Kd=1.0nM': 1, 'IC50=84uM': 1, 'Ki=250uM': 1, 'Kd=32.6nM': 1, 'Kd=0.88nM': 1, 'Kd=78.1uM': 1, 'Kd=60nM': 1, 'IC50=0.27uM': 1, 'IC50=7.9uM': 1, 'Kd=180uM': 1, 'Kd=192uM': 1, 'Ki=1.9uM': 1, 'Kd=1.06uM': 1, 'Kd=0.92uM': 1, 'IC50=0.37nM': 1, 'Kd=0.78uM': 1, 'Kd=33.4uM': 1, 'Kd=1.3mM': 1, 'Kd=25.4uM': 1, 'Kd=245uM': 1, 'Kd=1.2mM': 1, 'Kd=2.4mM': 1, 'Kd=276nM': 1, 'Kd=57uM': 1, 'Kd=108.4uM': 1, 'Kd~1.8uM': 1, 'Kd<0.1nM': 1, 'Kd=1.6nM': 1, 'Kd=1.40uM': 1, 'Kd=86.5uM': 1, 'Kd=30.5nM': 1, 'Ki=1.3nM': 1, 'Ki=274uM': 1, 'Kd=5.41uM': 1, 'Kd=770uM': 1, 'Kd=365nM': 1, 'Kd=1079nM': 1, 'Kd=9.63uM': 1, 'IC50<=3.5nM': 1, 'Kd=1.09uM': 1, 'Kd=159nM': 1, 'Kd=140.5uM': 1, 'Kd=635uM': 1, 'Kd=59.5uM': 1, 'Kd=19.4uM': 1, 'Kd=0.53uM': 1, 'Ki=500nM': 1, 'Kd=288uM': 1, 'Kd=109.4uM': 1, 'IC50=910uM': 1, 'Kd=67.4uM': 1, 'Kd=64.7uM': 1, 'Kd=42.2nM': 1, 'Kd=73nM': 1, 'Kd=1.23uM': 1, 'Kd=102uM': 1, 'Kd=56nM': 1, 'IC50=6nM': 1, 'Ki=271nM': 1, 'Kd=1051uM': 1, 'Kd=218uM': 1, 'Kd=16.7uM': 1, 'Kd=0.17uM': 1, 'Ki=1.7nM': 1, 'Kd=93.8uM': 1, 'Kd=9.17uM': 1, 'Kd=251uM': 1, 'Kd=24.5uM': 1, 'Kd~10nM': 1, 'Kd=0.025uM': 1, 'Kd=2.40uM': 1, 'Kd=13.4nM': 1, 'IC50=31uM': 1, 'Kd=0.37uM': 1, 'Kd=1.73uM': 1, 'IC50=435uM': 1, 'IC50=39.7uM': 1, 'IC50=85.2uM': 1, 'IC50=295.0uM': 1, 'IC50=65.8uM': 1, 'IC50=65nM': 1, 'IC50=5.7uM': 1, 'Kd=0.88uM': 1, 'Kd=1.49uM': 1, 'Kd=2.36uM': 1, 'Kd=41uM': 1, 'IC50=4.3uM': 1, 'IC50=1.7uM': 1, 'Kd=16.8uM': 1, 'Kd=1mM': 1, 'IC50=133uM': 1, 'IC50=95uM': 1, 'Kd=11.5uM': 1, 'Kd=5.5nM': 1, 'Kd=14.9uM': 1, 'Kd=6.0uM': 1, 'Kd=64.5uM': 1, 'Ki=41nM': 1, 'Kd=0.01mM': 1, 'IC50=6.8uM': 1, 'Kd=140nM': 1, 'Kd=63nM': 1, 'Kd=2200nM': 1, 'Kd=0.66uM': 1, 'Kd=0.447uM': 1, 'Kd=185uM': 1, 'Kd=0.445uM': 1, 'Kd=63.6uM': 1, 'Kd=490nM': 1, 'Kd=36.4uM': 1, 'Kd=13.2uM': 1, 'Kd=24.9uM': 1, 'Kd=38.7uM': 1, 'Kd=11.7uM': 1, 'Kd=25.1nM': 1, 'Kd=76.1uM': 1, 'Kd=235uM': 1, 'Kd=118uM': 1, 'Kd=400uM': 1, 'IC50=250nM': 1, 'IC50=110nM': 1, 'IC50=8nM': 1, 'IC50~0.49uM': 1, 'Kd=200uM': 1, 'IC50=0.71uM': 1, 'IC50=13.6uM': 1, 'Kd=380uM': 1, 'Kd=286.1nM': 1, 'Kd=384.8nM': 1, 'Ki=638.1nM': 1, 'Kd=42uM': 1, 'Kd=12.4uM': 1, 'IC50=2.6uM': 1, 'Kd=3.83uM': 1, 'Ki=2.5nM': 1, 'Ki=8.7uM': 1, 'Kd=183nM': 1, 'Kd=1.69uM': 1, 'Kd=48.7uM': 1, 'Kd=0.06uM': 1, 'Kd=157nM': 1, 'Kd=99uM': 1, 'Kd=69uM': 1, 'IC50=17nM': 1, 'Kd=139uM': 1, 'Ki=72.73uM': 1, 'Ki=41.24uM': 1, 'Ki=40.98uM': 1, 'IC50=27uM': 1, 'IC50=61uM': 1, 'Kd=2.20uM': 1, 'Kd=6.32uM': 1, 'Kd=52uM': 1, 'Kd=121uM': 1, 'Kd=19.8uM': 1, 'Kd=10.6uM': 1, 'IC50=57.2nM': 1, 'Kd=2.03uM': 1, 'Kd=0.10uM': 1, 'Kd=27.4uM': 1, 'Kd=109uM': 1, 'Kd=184uM': 1, 'Kd=23.3uM': 1, 'Kd=780nM': 1, 'Kd=124nM': 1, 'Kd=14.1uM': 1, 'Kd=441nM': 1, 'Kd=120uM': 1, 'IC50=0.36uM': 1, 'Kd=64nM': 1, 'Ki=76uM': 1, 'Ki=107uM': 1, 'Ki=6.3nM': 1, 'Ki=34nM': 1, 'Ki=4.0uM': 1, 'Kd=12.3uM': 1, 'Kd=103.0nM': 1, 'Kd=362uM': 1, 'Kd=68.0uM': 1, 'Kd=5.0nM': 1, 'Kd=0.069uM': 1, 'Kd=4.47uM': 1, 'IC50=670nM': 1, 'IC50=1000nM': 1, 'Kd>180uM': 1, 'Kd=100nM': 1, 'IC50=205nM': 1, 'IC50=14nM': 1, 'Kd=0.65uM': 1, 'Kd=17.9uM': 1, 'IC50=128nM': 1, 'Ki=0.9uM': 1, 'Kd=330nM': 1, 'Kd=1100nM': 1, 'Kd~102uM': 1, 'IC50=8uM': 1, 'Kd=326nM': 1, 'Kd=4.43uM': 1, 'Kd=155nM': 1, 'Kd=2.52uM': 1, 'Ki=0.5nM': 1, 'Ki=160nM': 1, 'Kd=54nM': 1, 'Kd=88nM': 1, 'Kd=103nM': 1, 'Kd=2762nM': 1, 'Kd=541nM': 1, 'Kd=2.26uM': 1, 'Kd=12.93uM': 1, 'Kd=0.5083mM': 1, 'Kd=115uM': 1, 'IC50=73nM': 1, 'Kd=45.6nM': 1, 'Kd=173nM': 1, 'Kd=184.3uM': 1, 'Ki=35nM': 1, 'Ki=0.54uM': 1, 'Ki=16.7pM': 1, 'Ki=977pM': 1, 'IC50=0.04uM': 1, 'IC50=0.40uM': 1, 'Kd=0.2nM': 1, 'Ki=238nM': 1, 'Ki=22.4pM': 1, 'Ki=1.17nM': 1, 'Ki=0.491nM': 1, 'Ki=1442nM': 1, 'Ki=14uM': 1, 'IC50=30nM': 1, 'IC50=4.42uM': 1, 'Kd=0.0095nM': 1, 'IC50=20.2uM': 1, 'IC50=1.9nM': 1, 'Kd=2.87uM': 1, 'Ki=33.7pM': 1, 'Kd=1500uM': 1, 'Ki=0.68nM': 1, 'Ki=1.05nM': 1, 'Ki=0.504nM': 1, 'Ki=0.618nM': 1, 'Ki=0.0538nM': 1, 'Kd=63.4nM': 1, 'Kd=0.81uM': 1, 'IC50=0.013uM': 1, 'IC50=0.06nM': 1, 'Ki=20.4nM': 1, 'IC50=400uM': 1, 'Ki=0.82nM': 1, 'Kd=37.7uM': 1, 'IC50=0.1nM': 1, 'Ki=8.27uM': 1, 'Kd=6nM': 1, 'Kd=9nM': 1, 'Kd=143.5uM': 1, 'Kd=9.8uM': 1})\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 1500x500 with 2 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "import numpy as np\n",
+    "import matplotlib.pyplot as plt\n",
+    "import seaborn as sns\n",
+    "from collections import Counter\n",
+    "\n",
+    "def analyze_binding_affinities(dataset):\n",
+    "    \"\"\"Analyze binding affinities and their measures to determine appropriate thresholds\"\"\"\n",
+    "    \n",
+    "    affinities = np.array(dataset['affinity'])\n",
+    "    measures = dataset['affinity_measure']\n",
+    "    \n",
+    "    # Print basic statistics\n",
+    "    print(\"Affinity Statistics:\")\n",
+    "    print(f\"Min: {affinities.min():.2f}\")\n",
+    "    print(f\"Max: {affinities.max():.2f}\")\n",
+    "    print(f\"Mean: {affinities.mean():.2f}\")\n",
+    "    print(f\"Median: {np.median(affinities):.2f}\")\n",
+    "    print(f\"\\nUnique affinity measures: {Counter(measures)}\")\n",
+    "    \n",
+    "    # Plot distribution\n",
+    "    plt.figure(figsize=(15, 5))\n",
+    "    \n",
+    "    # Overall distribution\n",
+    "    plt.subplot(1, 2, 1)\n",
+    "    sns.histplot(affinities, bins=50)\n",
+    "    plt.title('Distribution of Binding Affinities')\n",
+    "    plt.xlabel('Affinity Value')\n",
+    "    plt.ylabel('Count')\n",
+    "    \n",
+    "    # Box plot by measure type\n",
+    "    plt.subplot(1, 2, 2)\n",
+    "    sns.boxplot(x='affinity_measure', y='affinity', data={'affinity_measure': measures, 'affinity': affinities})\n",
+    "    plt.title('Affinity Distribution by Measure Type')\n",
+    "    plt.xticks(rotation=45)\n",
+    "    \n",
+    "    plt.tight_layout()\n",
+    "    plt.show()\n",
+    "    \n",
+    "    # Calculate measure-specific statistics\n",
+    "    measure_stats = {}\n",
+    "    for measure in set(measures):\n",
+    "        mask = np.array(measures) == measure\n",
+    "        measure_affinities = affinities[mask]\n",
+    "        measure_stats[measure] = {\n",
+    "            'count': len(measure_affinities),\n",
+    "            'mean': measure_affinities.mean(),\n",
+    "            'std': measure_affinities.std(),\n",
+    "            'median': np.median(measure_affinities),\n",
+    "            'q25': np.percentile(measure_affinities, 25),\n",
+    "            'q75': np.percentile(measure_affinities, 75)\n",
+    "        }\n",
+    "    \n",
+    "    return measure_stats\n",
+    "\n",
+    "measure_stats = analyze_binding_affinities(dataset)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 23,
+   "metadata": {
+    "metadata": {}
+   },
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 1000x600 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Min: 2.04\n",
+      "Q1: 4.71\n",
+      "Median: 5.92\n",
+      "Q3: 7.17\n",
+      "Max: 13.77\n"
+     ]
+    }
+   ],
+   "source": [
+    "import numpy as np\n",
+    "import matplotlib.pyplot as plt\n",
+    "\n",
+    "def analyze_affinity_distribution(affinities):\n",
+    "    plt.figure(figsize=(10, 6))\n",
+    "    plt.hist(affinities, bins=30)\n",
+    "    plt.axvline(x=np.percentile(affinities, 25), color='r', linestyle='--', label='Q1')\n",
+    "    plt.axvline(x=np.median(affinities), color='g', linestyle='--', label='Median')\n",
+    "    plt.axvline(x=np.percentile(affinities, 75), color='b', linestyle='--', label='Q3')\n",
+    "    plt.xlabel('-log(Affinity)')\n",
+    "    plt.ylabel('Count')\n",
+    "    plt.title('Distribution of Binding Affinities')\n",
+    "    plt.legend()\n",
+    "    plt.show()\n",
+    "    \n",
+    "    print(f\"Min: {np.min(affinities):.2f}\")\n",
+    "    print(f\"Q1: {np.percentile(affinities, 25):.2f}\")\n",
+    "    print(f\"Median: {np.median(affinities):.2f}\")\n",
+    "    print(f\"Q3: {np.percentile(affinities, 75):.2f}\")\n",
+    "    print(f\"Max: {np.max(affinities):.2f}\")\n",
+    "\n",
+    "# Get all affinity values\n",
+    "affinities = [float(stats['mean']) for stats in measure_stats.values()]\n",
+    "analyze_affinity_distribution(affinities)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 24,
+   "metadata": {
+    "metadata": {}
+   },
+   "outputs": [],
+   "source": [
+    "# Initialize model\n",
+    "device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 25,
+   "metadata": {
+    "metadata": {}
+   },
+   "outputs": [],
+   "source": [
+    "from sklearn.metrics import accuracy_score, f1_score\n",
+    "from scipy.stats import spearmanr\n",
+    "class ImprovedBindingPredictor(nn.Module):\n",
+    "    def __init__(self, \n",
+    "                 esm_dim=1280,\n",
+    "                 smiles_dim=768,\n",
+    "                 hidden_dim=512,\n",
+    "                 n_heads=8,\n",
+    "                 n_layers=3,\n",
+    "                 dropout=0.1):\n",
+    "        super().__init__()\n",
+    "        \n",
+    "        # Define binding thresholds\n",
+    "        self.tight_threshold = 7.5    # Kd/Ki/IC50 ≤ ~30nM\n",
+    "        self.weak_threshold = 6.0     # Kd/Ki/IC50 > 1μM\n",
+    "        \n",
+    "        # Project to same dimension\n",
+    "        self.smiles_projection = nn.Linear(smiles_dim, hidden_dim)\n",
+    "        self.protein_projection = nn.Linear(esm_dim, hidden_dim)\n",
+    "        self.protein_norm = nn.LayerNorm(hidden_dim)\n",
+    "        self.smiles_norm = nn.LayerNorm(hidden_dim)\n",
+    "        \n",
+    "        # Cross attention blocks with layer norm\n",
+    "        self.cross_attention_layers = nn.ModuleList([\n",
+    "            nn.ModuleDict({\n",
+    "                'attention': nn.MultiheadAttention(hidden_dim, n_heads, dropout=dropout),\n",
+    "                'norm1': nn.LayerNorm(hidden_dim),\n",
+    "                'ffn': nn.Sequential(\n",
+    "                    nn.Linear(hidden_dim, hidden_dim * 4),\n",
+    "                    nn.ReLU(),\n",
+    "                    nn.Dropout(dropout),\n",
+    "                    nn.Linear(hidden_dim * 4, hidden_dim)\n",
+    "                ),\n",
+    "                'norm2': nn.LayerNorm(hidden_dim)\n",
+    "            }) for _ in range(n_layers)\n",
+    "        ])\n",
+    "        \n",
+    "        # Prediction heads\n",
+    "        self.shared_head = nn.Sequential(\n",
+    "            nn.Linear(hidden_dim * 2, hidden_dim),\n",
+    "            nn.ReLU(),\n",
+    "            nn.Dropout(dropout),\n",
+    "        )\n",
+    "        \n",
+    "        # Regression head\n",
+    "        self.regression_head = nn.Linear(hidden_dim, 1)\n",
+    "        \n",
+    "        # Classification head (3 classes: tight, medium, loose binding)\n",
+    "        self.classification_head = nn.Linear(hidden_dim, 3)\n",
+    "        \n",
+    "    def get_binding_class(self, affinity):\n",
+    "        \"\"\"Convert affinity values to class indices\n",
+    "        0: tight binding (>= 7.5)\n",
+    "        1: medium binding (6.0-7.5)\n",
+    "        2: weak binding (< 6.0)\n",
+    "        \"\"\"\n",
+    "        if isinstance(affinity, torch.Tensor):\n",
+    "            tight_mask = affinity >= self.tight_threshold\n",
+    "            weak_mask = affinity < self.weak_threshold\n",
+    "            medium_mask = ~(tight_mask | weak_mask)\n",
+    "            \n",
+    "            classes = torch.zeros_like(affinity, dtype=torch.long)\n",
+    "            classes[medium_mask] = 1\n",
+    "            classes[weak_mask] = 2\n",
+    "            return classes\n",
+    "        else:\n",
+    "            if affinity >= self.tight_threshold:\n",
+    "                return 0  # tight binding\n",
+    "            elif affinity < self.weak_threshold:\n",
+    "                return 2  # weak binding\n",
+    "            else:\n",
+    "                return 1  # medium binding\n",
+    "        \n",
+    "    def forward(self, protein_emb, smiles_emb):\n",
+    "        protein = self.protein_norm(self.protein_projection(protein_emb))\n",
+    "        smiles = self.smiles_norm(self.smiles_projection(smiles_emb))\n",
+    "        \n",
+    "        protein = protein.transpose(0, 1)\n",
+    "        smiles = smiles.transpose(0, 1)\n",
+    "        \n",
+    "        # Cross attention layers\n",
+    "        for layer in self.cross_attention_layers:\n",
+    "            # Protein attending to SMILES\n",
+    "            attended_protein = layer['attention'](\n",
+    "                protein, smiles, smiles\n",
+    "            )[0]\n",
+    "            protein = layer['norm1'](protein + attended_protein)\n",
+    "            protein = layer['norm2'](protein + layer['ffn'](protein))\n",
+    "            \n",
+    "            # SMILES attending to protein\n",
+    "            attended_smiles = layer['attention'](\n",
+    "                smiles, protein, protein\n",
+    "            )[0]\n",
+    "            smiles = layer['norm1'](smiles + attended_smiles)\n",
+    "            smiles = layer['norm2'](smiles + layer['ffn'](smiles))\n",
+    "        \n",
+    "        # Get sequence-level representations\n",
+    "        protein_pool = torch.mean(protein, dim=0)\n",
+    "        smiles_pool = torch.mean(smiles, dim=0)\n",
+    "        \n",
+    "        # Concatenate both representations\n",
+    "        combined = torch.cat([protein_pool, smiles_pool], dim=-1)\n",
+    "        \n",
+    "        # Shared features\n",
+    "        shared_features = self.shared_head(combined)\n",
+    "        \n",
+    "        regression_output = self.regression_head(shared_features)\n",
+    "        classification_logits = self.classification_head(shared_features)\n",
+    "        \n",
+    "        return regression_output, classification_logits\n",
+    "\n",
+    "def train_epoch(model, dataloader, optimizer, criterion_reg, criterion_cls, device):\n",
+    "    model.train()\n",
+    "    total_loss = 0\n",
+    "    total_reg_loss = 0\n",
+    "    total_cls_loss = 0\n",
+    "    reg_predictions = []\n",
+    "    cls_predictions = []\n",
+    "    true_values = []\n",
+    "    cls_true = []\n",
+    "    \n",
+    "    for batch in dataloader:\n",
+    "        optimizer.zero_grad()\n",
+    "        \n",
+    "        protein_emb = batch['esm_embedding'].to(device)\n",
+    "        peptide_emb = batch['smiles_embedding'].to(device)\n",
+    "        affinity = batch['affinity'].to(device)\n",
+    "        \n",
+    "        true_classes = model.get_binding_class(affinity)\n",
+    "        \n",
+    "        reg_pred, cls_pred = model(protein_emb, peptide_emb)\n",
+    "        \n",
+    "        reg_loss = criterion_reg(reg_pred.squeeze(), affinity)\n",
+    "        cls_loss = criterion_cls(cls_pred, true_classes)\n",
+    "        \n",
+    "        # Combined loss with weighting\n",
+    "        loss = reg_loss + cls_loss\n",
+    "        \n",
+    "        loss.backward()\n",
+    "        torch.nn.utils.clip_grad_norm_(model.parameters(), max_norm=1.0)\n",
+    "        optimizer.step()\n",
+    "        \n",
+    "        total_loss += loss.item()\n",
+    "        total_reg_loss += reg_loss.item()\n",
+    "        total_cls_loss += cls_loss.item()\n",
+    "        \n",
+    "        reg_predictions.extend(reg_pred.detach().cpu().numpy())\n",
+    "        true_values.extend(affinity.cpu().numpy())\n",
+    "        cls_predictions.extend(torch.argmax(cls_pred, dim=1).cpu().numpy())\n",
+    "        cls_true.extend(true_classes.cpu().numpy())\n",
+    "    \n",
+    "\n",
+    "    avg_loss = total_loss / len(dataloader)\n",
+    "    avg_reg_loss = total_reg_loss / len(dataloader)\n",
+    "    avg_cls_loss = total_cls_loss / len(dataloader)\n",
+    "    correlation = spearmanr(reg_predictions, true_values)[0]\n",
+    "    cls_accuracy = accuracy_score(cls_true, cls_predictions)\n",
+    "    f1 = f1_score(cls_true, cls_predictions, average='weighted')\n",
+    "    \n",
+    "    # Calculate per-class metrics\n",
+    "    class_names = ['Tight', 'Medium', 'Weak']\n",
+    "    per_class_f1 = f1_score(cls_true, cls_predictions, average=None)\n",
+    "    per_class_metrics = {\n",
+    "        name: score for name, score in zip(class_names, per_class_f1)\n",
+    "    }\n",
+    "    \n",
+    "    return {\n",
+    "        'loss': avg_loss,\n",
+    "        'reg_loss': avg_reg_loss,\n",
+    "        'cls_loss': avg_cls_loss,\n",
+    "        'correlation': correlation,\n",
+    "        'accuracy': cls_accuracy,\n",
+    "        'f1_score': f1,\n",
+    "        'per_class_f1': per_class_metrics,\n",
+    "        'reg_predictions': reg_predictions,\n",
+    "        'cls_predictions': cls_predictions,\n",
+    "        'true_values': true_values,\n",
+    "        'cls_true': cls_true\n",
+    "    }\n",
+    "\n",
+    "def validate_epoch(model, dataloader, criterion_reg, criterion_cls, device):\n",
+    "    model.eval()\n",
+    "    total_loss = 0\n",
+    "    total_reg_loss = 0\n",
+    "    total_cls_loss = 0\n",
+    "    reg_predictions = []\n",
+    "    cls_predictions = []\n",
+    "    true_values = []\n",
+    "    cls_true = []\n",
+    "    \n",
+    "    with torch.no_grad():\n",
+    "        for batch in dataloader:\n",
+    "            protein_emb = batch['esm_embedding'].to(device)\n",
+    "            peptide_emb = batch['smiles_embedding'].to(device)\n",
+    "            affinity = batch['affinity'].to(device)\n",
+    "            \n",
+    "            true_classes = model.get_binding_class(affinity)\n",
+    "            \n",
+    "            reg_pred, cls_pred = model(protein_emb, peptide_emb)\n",
+    "            \n",
+    "            reg_loss = criterion_reg(reg_pred.squeeze(), affinity)\n",
+    "            cls_loss = criterion_cls(cls_pred, true_classes)\n",
+    "            loss = reg_loss + cls_loss\n",
+    "            \n",
+    "            total_loss += loss.item()\n",
+    "            total_reg_loss += reg_loss.item()\n",
+    "            total_cls_loss += cls_loss.item()\n",
+    "            \n",
+    "            reg_predictions.extend(reg_pred.cpu().numpy())\n",
+    "            true_values.extend(affinity.cpu().numpy())\n",
+    "            cls_predictions.extend(torch.argmax(cls_pred, dim=1).cpu().numpy())\n",
+    "            cls_true.extend(true_classes.cpu().numpy())\n",
+    "    \n",
+    "    avg_loss = total_loss / len(dataloader)\n",
+    "    avg_reg_loss = total_reg_loss / len(dataloader)\n",
+    "    avg_cls_loss = total_cls_loss / len(dataloader)\n",
+    "    correlation = spearmanr(reg_predictions, true_values)[0]\n",
+    "    cls_accuracy = accuracy_score(cls_true, cls_predictions)\n",
+    "    f1 = f1_score(cls_true, cls_predictions, average='weighted')\n",
+    "    \n",
+    "    class_names = ['Tight', 'Medium', 'Weak']\n",
+    "    per_class_f1 = f1_score(cls_true, cls_predictions, average=None)\n",
+    "    per_class_metrics = {\n",
+    "        name: score for name, score in zip(class_names, per_class_f1)\n",
+    "    }\n",
+    "    \n",
+    "    return {\n",
+    "        'loss': avg_loss,\n",
+    "        'reg_loss': avg_reg_loss,\n",
+    "        'cls_loss': avg_cls_loss,\n",
+    "        'correlation': correlation,\n",
+    "        'accuracy': cls_accuracy,\n",
+    "        'f1_score': f1,\n",
+    "        'per_class_f1': per_class_metrics,\n",
+    "        'reg_predictions': reg_predictions,\n",
+    "        'cls_predictions': cls_predictions,\n",
+    "        'true_values': true_values,\n",
+    "        'cls_true': cls_true\n",
+    "    }"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 26,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def plot_correlation_with_classes(reg_predictions, true_values, epoch, set_name=\"Validation\"):\n",
+    "    plt.figure(figsize=(10, 8))\n",
+    "    \n",
+    "    plt.scatter(true_values, reg_predictions, alpha=0.5)\n",
+    "    \n",
+    "    # Perfect correlation line\n",
+    "    min_val = min(min(true_values), min(reg_predictions))\n",
+    "    max_val = max(max(true_values), max(reg_predictions))\n",
+    "    plt.plot([min_val, max_val], [min_val, max_val], 'r--', label='Perfect Correlation')\n",
+    "    \n",
+    "    # Class boundary lines\n",
+    "    plt.axvline(x=7.5, color='g', linestyle='--', alpha=0.5)\n",
+    "    plt.axvline(x=6.0, color='b', linestyle='--', alpha=0.5)\n",
+    "    \n",
+    "    # Add class labels\n",
+    "    plt.text(7.7, plt.ylim()[0], 'Tight Binding\\n(≤ ~30nM)', rotation=90, verticalalignment='bottom')\n",
+    "    plt.text(6.2, plt.ylim()[0], 'Medium Binding\\n(~30nM - 1μM)', rotation=90, verticalalignment='bottom')\n",
+    "    plt.text(5.7, plt.ylim()[0], 'Weak Binding\\n(> 1μM)', rotation=90, verticalalignment='bottom')\n",
+    "    \n",
+    "    # Calculate correlation\n",
+    "    correlation = spearmanr(true_values, reg_predictions)[0]\n",
+    "    \n",
+    "    plt.xlabel('True Affinity (-log scale)')\n",
+    "    plt.ylabel('Predicted Affinity (-log scale)')\n",
+    "    plt.title(f'{set_name} Set Correlation Plot (Epoch {epoch})\\nSpearman Correlation: {correlation:.3f}')\n",
+    "    plt.grid(True, alpha=0.3)\n",
+    "    \n",
+    "    # Add counts for each class\n",
+    "    tight = sum(1 for x in true_values if x >= 7.5)\n",
+    "    medium = sum(1 for x in true_values if 6.0 <= x < 7.5)\n",
+    "    weak = sum(1 for x in true_values if x < 6.0)\n",
+    "    total = len(true_values)\n",
+    "    \n",
+    "    plt.text(0.02, 0.98, \n",
+    "             f'Class Distribution:\\nTight: {tight} ({tight/total*100:.1f}%)\\n'\n",
+    "             f'Medium: {medium} ({medium/total*100:.1f}%)\\n'\n",
+    "             f'Weak: {weak} ({weak/total*100:.1f}%)',\n",
+    "             transform=plt.gca().transAxes,\n",
+    "             verticalalignment='top',\n",
+    "             bbox=dict(boxstyle='round', facecolor='white', alpha=0.8))\n",
+    "    \n",
+    "    plt.tight_layout()\n",
+    "    plt.show()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 27,
+   "metadata": {
+    "metadata": {}
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Epoch 1/50:\n",
+      "Train - Reg Loss: 3.9224, Cls Loss: 1.0055\n",
+      "Train - Correlation: 0.1038, Accuracy: 0.5403\n",
+      "Train - Per-class F1: {'Tight': 0.0, 'Medium': 0.11218568665377177, 'Weak': 0.7003272557269752}\n",
+      "Val - Reg Loss: 1.8838, Cls Loss: 0.9746\n",
+      "Val - Correlation: 0.4910, Accuracy: 0.5429\n",
+      "Val - Per-class F1: {'Tight': 0.0, 'Medium': 0.0, 'Weak': 0.703770197486535}\n",
+      "✓ Saved best model at epoch 1 with validation correlation: 0.4910\n",
+      "✓ Saved correlation plots: best_model_val_correlation.png, best_model_train_correlation.png\n",
+      "✓ Saved correlation plots: best_model_val_correlation.png, best_model_train_correlation.png\n",
+      "Epoch 2/50:\n",
+      "Train - Reg Loss: 2.0297, Cls Loss: 0.9133\n",
+      "Train - Correlation: 0.4269, Accuracy: 0.5708\n",
+      "Train - Per-class F1: {'Tight': 0.15267175572519084, 'Medium': 0.13857677902621723, 'Weak': 0.7341650671785028}\n",
+      "Val - Reg Loss: 1.7378, Cls Loss: 0.8810\n",
+      "Val - Correlation: 0.5158, Accuracy: 0.5873\n",
+      "Val - Per-class F1: {'Tight': 0.06349206349206349, 'Medium': 0.35714285714285715, 'Weak': 0.755939524838013}\n",
+      "✓ Saved best model at epoch 2 with validation correlation: 0.5158\n",
+      "Epoch 2/50:\n",
+      "Train - Reg Loss: 2.0297, Cls Loss: 0.9133\n",
+      "Train - Correlation: 0.4269, Accuracy: 0.5708\n",
+      "Train - Per-class F1: {'Tight': 0.15267175572519084, 'Medium': 0.13857677902621723, 'Weak': 0.7341650671785028}\n",
+      "Val - Reg Loss: 1.7378, Cls Loss: 0.8810\n",
+      "Val - Correlation: 0.5158, Accuracy: 0.5873\n",
+      "Val - Per-class F1: {'Tight': 0.06349206349206349, 'Medium': 0.35714285714285715, 'Weak': 0.755939524838013}\n",
+      "✓ Saved best model at epoch 2 with validation correlation: 0.5158\n",
+      "✓ Saved correlation plots: best_model_val_correlation.png, best_model_train_correlation.png\n",
+      "✓ Saved correlation plots: best_model_val_correlation.png, best_model_train_correlation.png\n",
+      "Epoch 3/50:\n",
+      "Train - Reg Loss: 1.8158, Cls Loss: 0.8444\n",
+      "Train - Correlation: 0.4794, Accuracy: 0.5986\n",
+      "Train - Per-class F1: {'Tight': 0.37790697674418605, 'Medium': 0.2732732732732733, 'Weak': 0.7550802139037434}\n",
+      "Val - Reg Loss: 1.7881, Cls Loss: 0.8570\n",
+      "Val - Correlation: 0.5060, Accuracy: 0.5900\n",
+      "Val - Per-class F1: {'Tight': 0.2926829268292683, 'Medium': 0.25, 'Weak': 0.7541666666666667}\n",
+      "Epoch 3/50:\n",
+      "Train - Reg Loss: 1.8158, Cls Loss: 0.8444\n",
+      "Train - Correlation: 0.4794, Accuracy: 0.5986\n",
+      "Train - Per-class F1: {'Tight': 0.37790697674418605, 'Medium': 0.2732732732732733, 'Weak': 0.7550802139037434}\n",
+      "Val - Reg Loss: 1.7881, Cls Loss: 0.8570\n",
+      "Val - Correlation: 0.5060, Accuracy: 0.5900\n",
+      "Val - Per-class F1: {'Tight': 0.2926829268292683, 'Medium': 0.25, 'Weak': 0.7541666666666667}\n",
+      "Epoch 4/50:\n",
+      "Train - Reg Loss: 1.6833, Cls Loss: 0.8409\n",
+      "Train - Correlation: 0.5224, Accuracy: 0.6083\n",
+      "Train - Per-class F1: {'Tight': 0.4450402144772118, 'Medium': 0.30906389301634474, 'Weak': 0.7513631406761178}\n",
+      "Val - Reg Loss: 2.3813, Cls Loss: 0.8922\n",
+      "Val - Correlation: 0.5501, Accuracy: 0.5762\n",
+      "Val - Per-class F1: {'Tight': 0.4175824175824176, 'Medium': 0.5119453924914675, 'Weak': 0.6745562130177515}\n",
+      "✓ Saved best model at epoch 4 with validation correlation: 0.5501\n",
+      "Epoch 4/50:\n",
+      "Train - Reg Loss: 1.6833, Cls Loss: 0.8409\n",
+      "Train - Correlation: 0.5224, Accuracy: 0.6083\n",
+      "Train - Per-class F1: {'Tight': 0.4450402144772118, 'Medium': 0.30906389301634474, 'Weak': 0.7513631406761178}\n",
+      "Val - Reg Loss: 2.3813, Cls Loss: 0.8922\n",
+      "Val - Correlation: 0.5501, Accuracy: 0.5762\n",
+      "Val - Per-class F1: {'Tight': 0.4175824175824176, 'Medium': 0.5119453924914675, 'Weak': 0.6745562130177515}\n",
+      "✓ Saved best model at epoch 4 with validation correlation: 0.5501\n",
+      "✓ Saved correlation plots: best_model_val_correlation.png, best_model_train_correlation.png\n",
+      "✓ Saved correlation plots: best_model_val_correlation.png, best_model_train_correlation.png\n",
+      "Epoch 5/50:\n",
+      "Train - Reg Loss: 1.6100, Cls Loss: 0.8059\n",
+      "Train - Correlation: 0.5389, Accuracy: 0.6222\n",
+      "Train - Per-class F1: {'Tight': 0.5169712793733682, 'Medium': 0.3716577540106952, 'Weak': 0.7524299599771298}\n",
+      "Val - Reg Loss: 1.6478, Cls Loss: 0.8333\n",
+      "Val - Correlation: 0.5423, Accuracy: 0.5956\n",
+      "Val - Per-class F1: {'Tight': 0.48695652173913045, 'Medium': 0.13533834586466165, 'Weak': 0.7510548523206751}\n",
+      "Epoch 5/50:\n",
+      "Train - Reg Loss: 1.6100, Cls Loss: 0.8059\n",
+      "Train - Correlation: 0.5389, Accuracy: 0.6222\n",
+      "Train - Per-class F1: {'Tight': 0.5169712793733682, 'Medium': 0.3716577540106952, 'Weak': 0.7524299599771298}\n",
+      "Val - Reg Loss: 1.6478, Cls Loss: 0.8333\n",
+      "Val - Correlation: 0.5423, Accuracy: 0.5956\n",
+      "Val - Per-class F1: {'Tight': 0.48695652173913045, 'Medium': 0.13533834586466165, 'Weak': 0.7510548523206751}\n",
+      "Epoch 6/50:\n",
+      "Train - Reg Loss: 1.4593, Cls Loss: 0.7777\n",
+      "Train - Correlation: 0.5894, Accuracy: 0.6417\n",
+      "Train - Per-class F1: {'Tight': 0.5580246913580247, 'Medium': 0.31309904153354634, 'Weak': 0.7712276906435911}\n",
+      "Val - Reg Loss: 1.6347, Cls Loss: 0.8518\n",
+      "Val - Correlation: 0.5675, Accuracy: 0.6233\n",
+      "Val - Per-class F1: {'Tight': 0.3116883116883117, 'Medium': 0.48559670781893005, 'Weak': 0.7661691542288557}\n",
+      "✓ Saved best model at epoch 6 with validation correlation: 0.5675\n",
+      "Epoch 6/50:\n",
+      "Train - Reg Loss: 1.4593, Cls Loss: 0.7777\n",
+      "Train - Correlation: 0.5894, Accuracy: 0.6417\n",
+      "Train - Per-class F1: {'Tight': 0.5580246913580247, 'Medium': 0.31309904153354634, 'Weak': 0.7712276906435911}\n",
+      "Val - Reg Loss: 1.6347, Cls Loss: 0.8518\n",
+      "Val - Correlation: 0.5675, Accuracy: 0.6233\n",
+      "Val - Per-class F1: {'Tight': 0.3116883116883117, 'Medium': 0.48559670781893005, 'Weak': 0.7661691542288557}\n",
+      "✓ Saved best model at epoch 6 with validation correlation: 0.5675\n",
+      "✓ Saved correlation plots: best_model_val_correlation.png, best_model_train_correlation.png\n",
+      "✓ Saved correlation plots: best_model_val_correlation.png, best_model_train_correlation.png\n",
+      "Epoch 7/50:\n",
+      "Train - Reg Loss: 1.5465, Cls Loss: 0.7802\n",
+      "Train - Correlation: 0.5603, Accuracy: 0.6361\n",
+      "Train - Per-class F1: {'Tight': 0.5236907730673317, 'Medium': 0.39348710990502034, 'Weak': 0.764638346727899}\n",
+      "Val - Reg Loss: 1.6907, Cls Loss: 0.8079\n",
+      "Val - Correlation: 0.5505, Accuracy: 0.6371\n",
+      "Val - Per-class F1: {'Tight': 0.44, 'Medium': 0.4365482233502538, 'Weak': 0.7764705882352941}\n",
+      "Epoch 7/50:\n",
+      "Train - Reg Loss: 1.5465, Cls Loss: 0.7802\n",
+      "Train - Correlation: 0.5603, Accuracy: 0.6361\n",
+      "Train - Per-class F1: {'Tight': 0.5236907730673317, 'Medium': 0.39348710990502034, 'Weak': 0.764638346727899}\n",
+      "Val - Reg Loss: 1.6907, Cls Loss: 0.8079\n",
+      "Val - Correlation: 0.5505, Accuracy: 0.6371\n",
+      "Val - Per-class F1: {'Tight': 0.44, 'Medium': 0.4365482233502538, 'Weak': 0.7764705882352941}\n",
+      "Epoch 8/50:\n",
+      "Train - Reg Loss: 1.3665, Cls Loss: 0.7463\n",
+      "Train - Correlation: 0.6111, Accuracy: 0.6625\n",
+      "Train - Per-class F1: {'Tight': 0.594847775175644, 'Medium': 0.3854014598540146, 'Weak': 0.7861990950226244}\n",
+      "Val - Reg Loss: 1.6440, Cls Loss: 0.8222\n",
+      "Val - Correlation: 0.5450, Accuracy: 0.6316\n",
+      "Val - Per-class F1: {'Tight': 0.4090909090909091, 'Medium': 0.4782608695652174, 'Weak': 0.7673267326732673}\n",
+      "Epoch 8/50:\n",
+      "Train - Reg Loss: 1.3665, Cls Loss: 0.7463\n",
+      "Train - Correlation: 0.6111, Accuracy: 0.6625\n",
+      "Train - Per-class F1: {'Tight': 0.594847775175644, 'Medium': 0.3854014598540146, 'Weak': 0.7861990950226244}\n",
+      "Val - Reg Loss: 1.6440, Cls Loss: 0.8222\n",
+      "Val - Correlation: 0.5450, Accuracy: 0.6316\n",
+      "Val - Per-class F1: {'Tight': 0.4090909090909091, 'Medium': 0.4782608695652174, 'Weak': 0.7673267326732673}\n",
+      "Epoch 9/50:\n",
+      "Train - Reg Loss: 1.3592, Cls Loss: 0.7340\n",
+      "Train - Correlation: 0.6147, Accuracy: 0.6549\n",
+      "Train - Per-class F1: {'Tight': 0.5876543209876544, 'Medium': 0.40934065934065933, 'Weak': 0.7727532913566113}\n",
+      "Val - Reg Loss: 1.6071, Cls Loss: 0.8304\n",
+      "Val - Correlation: 0.5584, Accuracy: 0.5900\n",
+      "Val - Per-class F1: {'Tight': 0.345679012345679, 'Medium': 0.2905027932960894, 'Weak': 0.7489177489177489}\n",
+      "Epoch 9/50:\n",
+      "Train - Reg Loss: 1.3592, Cls Loss: 0.7340\n",
+      "Train - Correlation: 0.6147, Accuracy: 0.6549\n",
+      "Train - Per-class F1: {'Tight': 0.5876543209876544, 'Medium': 0.40934065934065933, 'Weak': 0.7727532913566113}\n",
+      "Val - Reg Loss: 1.6071, Cls Loss: 0.8304\n",
+      "Val - Correlation: 0.5584, Accuracy: 0.5900\n",
+      "Val - Per-class F1: {'Tight': 0.345679012345679, 'Medium': 0.2905027932960894, 'Weak': 0.7489177489177489}\n",
+      "Epoch 10/50:\n",
+      "Train - Reg Loss: 1.2531, Cls Loss: 0.7026\n",
+      "Train - Correlation: 0.6511, Accuracy: 0.6521\n",
+      "Train - Per-class F1: {'Tight': 0.5692307692307692, 'Medium': 0.3883495145631068, 'Weak': 0.7778405879027699}\n",
+      "Val - Reg Loss: 1.6013, Cls Loss: 0.8312\n",
+      "Val - Correlation: 0.5382, Accuracy: 0.6094\n",
+      "Val - Per-class F1: {'Tight': 0.5148514851485149, 'Medium': 0.23225806451612904, 'Weak': 0.7553648068669528}\n",
+      "Epoch 10/50:\n",
+      "Train - Reg Loss: 1.2531, Cls Loss: 0.7026\n",
+      "Train - Correlation: 0.6511, Accuracy: 0.6521\n",
+      "Train - Per-class F1: {'Tight': 0.5692307692307692, 'Medium': 0.3883495145631068, 'Weak': 0.7778405879027699}\n",
+      "Val - Reg Loss: 1.6013, Cls Loss: 0.8312\n",
+      "Val - Correlation: 0.5382, Accuracy: 0.6094\n",
+      "Val - Per-class F1: {'Tight': 0.5148514851485149, 'Medium': 0.23225806451612904, 'Weak': 0.7553648068669528}\n",
+      "Epoch 11/50:\n",
+      "Train - Reg Loss: 1.1266, Cls Loss: 0.6790\n",
+      "Train - Correlation: 0.6856, Accuracy: 0.6833\n",
+      "Train - Per-class F1: {'Tight': 0.6297229219143576, 'Medium': 0.45526315789473687, 'Weak': 0.796285548461985}\n",
+      "Val - Reg Loss: 1.7241, Cls Loss: 0.8286\n",
+      "Val - Correlation: 0.5590, Accuracy: 0.6565\n",
+      "Val - Per-class F1: {'Tight': 0.5370370370370371, 'Medium': 0.46632124352331605, 'Weak': 0.7743467933491687}\n",
+      "Epoch 11/50:\n",
+      "Train - Reg Loss: 1.1266, Cls Loss: 0.6790\n",
+      "Train - Correlation: 0.6856, Accuracy: 0.6833\n",
+      "Train - Per-class F1: {'Tight': 0.6297229219143576, 'Medium': 0.45526315789473687, 'Weak': 0.796285548461985}\n",
+      "Val - Reg Loss: 1.7241, Cls Loss: 0.8286\n",
+      "Val - Correlation: 0.5590, Accuracy: 0.6565\n",
+      "Val - Per-class F1: {'Tight': 0.5370370370370371, 'Medium': 0.46632124352331605, 'Weak': 0.7743467933491687}\n",
+      "Epoch 12/50:\n",
+      "Train - Reg Loss: 1.1405, Cls Loss: 0.6671\n",
+      "Train - Correlation: 0.6678, Accuracy: 0.6875\n",
+      "Train - Per-class F1: {'Tight': 0.6585956416464891, 'Medium': 0.4410958904109589, 'Weak': 0.7979274611398963}\n",
+      "Val - Reg Loss: 1.5747, Cls Loss: 0.8214\n",
+      "Val - Correlation: 0.5705, Accuracy: 0.6676\n",
+      "Val - Per-class F1: {'Tight': 0.5517241379310345, 'Medium': 0.4624277456647399, 'Weak': 0.7806004618937644}\n",
+      "✓ Saved best model at epoch 12 with validation correlation: 0.5705\n",
+      "Epoch 12/50:\n",
+      "Train - Reg Loss: 1.1405, Cls Loss: 0.6671\n",
+      "Train - Correlation: 0.6678, Accuracy: 0.6875\n",
+      "Train - Per-class F1: {'Tight': 0.6585956416464891, 'Medium': 0.4410958904109589, 'Weak': 0.7979274611398963}\n",
+      "Val - Reg Loss: 1.5747, Cls Loss: 0.8214\n",
+      "Val - Correlation: 0.5705, Accuracy: 0.6676\n",
+      "Val - Per-class F1: {'Tight': 0.5517241379310345, 'Medium': 0.4624277456647399, 'Weak': 0.7806004618937644}\n",
+      "✓ Saved best model at epoch 12 with validation correlation: 0.5705\n",
+      "✓ Saved correlation plots: best_model_val_correlation.png, best_model_train_correlation.png\n",
+      "✓ Saved correlation plots: best_model_val_correlation.png, best_model_train_correlation.png\n",
+      "Epoch 13/50:\n",
+      "Train - Reg Loss: 1.1291, Cls Loss: 0.6631\n",
+      "Train - Correlation: 0.6965, Accuracy: 0.6868\n",
+      "Train - Per-class F1: {'Tight': 0.6374133949191686, 'Medium': 0.4609375, 'Weak': 0.8028588445503275}\n",
+      "Val - Reg Loss: 1.5680, Cls Loss: 0.9303\n",
+      "Val - Correlation: 0.5921, Accuracy: 0.5152\n",
+      "Val - Per-class F1: {'Tight': 0.345679012345679, 'Medium': 0.5029239766081871, 'Weak': 0.5752508361204013}\n",
+      "✓ Saved best model at epoch 13 with validation correlation: 0.5921\n",
+      "Epoch 13/50:\n",
+      "Train - Reg Loss: 1.1291, Cls Loss: 0.6631\n",
+      "Train - Correlation: 0.6965, Accuracy: 0.6868\n",
+      "Train - Per-class F1: {'Tight': 0.6374133949191686, 'Medium': 0.4609375, 'Weak': 0.8028588445503275}\n",
+      "Val - Reg Loss: 1.5680, Cls Loss: 0.9303\n",
+      "Val - Correlation: 0.5921, Accuracy: 0.5152\n",
+      "Val - Per-class F1: {'Tight': 0.345679012345679, 'Medium': 0.5029239766081871, 'Weak': 0.5752508361204013}\n",
+      "✓ Saved best model at epoch 13 with validation correlation: 0.5921\n",
+      "✓ Saved correlation plots: best_model_val_correlation.png, best_model_train_correlation.png\n",
+      "✓ Saved correlation plots: best_model_val_correlation.png, best_model_train_correlation.png\n",
+      "Epoch 14/50:\n",
+      "Train - Reg Loss: 1.1646, Cls Loss: 0.6602\n",
+      "Train - Correlation: 0.6792, Accuracy: 0.7007\n",
+      "Train - Per-class F1: {'Tight': 0.6682808716707022, 'Medium': 0.4778523489932886, 'Weak': 0.8048780487804879}\n",
+      "Val - Reg Loss: 1.5833, Cls Loss: 0.8376\n",
+      "Val - Correlation: 0.5928, Accuracy: 0.6648\n",
+      "Val - Per-class F1: {'Tight': 0.5544554455445545, 'Medium': 0.43956043956043955, 'Weak': 0.7835990888382688}\n",
+      "✓ Saved best model at epoch 14 with validation correlation: 0.5928\n",
+      "Epoch 14/50:\n",
+      "Train - Reg Loss: 1.1646, Cls Loss: 0.6602\n",
+      "Train - Correlation: 0.6792, Accuracy: 0.7007\n",
+      "Train - Per-class F1: {'Tight': 0.6682808716707022, 'Medium': 0.4778523489932886, 'Weak': 0.8048780487804879}\n",
+      "Val - Reg Loss: 1.5833, Cls Loss: 0.8376\n",
+      "Val - Correlation: 0.5928, Accuracy: 0.6648\n",
+      "Val - Per-class F1: {'Tight': 0.5544554455445545, 'Medium': 0.43956043956043955, 'Weak': 0.7835990888382688}\n",
+      "✓ Saved best model at epoch 14 with validation correlation: 0.5928\n",
+      "✓ Saved correlation plots: best_model_val_correlation.png, best_model_train_correlation.png\n",
+      "✓ Saved correlation plots: best_model_val_correlation.png, best_model_train_correlation.png\n",
+      "Epoch 15/50:\n",
+      "Train - Reg Loss: 0.9602, Cls Loss: 0.6058\n",
+      "Train - Correlation: 0.7440, Accuracy: 0.7215\n",
+      "Train - Per-class F1: {'Tight': 0.672645739910314, 'Medium': 0.5363984674329502, 'Weak': 0.8225317989097517}\n",
+      "Val - Reg Loss: 1.5925, Cls Loss: 0.9322\n",
+      "Val - Correlation: 0.5860, Accuracy: 0.6343\n",
+      "Val - Per-class F1: {'Tight': 0.5490196078431373, 'Medium': 0.3, 'Weak': 0.7695652173913043}\n",
+      "Epoch 15/50:\n",
+      "Train - Reg Loss: 0.9602, Cls Loss: 0.6058\n",
+      "Train - Correlation: 0.7440, Accuracy: 0.7215\n",
+      "Train - Per-class F1: {'Tight': 0.672645739910314, 'Medium': 0.5363984674329502, 'Weak': 0.8225317989097517}\n",
+      "Val - Reg Loss: 1.5925, Cls Loss: 0.9322\n",
+      "Val - Correlation: 0.5860, Accuracy: 0.6343\n",
+      "Val - Per-class F1: {'Tight': 0.5490196078431373, 'Medium': 0.3, 'Weak': 0.7695652173913043}\n",
+      "Epoch 16/50:\n",
+      "Train - Reg Loss: 0.9619, Cls Loss: 0.5933\n",
+      "Train - Correlation: 0.7410, Accuracy: 0.7236\n",
+      "Train - Per-class F1: {'Tight': 0.689156626506024, 'Medium': 0.5257731958762887, 'Weak': 0.8229721728833629}\n",
+      "Val - Reg Loss: 1.7476, Cls Loss: 0.9324\n",
+      "Val - Correlation: 0.5812, Accuracy: 0.6454\n",
+      "Val - Per-class F1: {'Tight': 0.5225225225225225, 'Medium': 0.3742690058479532, 'Weak': 0.7818181818181819}\n",
+      "Epoch 16/50:\n",
+      "Train - Reg Loss: 0.9619, Cls Loss: 0.5933\n",
+      "Train - Correlation: 0.7410, Accuracy: 0.7236\n",
+      "Train - Per-class F1: {'Tight': 0.689156626506024, 'Medium': 0.5257731958762887, 'Weak': 0.8229721728833629}\n",
+      "Val - Reg Loss: 1.7476, Cls Loss: 0.9324\n",
+      "Val - Correlation: 0.5812, Accuracy: 0.6454\n",
+      "Val - Per-class F1: {'Tight': 0.5225225225225225, 'Medium': 0.3742690058479532, 'Weak': 0.7818181818181819}\n",
+      "Epoch 17/50:\n",
+      "Train - Reg Loss: 0.9729, Cls Loss: 0.5900\n",
+      "Train - Correlation: 0.7418, Accuracy: 0.7382\n",
+      "Train - Per-class F1: {'Tight': 0.6681922196796338, 'Medium': 0.5492772667542707, 'Weak': 0.8418549346016647}\n",
+      "Val - Reg Loss: 1.5528, Cls Loss: 0.9079\n",
+      "Val - Correlation: 0.5854, Accuracy: 0.6288\n",
+      "Val - Per-class F1: {'Tight': 0.3855421686746988, 'Medium': 0.49166666666666664, 'Weak': 0.7619047619047619}\n",
+      "Epoch 17/50:\n",
+      "Train - Reg Loss: 0.9729, Cls Loss: 0.5900\n",
+      "Train - Correlation: 0.7418, Accuracy: 0.7382\n",
+      "Train - Per-class F1: {'Tight': 0.6681922196796338, 'Medium': 0.5492772667542707, 'Weak': 0.8418549346016647}\n",
+      "Val - Reg Loss: 1.5528, Cls Loss: 0.9079\n",
+      "Val - Correlation: 0.5854, Accuracy: 0.6288\n",
+      "Val - Per-class F1: {'Tight': 0.3855421686746988, 'Medium': 0.49166666666666664, 'Weak': 0.7619047619047619}\n",
+      "Epoch 18/50:\n",
+      "Train - Reg Loss: 0.8553, Cls Loss: 0.5584\n",
+      "Train - Correlation: 0.7762, Accuracy: 0.7403\n",
+      "Train - Per-class F1: {'Tight': 0.7095238095238096, 'Medium': 0.5755743651753326, 'Weak': 0.8315982853643601}\n",
+      "Val - Reg Loss: 1.7244, Cls Loss: 0.9134\n",
+      "Val - Correlation: 0.5720, Accuracy: 0.6011\n",
+      "Val - Per-class F1: {'Tight': 0.4583333333333333, 'Medium': 0.49808429118773945, 'Weak': 0.7123287671232876}\n",
+      "Epoch 18/50:\n",
+      "Train - Reg Loss: 0.8553, Cls Loss: 0.5584\n",
+      "Train - Correlation: 0.7762, Accuracy: 0.7403\n",
+      "Train - Per-class F1: {'Tight': 0.7095238095238096, 'Medium': 0.5755743651753326, 'Weak': 0.8315982853643601}\n",
+      "Val - Reg Loss: 1.7244, Cls Loss: 0.9134\n",
+      "Val - Correlation: 0.5720, Accuracy: 0.6011\n",
+      "Val - Per-class F1: {'Tight': 0.4583333333333333, 'Medium': 0.49808429118773945, 'Weak': 0.7123287671232876}\n",
+      "Epoch 19/50:\n",
+      "Train - Reg Loss: 0.8436, Cls Loss: 0.5634\n",
+      "Train - Correlation: 0.7801, Accuracy: 0.7431\n",
+      "Train - Per-class F1: {'Tight': 0.6906474820143885, 'Medium': 0.5755743651753326, 'Weak': 0.8410757946210269}\n",
+      "Val - Reg Loss: 1.6332, Cls Loss: 0.8848\n",
+      "Val - Correlation: 0.5950, Accuracy: 0.6648\n",
+      "Val - Per-class F1: {'Tight': 0.5523809523809524, 'Medium': 0.44329896907216493, 'Weak': 0.7943262411347518}\n",
+      "✓ Saved best model at epoch 19 with validation correlation: 0.5950\n",
+      "Epoch 19/50:\n",
+      "Train - Reg Loss: 0.8436, Cls Loss: 0.5634\n",
+      "Train - Correlation: 0.7801, Accuracy: 0.7431\n",
+      "Train - Per-class F1: {'Tight': 0.6906474820143885, 'Medium': 0.5755743651753326, 'Weak': 0.8410757946210269}\n",
+      "Val - Reg Loss: 1.6332, Cls Loss: 0.8848\n",
+      "Val - Correlation: 0.5950, Accuracy: 0.6648\n",
+      "Val - Per-class F1: {'Tight': 0.5523809523809524, 'Medium': 0.44329896907216493, 'Weak': 0.7943262411347518}\n",
+      "✓ Saved best model at epoch 19 with validation correlation: 0.5950\n",
+      "✓ Saved correlation plots: best_model_val_correlation.png, best_model_train_correlation.png\n",
+      "✓ Saved correlation plots: best_model_val_correlation.png, best_model_train_correlation.png\n",
+      "Epoch 20/50:\n",
+      "Train - Reg Loss: 0.7833, Cls Loss: 0.5067\n",
+      "Train - Correlation: 0.7945, Accuracy: 0.7778\n",
+      "Train - Per-class F1: {'Tight': 0.7552447552447552, 'Medium': 0.6219974715549936, 'Weak': 0.8578313253012049}\n",
+      "Val - Reg Loss: 1.5962, Cls Loss: 0.9356\n",
+      "Val - Correlation: 0.5886, Accuracy: 0.6177\n",
+      "Val - Per-class F1: {'Tight': 0.4044943820224719, 'Medium': 0.5171102661596958, 'Weak': 0.7405405405405405}\n",
+      "Epoch 20/50:\n",
+      "Train - Reg Loss: 0.7833, Cls Loss: 0.5067\n",
+      "Train - Correlation: 0.7945, Accuracy: 0.7778\n",
+      "Train - Per-class F1: {'Tight': 0.7552447552447552, 'Medium': 0.6219974715549936, 'Weak': 0.8578313253012049}\n",
+      "Val - Reg Loss: 1.5962, Cls Loss: 0.9356\n",
+      "Val - Correlation: 0.5886, Accuracy: 0.6177\n",
+      "Val - Per-class F1: {'Tight': 0.4044943820224719, 'Medium': 0.5171102661596958, 'Weak': 0.7405405405405405}\n",
+      "Epoch 21/50:\n",
+      "Train - Reg Loss: 0.7506, Cls Loss: 0.5018\n",
+      "Train - Correlation: 0.8054, Accuracy: 0.7771\n",
+      "Train - Per-class F1: {'Tight': 0.7398568019093079, 'Medium': 0.628992628992629, 'Weak': 0.8597449908925319}\n",
+      "Val - Reg Loss: 1.6593, Cls Loss: 0.9450\n",
+      "Val - Correlation: 0.6109, Accuracy: 0.6177\n",
+      "Val - Per-class F1: {'Tight': 0.4731182795698925, 'Medium': 0.518796992481203, 'Weak': 0.7272727272727273}\n",
+      "✓ Saved best model at epoch 21 with validation correlation: 0.6109\n",
+      "Epoch 21/50:\n",
+      "Train - Reg Loss: 0.7506, Cls Loss: 0.5018\n",
+      "Train - Correlation: 0.8054, Accuracy: 0.7771\n",
+      "Train - Per-class F1: {'Tight': 0.7398568019093079, 'Medium': 0.628992628992629, 'Weak': 0.8597449908925319}\n",
+      "Val - Reg Loss: 1.6593, Cls Loss: 0.9450\n",
+      "Val - Correlation: 0.6109, Accuracy: 0.6177\n",
+      "Val - Per-class F1: {'Tight': 0.4731182795698925, 'Medium': 0.518796992481203, 'Weak': 0.7272727272727273}\n",
+      "✓ Saved best model at epoch 21 with validation correlation: 0.6109\n",
+      "✓ Saved correlation plots: best_model_val_correlation.png, best_model_train_correlation.png\n",
+      "✓ Saved correlation plots: best_model_val_correlation.png, best_model_train_correlation.png\n",
+      "Epoch 22/50:\n",
+      "Train - Reg Loss: 0.7011, Cls Loss: 0.4680\n",
+      "Train - Correlation: 0.8194, Accuracy: 0.7972\n",
+      "Train - Per-class F1: {'Tight': 0.7652582159624414, 'Medium': 0.6658653846153846, 'Weak': 0.872996300863132}\n",
+      "Val - Reg Loss: 1.7312, Cls Loss: 0.9957\n",
+      "Val - Correlation: 0.5871, Accuracy: 0.6371\n",
+      "Val - Per-class F1: {'Tight': 0.4090909090909091, 'Medium': 0.5390625, 'Weak': 0.7566137566137566}\n",
+      "Epoch 22/50:\n",
+      "Train - Reg Loss: 0.7011, Cls Loss: 0.4680\n",
+      "Train - Correlation: 0.8194, Accuracy: 0.7972\n",
+      "Train - Per-class F1: {'Tight': 0.7652582159624414, 'Medium': 0.6658653846153846, 'Weak': 0.872996300863132}\n",
+      "Val - Reg Loss: 1.7312, Cls Loss: 0.9957\n",
+      "Val - Correlation: 0.5871, Accuracy: 0.6371\n",
+      "Val - Per-class F1: {'Tight': 0.4090909090909091, 'Medium': 0.5390625, 'Weak': 0.7566137566137566}\n",
+      "Epoch 23/50:\n",
+      "Train - Reg Loss: 0.6680, Cls Loss: 0.4425\n",
+      "Train - Correlation: 0.8347, Accuracy: 0.8042\n",
+      "Train - Per-class F1: {'Tight': 0.7749419953596288, 'Medium': 0.6714975845410628, 'Weak': 0.8797038864898211}\n",
+      "Val - Reg Loss: 1.6876, Cls Loss: 0.9984\n",
+      "Val - Correlation: 0.5762, Accuracy: 0.6288\n",
+      "Val - Per-class F1: {'Tight': 0.48484848484848486, 'Medium': 0.4935064935064935, 'Weak': 0.7448979591836735}\n",
+      "Epoch 23/50:\n",
+      "Train - Reg Loss: 0.6680, Cls Loss: 0.4425\n",
+      "Train - Correlation: 0.8347, Accuracy: 0.8042\n",
+      "Train - Per-class F1: {'Tight': 0.7749419953596288, 'Medium': 0.6714975845410628, 'Weak': 0.8797038864898211}\n",
+      "Val - Reg Loss: 1.6876, Cls Loss: 0.9984\n",
+      "Val - Correlation: 0.5762, Accuracy: 0.6288\n",
+      "Val - Per-class F1: {'Tight': 0.48484848484848486, 'Medium': 0.4935064935064935, 'Weak': 0.7448979591836735}\n",
+      "Epoch 24/50:\n",
+      "Train - Reg Loss: 0.5978, Cls Loss: 0.4078\n",
+      "Train - Correlation: 0.8469, Accuracy: 0.8160\n",
+      "Train - Per-class F1: {'Tight': 0.7889908256880734, 'Medium': 0.6905940594059405, 'Weak': 0.8850855745721271}\n",
+      "Val - Reg Loss: 1.7259, Cls Loss: 1.1171\n",
+      "Val - Correlation: 0.5879, Accuracy: 0.6371\n",
+      "Val - Per-class F1: {'Tight': 0.53125, 'Medium': 0.44329896907216493, 'Weak': 0.765}\n",
+      "Epoch 24/50:\n",
+      "Train - Reg Loss: 0.5978, Cls Loss: 0.4078\n",
+      "Train - Correlation: 0.8469, Accuracy: 0.8160\n",
+      "Train - Per-class F1: {'Tight': 0.7889908256880734, 'Medium': 0.6905940594059405, 'Weak': 0.8850855745721271}\n",
+      "Val - Reg Loss: 1.7259, Cls Loss: 1.1171\n",
+      "Val - Correlation: 0.5879, Accuracy: 0.6371\n",
+      "Val - Per-class F1: {'Tight': 0.53125, 'Medium': 0.44329896907216493, 'Weak': 0.765}\n",
+      "Epoch 25/50:\n",
+      "Train - Reg Loss: 0.5982, Cls Loss: 0.4047\n",
+      "Train - Correlation: 0.8461, Accuracy: 0.8326\n",
+      "Train - Per-class F1: {'Tight': 0.8, 'Medium': 0.7218225419664268, 'Weak': 0.897908979089791}\n",
+      "Val - Reg Loss: 1.6857, Cls Loss: 1.1744\n",
+      "Val - Correlation: 0.5793, Accuracy: 0.6371\n",
+      "Val - Per-class F1: {'Tight': 0.4444444444444444, 'Medium': 0.36046511627906974, 'Weak': 0.7849223946784922}\n",
+      "Epoch 25/50:\n",
+      "Train - Reg Loss: 0.5982, Cls Loss: 0.4047\n",
+      "Train - Correlation: 0.8461, Accuracy: 0.8326\n",
+      "Train - Per-class F1: {'Tight': 0.8, 'Medium': 0.7218225419664268, 'Weak': 0.897908979089791}\n",
+      "Val - Reg Loss: 1.6857, Cls Loss: 1.1744\n",
+      "Val - Correlation: 0.5793, Accuracy: 0.6371\n",
+      "Val - Per-class F1: {'Tight': 0.4444444444444444, 'Medium': 0.36046511627906974, 'Weak': 0.7849223946784922}\n",
+      "Epoch 26/50:\n",
+      "Train - Reg Loss: 0.6350, Cls Loss: 0.4045\n",
+      "Train - Correlation: 0.8393, Accuracy: 0.8264\n",
+      "Train - Per-class F1: {'Tight': 0.7955555555555556, 'Medium': 0.7177033492822966, 'Weak': 0.8920953575909661}\n",
+      "Val - Reg Loss: 1.5956, Cls Loss: 1.1335\n",
+      "Val - Correlation: 0.5913, Accuracy: 0.6731\n",
+      "Val - Per-class F1: {'Tight': 0.4835164835164835, 'Medium': 0.5233644859813084, 'Weak': 0.7913669064748201}\n",
+      "Epoch 26/50:\n",
+      "Train - Reg Loss: 0.6350, Cls Loss: 0.4045\n",
+      "Train - Correlation: 0.8393, Accuracy: 0.8264\n",
+      "Train - Per-class F1: {'Tight': 0.7955555555555556, 'Medium': 0.7177033492822966, 'Weak': 0.8920953575909661}\n",
+      "Val - Reg Loss: 1.5956, Cls Loss: 1.1335\n",
+      "Val - Correlation: 0.5913, Accuracy: 0.6731\n",
+      "Val - Per-class F1: {'Tight': 0.4835164835164835, 'Medium': 0.5233644859813084, 'Weak': 0.7913669064748201}\n",
+      "Epoch 27/50:\n",
+      "Train - Reg Loss: 0.5977, Cls Loss: 0.3590\n",
+      "Train - Correlation: 0.8561, Accuracy: 0.8528\n",
+      "Train - Per-class F1: {'Tight': 0.7933491686460807, 'Medium': 0.764774044032445, 'Weak': 0.9160401002506265}\n",
+      "Val - Reg Loss: 1.7333, Cls Loss: 1.1621\n",
+      "Val - Correlation: 0.5950, Accuracy: 0.6620\n",
+      "Val - Per-class F1: {'Tight': 0.5098039215686274, 'Medium': 0.41935483870967744, 'Weak': 0.8018433179723502}\n",
+      "Epoch 27/50:\n",
+      "Train - Reg Loss: 0.5977, Cls Loss: 0.3590\n",
+      "Train - Correlation: 0.8561, Accuracy: 0.8528\n",
+      "Train - Per-class F1: {'Tight': 0.7933491686460807, 'Medium': 0.764774044032445, 'Weak': 0.9160401002506265}\n",
+      "Val - Reg Loss: 1.7333, Cls Loss: 1.1621\n",
+      "Val - Correlation: 0.5950, Accuracy: 0.6620\n",
+      "Val - Per-class F1: {'Tight': 0.5098039215686274, 'Medium': 0.41935483870967744, 'Weak': 0.8018433179723502}\n",
+      "Epoch 28/50:\n",
+      "Train - Reg Loss: 0.5340, Cls Loss: 0.3483\n",
+      "Train - Correlation: 0.8675, Accuracy: 0.8535\n",
+      "Train - Per-class F1: {'Tight': 0.8288288288288288, 'Medium': 0.7624703087885986, 'Weak': 0.9084065244667503}\n",
+      "Val - Reg Loss: 1.7225, Cls Loss: 1.2035\n",
+      "Val - Correlation: 0.5659, Accuracy: 0.6343\n",
+      "Val - Per-class F1: {'Tight': 0.4, 'Medium': 0.5166666666666667, 'Weak': 0.7556675062972292}\n",
+      "Epoch 28/50:\n",
+      "Train - Reg Loss: 0.5340, Cls Loss: 0.3483\n",
+      "Train - Correlation: 0.8675, Accuracy: 0.8535\n",
+      "Train - Per-class F1: {'Tight': 0.8288288288288288, 'Medium': 0.7624703087885986, 'Weak': 0.9084065244667503}\n",
+      "Val - Reg Loss: 1.7225, Cls Loss: 1.2035\n",
+      "Val - Correlation: 0.5659, Accuracy: 0.6343\n",
+      "Val - Per-class F1: {'Tight': 0.4, 'Medium': 0.5166666666666667, 'Weak': 0.7556675062972292}\n",
+      "Epoch 29/50:\n",
+      "Train - Reg Loss: 0.4901, Cls Loss: 0.3211\n",
+      "Train - Correlation: 0.8832, Accuracy: 0.8674\n",
+      "Train - Per-class F1: {'Tight': 0.8262910798122066, 'Medium': 0.7850467289719626, 'Weak': 0.9224030037546934}\n",
+      "Val - Reg Loss: 1.9583, Cls Loss: 1.3001\n",
+      "Val - Correlation: 0.5899, Accuracy: 0.6150\n",
+      "Val - Per-class F1: {'Tight': 0.52, 'Medium': 0.5095057034220533, 'Weak': 0.7186629526462396}\n",
+      "Epoch 29/50:\n",
+      "Train - Reg Loss: 0.4901, Cls Loss: 0.3211\n",
+      "Train - Correlation: 0.8832, Accuracy: 0.8674\n",
+      "Train - Per-class F1: {'Tight': 0.8262910798122066, 'Medium': 0.7850467289719626, 'Weak': 0.9224030037546934}\n",
+      "Val - Reg Loss: 1.9583, Cls Loss: 1.3001\n",
+      "Val - Correlation: 0.5899, Accuracy: 0.6150\n",
+      "Val - Per-class F1: {'Tight': 0.52, 'Medium': 0.5095057034220533, 'Weak': 0.7186629526462396}\n",
+      "Epoch 30/50:\n",
+      "Train - Reg Loss: 0.4755, Cls Loss: 0.3126\n",
+      "Train - Correlation: 0.8796, Accuracy: 0.8694\n",
+      "Train - Per-class F1: {'Tight': 0.8473804100227791, 'Medium': 0.7768395657418576, 'Weak': 0.9230769230769231}\n",
+      "Val - Reg Loss: 1.5722, Cls Loss: 1.1841\n",
+      "Val - Correlation: 0.5984, Accuracy: 0.6454\n",
+      "Val - Per-class F1: {'Tight': 0.5471698113207547, 'Medium': 0.5128205128205128, 'Weak': 0.7539267015706806}\n",
+      "Epoch 30/50:\n",
+      "Train - Reg Loss: 0.4755, Cls Loss: 0.3126\n",
+      "Train - Correlation: 0.8796, Accuracy: 0.8694\n",
+      "Train - Per-class F1: {'Tight': 0.8473804100227791, 'Medium': 0.7768395657418576, 'Weak': 0.9230769230769231}\n",
+      "Val - Reg Loss: 1.5722, Cls Loss: 1.1841\n",
+      "Val - Correlation: 0.5984, Accuracy: 0.6454\n",
+      "Val - Per-class F1: {'Tight': 0.5471698113207547, 'Medium': 0.5128205128205128, 'Weak': 0.7539267015706806}\n",
+      "Epoch 31/50:\n",
+      "Train - Reg Loss: 0.4378, Cls Loss: 0.2913\n",
+      "Train - Correlation: 0.8934, Accuracy: 0.8806\n",
+      "Train - Per-class F1: {'Tight': 0.8302752293577982, 'Medium': 0.8060747663551402, 'Weak': 0.9345088161209067}\n",
+      "Val - Reg Loss: 1.8473, Cls Loss: 1.2531\n",
+      "Val - Correlation: 0.5537, Accuracy: 0.5817\n",
+      "Val - Per-class F1: {'Tight': 0.4727272727272727, 'Medium': 0.4380165289256198, 'Weak': 0.7081081081081081}\n",
+      "Epoch 31/50:\n",
+      "Train - Reg Loss: 0.4378, Cls Loss: 0.2913\n",
+      "Train - Correlation: 0.8934, Accuracy: 0.8806\n",
+      "Train - Per-class F1: {'Tight': 0.8302752293577982, 'Medium': 0.8060747663551402, 'Weak': 0.9345088161209067}\n",
+      "Val - Reg Loss: 1.8473, Cls Loss: 1.2531\n",
+      "Val - Correlation: 0.5537, Accuracy: 0.5817\n",
+      "Val - Per-class F1: {'Tight': 0.4727272727272727, 'Medium': 0.4380165289256198, 'Weak': 0.7081081081081081}\n",
+      "Epoch 32/50:\n",
+      "Train - Reg Loss: 0.4320, Cls Loss: 0.2898\n",
+      "Train - Correlation: 0.8937, Accuracy: 0.8833\n",
+      "Train - Per-class F1: {'Tight': 0.8372093023255814, 'Medium': 0.8083623693379791, 'Weak': 0.9364380113278792}\n",
+      "Val - Reg Loss: 1.9495, Cls Loss: 1.3266\n",
+      "Val - Correlation: 0.5945, Accuracy: 0.6316\n",
+      "Val - Per-class F1: {'Tight': 0.5689655172413793, 'Medium': 0.4423076923076923, 'Weak': 0.7487437185929648}\n",
+      "Epoch 32/50:\n",
+      "Train - Reg Loss: 0.4320, Cls Loss: 0.2898\n",
+      "Train - Correlation: 0.8937, Accuracy: 0.8833\n",
+      "Train - Per-class F1: {'Tight': 0.8372093023255814, 'Medium': 0.8083623693379791, 'Weak': 0.9364380113278792}\n",
+      "Val - Reg Loss: 1.9495, Cls Loss: 1.3266\n",
+      "Val - Correlation: 0.5945, Accuracy: 0.6316\n",
+      "Val - Per-class F1: {'Tight': 0.5689655172413793, 'Medium': 0.4423076923076923, 'Weak': 0.7487437185929648}\n",
+      "Epoch 33/50:\n",
+      "Train - Reg Loss: 0.4233, Cls Loss: 0.2512\n",
+      "Train - Correlation: 0.9010, Accuracy: 0.9014\n",
+      "Train - Per-class F1: {'Tight': 0.8663594470046083, 'Medium': 0.8352941176470589, 'Weak': 0.9461152882205514}\n",
+      "Val - Reg Loss: 1.6778, Cls Loss: 1.4054\n",
+      "Val - Correlation: 0.5923, Accuracy: 0.6260\n",
+      "Val - Per-class F1: {'Tight': 0.4444444444444444, 'Medium': 0.5080645161290323, 'Weak': 0.7447916666666666}\n",
+      "Epoch 33/50:\n",
+      "Train - Reg Loss: 0.4233, Cls Loss: 0.2512\n",
+      "Train - Correlation: 0.9010, Accuracy: 0.9014\n",
+      "Train - Per-class F1: {'Tight': 0.8663594470046083, 'Medium': 0.8352941176470589, 'Weak': 0.9461152882205514}\n",
+      "Val - Reg Loss: 1.6778, Cls Loss: 1.4054\n",
+      "Val - Correlation: 0.5923, Accuracy: 0.6260\n",
+      "Val - Per-class F1: {'Tight': 0.4444444444444444, 'Medium': 0.5080645161290323, 'Weak': 0.7447916666666666}\n",
+      "Epoch 34/50:\n",
+      "Train - Reg Loss: 0.4325, Cls Loss: 0.2690\n",
+      "Train - Correlation: 0.9024, Accuracy: 0.8938\n",
+      "Train - Per-class F1: {'Tight': 0.8457943925233645, 'Medium': 0.8283062645011601, 'Weak': 0.9421383647798742}\n",
+      "Val - Reg Loss: 1.6346, Cls Loss: 1.2990\n",
+      "Val - Correlation: 0.5957, Accuracy: 0.6066\n",
+      "Val - Per-class F1: {'Tight': 0.48598130841121495, 'Medium': 0.44155844155844154, 'Weak': 0.7395833333333334}\n",
+      "Epoch 34/50:\n",
+      "Train - Reg Loss: 0.4325, Cls Loss: 0.2690\n",
+      "Train - Correlation: 0.9024, Accuracy: 0.8938\n",
+      "Train - Per-class F1: {'Tight': 0.8457943925233645, 'Medium': 0.8283062645011601, 'Weak': 0.9421383647798742}\n",
+      "Val - Reg Loss: 1.6346, Cls Loss: 1.2990\n",
+      "Val - Correlation: 0.5957, Accuracy: 0.6066\n",
+      "Val - Per-class F1: {'Tight': 0.48598130841121495, 'Medium': 0.44155844155844154, 'Weak': 0.7395833333333334}\n",
+      "Epoch 35/50:\n",
+      "Train - Reg Loss: 0.3902, Cls Loss: 0.2573\n",
+      "Train - Correlation: 0.9085, Accuracy: 0.8972\n",
+      "Train - Per-class F1: {'Tight': 0.8671328671328671, 'Medium': 0.8301886792452831, 'Weak': 0.940736119775421}\n",
+      "Val - Reg Loss: 1.7511, Cls Loss: 1.2812\n",
+      "Val - Correlation: 0.5829, Accuracy: 0.6150\n",
+      "Val - Per-class F1: {'Tight': 0.5087719298245614, 'Medium': 0.39603960396039606, 'Weak': 0.7536945812807881}\n",
+      "Epoch 35/50:\n",
+      "Train - Reg Loss: 0.3902, Cls Loss: 0.2573\n",
+      "Train - Correlation: 0.9085, Accuracy: 0.8972\n",
+      "Train - Per-class F1: {'Tight': 0.8671328671328671, 'Medium': 0.8301886792452831, 'Weak': 0.940736119775421}\n",
+      "Val - Reg Loss: 1.7511, Cls Loss: 1.2812\n",
+      "Val - Correlation: 0.5829, Accuracy: 0.6150\n",
+      "Val - Per-class F1: {'Tight': 0.5087719298245614, 'Medium': 0.39603960396039606, 'Weak': 0.7536945812807881}\n",
+      "Epoch 36/50:\n",
+      "Train - Reg Loss: 0.3583, Cls Loss: 0.2220\n",
+      "Train - Correlation: 0.9127, Accuracy: 0.9118\n",
+      "Train - Per-class F1: {'Tight': 0.8498845265588915, 'Medium': 0.8554641598119859, 'Weak': 0.9586466165413534}\n",
+      "Val - Reg Loss: 1.6853, Cls Loss: 1.3887\n",
+      "Val - Correlation: 0.5670, Accuracy: 0.6427\n",
+      "Val - Per-class F1: {'Tight': 0.4186046511627907, 'Medium': 0.552, 'Weak': 0.7512953367875648}\n",
+      "Epoch 36/50:\n",
+      "Train - Reg Loss: 0.3583, Cls Loss: 0.2220\n",
+      "Train - Correlation: 0.9127, Accuracy: 0.9118\n",
+      "Train - Per-class F1: {'Tight': 0.8498845265588915, 'Medium': 0.8554641598119859, 'Weak': 0.9586466165413534}\n",
+      "Val - Reg Loss: 1.6853, Cls Loss: 1.3887\n",
+      "Val - Correlation: 0.5670, Accuracy: 0.6427\n",
+      "Val - Per-class F1: {'Tight': 0.4186046511627907, 'Medium': 0.552, 'Weak': 0.7512953367875648}\n",
+      "Epoch 37/50:\n",
+      "Train - Reg Loss: 0.4009, Cls Loss: 0.2767\n",
+      "Train - Correlation: 0.9058, Accuracy: 0.8924\n",
+      "Train - Per-class F1: {'Tight': 0.8466819221967964, 'Medium': 0.8232502965599051, 'Weak': 0.94125}\n",
+      "Val - Reg Loss: 1.6305, Cls Loss: 1.3489\n",
+      "Val - Correlation: 0.5750, Accuracy: 0.6177\n",
+      "Val - Per-class F1: {'Tight': 0.5, 'Medium': 0.3567567567567568, 'Weak': 0.7575057736720554}\n",
+      "Epoch 37/50:\n",
+      "Train - Reg Loss: 0.4009, Cls Loss: 0.2767\n",
+      "Train - Correlation: 0.9058, Accuracy: 0.8924\n",
+      "Train - Per-class F1: {'Tight': 0.8466819221967964, 'Medium': 0.8232502965599051, 'Weak': 0.94125}\n",
+      "Val - Reg Loss: 1.6305, Cls Loss: 1.3489\n",
+      "Val - Correlation: 0.5750, Accuracy: 0.6177\n",
+      "Val - Per-class F1: {'Tight': 0.5, 'Medium': 0.3567567567567568, 'Weak': 0.7575057736720554}\n",
+      "Epoch 38/50:\n",
+      "Train - Reg Loss: 0.3749, Cls Loss: 0.2046\n",
+      "Train - Correlation: 0.9105, Accuracy: 0.9139\n",
+      "Train - Per-class F1: {'Tight': 0.8693693693693694, 'Medium': 0.8557806912991657, 'Weak': 0.9567939887288667}\n",
+      "Val - Reg Loss: 1.8119, Cls Loss: 1.4817\n",
+      "Val - Correlation: 0.5917, Accuracy: 0.5873\n",
+      "Val - Per-class F1: {'Tight': 0.38636363636363635, 'Medium': 0.5071428571428571, 'Weak': 0.7005649717514124}\n",
+      "Epoch 38/50:\n",
+      "Train - Reg Loss: 0.3749, Cls Loss: 0.2046\n",
+      "Train - Correlation: 0.9105, Accuracy: 0.9139\n",
+      "Train - Per-class F1: {'Tight': 0.8693693693693694, 'Medium': 0.8557806912991657, 'Weak': 0.9567939887288667}\n",
+      "Val - Reg Loss: 1.8119, Cls Loss: 1.4817\n",
+      "Val - Correlation: 0.5917, Accuracy: 0.5873\n",
+      "Val - Per-class F1: {'Tight': 0.38636363636363635, 'Medium': 0.5071428571428571, 'Weak': 0.7005649717514124}\n",
+      "Epoch 39/50:\n",
+      "Train - Reg Loss: 0.3424, Cls Loss: 0.1967\n",
+      "Train - Correlation: 0.9198, Accuracy: 0.9236\n",
+      "Train - Per-class F1: {'Tight': 0.8679245283018868, 'Medium': 0.875725900116144, 'Weak': 0.9642633228840125}\n",
+      "Val - Reg Loss: 1.8146, Cls Loss: 1.5760\n",
+      "Val - Correlation: 0.5988, Accuracy: 0.6399\n",
+      "Val - Per-class F1: {'Tight': 0.5, 'Medium': 0.43157894736842106, 'Weak': 0.7714285714285715}\n",
+      "Epoch 39/50:\n",
+      "Train - Reg Loss: 0.3424, Cls Loss: 0.1967\n",
+      "Train - Correlation: 0.9198, Accuracy: 0.9236\n",
+      "Train - Per-class F1: {'Tight': 0.8679245283018868, 'Medium': 0.875725900116144, 'Weak': 0.9642633228840125}\n",
+      "Val - Reg Loss: 1.8146, Cls Loss: 1.5760\n",
+      "Val - Correlation: 0.5988, Accuracy: 0.6399\n",
+      "Val - Per-class F1: {'Tight': 0.5, 'Medium': 0.43157894736842106, 'Weak': 0.7714285714285715}\n",
+      "Epoch 40/50:\n",
+      "Train - Reg Loss: 0.3737, Cls Loss: 0.2118\n",
+      "Train - Correlation: 0.9081, Accuracy: 0.9160\n",
+      "Train - Per-class F1: {'Tight': 0.8764568764568764, 'Medium': 0.8648018648018648, 'Weak': 0.9541745134965474}\n",
+      "Val - Reg Loss: 1.8482, Cls Loss: 1.4576\n",
+      "Val - Correlation: 0.5668, Accuracy: 0.6122\n",
+      "Val - Per-class F1: {'Tight': 0.4090909090909091, 'Medium': 0.5151515151515151, 'Weak': 0.7297297297297297}\n",
+      "Epoch 40/50:\n",
+      "Train - Reg Loss: 0.3737, Cls Loss: 0.2118\n",
+      "Train - Correlation: 0.9081, Accuracy: 0.9160\n",
+      "Train - Per-class F1: {'Tight': 0.8764568764568764, 'Medium': 0.8648018648018648, 'Weak': 0.9541745134965474}\n",
+      "Val - Reg Loss: 1.8482, Cls Loss: 1.4576\n",
+      "Val - Correlation: 0.5668, Accuracy: 0.6122\n",
+      "Val - Per-class F1: {'Tight': 0.4090909090909091, 'Medium': 0.5151515151515151, 'Weak': 0.7297297297297297}\n",
+      "Epoch 41/50:\n",
+      "Train - Reg Loss: 0.3208, Cls Loss: 0.1951\n",
+      "Train - Correlation: 0.9232, Accuracy: 0.9292\n",
+      "Train - Per-class F1: {'Tight': 0.8914549653579676, 'Medium': 0.8865497076023392, 'Weak': 0.9623115577889447}\n",
+      "Val - Reg Loss: 1.6777, Cls Loss: 1.5731\n",
+      "Val - Correlation: 0.5843, Accuracy: 0.6454\n",
+      "Val - Per-class F1: {'Tight': 0.4842105263157895, 'Medium': 0.46534653465346537, 'Weak': 0.7670588235294118}\n",
+      "Epoch 41/50:\n",
+      "Train - Reg Loss: 0.3208, Cls Loss: 0.1951\n",
+      "Train - Correlation: 0.9232, Accuracy: 0.9292\n",
+      "Train - Per-class F1: {'Tight': 0.8914549653579676, 'Medium': 0.8865497076023392, 'Weak': 0.9623115577889447}\n",
+      "Val - Reg Loss: 1.6777, Cls Loss: 1.5731\n",
+      "Val - Correlation: 0.5843, Accuracy: 0.6454\n",
+      "Val - Per-class F1: {'Tight': 0.4842105263157895, 'Medium': 0.46534653465346537, 'Weak': 0.7670588235294118}\n",
+      "Epoch 42/50:\n",
+      "Train - Reg Loss: 0.3359, Cls Loss: 0.2023\n",
+      "Train - Correlation: 0.9204, Accuracy: 0.9243\n",
+      "Train - Per-class F1: {'Tight': 0.8796296296296297, 'Medium': 0.8751486325802615, 'Weak': 0.9620410703173615}\n",
+      "Val - Reg Loss: 1.7456, Cls Loss: 1.6236\n",
+      "Val - Correlation: 0.5770, Accuracy: 0.6177\n",
+      "Val - Per-class F1: {'Tight': 0.5094339622641509, 'Medium': 0.4434389140271493, 'Weak': 0.7443037974683544}\n",
+      "Epoch 42/50:\n",
+      "Train - Reg Loss: 0.3359, Cls Loss: 0.2023\n",
+      "Train - Correlation: 0.9204, Accuracy: 0.9243\n",
+      "Train - Per-class F1: {'Tight': 0.8796296296296297, 'Medium': 0.8751486325802615, 'Weak': 0.9620410703173615}\n",
+      "Val - Reg Loss: 1.7456, Cls Loss: 1.6236\n",
+      "Val - Correlation: 0.5770, Accuracy: 0.6177\n",
+      "Val - Per-class F1: {'Tight': 0.5094339622641509, 'Medium': 0.4434389140271493, 'Weak': 0.7443037974683544}\n",
+      "Epoch 43/50:\n",
+      "Train - Reg Loss: 0.3033, Cls Loss: 0.1894\n",
+      "Train - Correlation: 0.9244, Accuracy: 0.9229\n",
+      "Train - Per-class F1: {'Tight': 0.8687782805429864, 'Medium': 0.8718562874251496, 'Weak': 0.9644416718652526}\n",
+      "Val - Reg Loss: 1.5880, Cls Loss: 1.4638\n",
+      "Val - Correlation: 0.5978, Accuracy: 0.6399\n",
+      "Val - Per-class F1: {'Tight': 0.5420560747663551, 'Medium': 0.41450777202072536, 'Weak': 0.7677725118483413}\n",
+      "Epoch 43/50:\n",
+      "Train - Reg Loss: 0.3033, Cls Loss: 0.1894\n",
+      "Train - Correlation: 0.9244, Accuracy: 0.9229\n",
+      "Train - Per-class F1: {'Tight': 0.8687782805429864, 'Medium': 0.8718562874251496, 'Weak': 0.9644416718652526}\n",
+      "Val - Reg Loss: 1.5880, Cls Loss: 1.4638\n",
+      "Val - Correlation: 0.5978, Accuracy: 0.6399\n",
+      "Val - Per-class F1: {'Tight': 0.5420560747663551, 'Medium': 0.41450777202072536, 'Weak': 0.7677725118483413}\n",
+      "Epoch 44/50:\n",
+      "Train - Reg Loss: 0.3102, Cls Loss: 0.1910\n",
+      "Train - Correlation: 0.9253, Accuracy: 0.9250\n",
+      "Train - Per-class F1: {'Tight': 0.9041095890410958, 'Medium': 0.8773584905660378, 'Weak': 0.9560853199498118}\n",
+      "Val - Reg Loss: 1.7503, Cls Loss: 1.5762\n",
+      "Val - Correlation: 0.5861, Accuracy: 0.6316\n",
+      "Val - Per-class F1: {'Tight': 0.5094339622641509, 'Medium': 0.4608294930875576, 'Weak': 0.7568922305764411}\n",
+      "Epoch 44/50:\n",
+      "Train - Reg Loss: 0.3102, Cls Loss: 0.1910\n",
+      "Train - Correlation: 0.9253, Accuracy: 0.9250\n",
+      "Train - Per-class F1: {'Tight': 0.9041095890410958, 'Medium': 0.8773584905660378, 'Weak': 0.9560853199498118}\n",
+      "Val - Reg Loss: 1.7503, Cls Loss: 1.5762\n",
+      "Val - Correlation: 0.5861, Accuracy: 0.6316\n",
+      "Val - Per-class F1: {'Tight': 0.5094339622641509, 'Medium': 0.4608294930875576, 'Weak': 0.7568922305764411}\n",
+      "Epoch 45/50:\n",
+      "Train - Reg Loss: 0.3191, Cls Loss: 0.1906\n",
+      "Train - Correlation: 0.9254, Accuracy: 0.9319\n",
+      "Train - Per-class F1: {'Tight': 0.8752834467120182, 'Medium': 0.8904761904761904, 'Weak': 0.9693558474046279}\n",
+      "Val - Reg Loss: 1.6623, Cls Loss: 1.4726\n",
+      "Val - Correlation: 0.5933, Accuracy: 0.6371\n",
+      "Val - Per-class F1: {'Tight': 0.5052631578947369, 'Medium': 0.5041322314049587, 'Weak': 0.7532467532467533}\n",
+      "Epoch 45/50:\n",
+      "Train - Reg Loss: 0.3191, Cls Loss: 0.1906\n",
+      "Train - Correlation: 0.9254, Accuracy: 0.9319\n",
+      "Train - Per-class F1: {'Tight': 0.8752834467120182, 'Medium': 0.8904761904761904, 'Weak': 0.9693558474046279}\n",
+      "Val - Reg Loss: 1.6623, Cls Loss: 1.4726\n",
+      "Val - Correlation: 0.5933, Accuracy: 0.6371\n",
+      "Val - Per-class F1: {'Tight': 0.5052631578947369, 'Medium': 0.5041322314049587, 'Weak': 0.7532467532467533}\n",
+      "Epoch 46/50:\n",
+      "Train - Reg Loss: 0.2731, Cls Loss: 0.1769\n",
+      "Train - Correlation: 0.9332, Accuracy: 0.9340\n",
+      "Train - Per-class F1: {'Tight': 0.8857808857808858, 'Medium': 0.8943089430894309, 'Weak': 0.9685534591194969}\n",
+      "Val - Reg Loss: 1.8729, Cls Loss: 1.7549\n",
+      "Val - Correlation: 0.5643, Accuracy: 0.6122\n",
+      "Val - Per-class F1: {'Tight': 0.5309734513274337, 'Medium': 0.3838383838383838, 'Weak': 0.7445255474452555}\n",
+      "Epoch 46/50:\n",
+      "Train - Reg Loss: 0.2731, Cls Loss: 0.1769\n",
+      "Train - Correlation: 0.9332, Accuracy: 0.9340\n",
+      "Train - Per-class F1: {'Tight': 0.8857808857808858, 'Medium': 0.8943089430894309, 'Weak': 0.9685534591194969}\n",
+      "Val - Reg Loss: 1.8729, Cls Loss: 1.7549\n",
+      "Val - Correlation: 0.5643, Accuracy: 0.6122\n",
+      "Val - Per-class F1: {'Tight': 0.5309734513274337, 'Medium': 0.3838383838383838, 'Weak': 0.7445255474452555}\n",
+      "Epoch 47/50:\n",
+      "Train - Reg Loss: 0.2866, Cls Loss: 0.1625\n",
+      "Train - Correlation: 0.9313, Accuracy: 0.9410\n",
+      "Train - Per-class F1: {'Tight': 0.9178082191780822, 'Medium': 0.901072705601907, 'Weak': 0.9681846537741734}\n",
+      "Val - Reg Loss: 1.7201, Cls Loss: 1.6087\n",
+      "Val - Correlation: 0.5733, Accuracy: 0.6039\n",
+      "Val - Per-class F1: {'Tight': 0.41304347826086957, 'Medium': 0.48627450980392156, 'Weak': 0.7306666666666667}\n",
+      "Epoch 47/50:\n",
+      "Train - Reg Loss: 0.2866, Cls Loss: 0.1625\n",
+      "Train - Correlation: 0.9313, Accuracy: 0.9410\n",
+      "Train - Per-class F1: {'Tight': 0.9178082191780822, 'Medium': 0.901072705601907, 'Weak': 0.9681846537741734}\n",
+      "Val - Reg Loss: 1.7201, Cls Loss: 1.6087\n",
+      "Val - Correlation: 0.5733, Accuracy: 0.6039\n",
+      "Val - Per-class F1: {'Tight': 0.41304347826086957, 'Medium': 0.48627450980392156, 'Weak': 0.7306666666666667}\n",
+      "Epoch 48/50:\n",
+      "Train - Reg Loss: 0.2892, Cls Loss: 0.1683\n",
+      "Train - Correlation: 0.9337, Accuracy: 0.9382\n",
+      "Train - Per-class F1: {'Tight': 0.8939051918735892, 'Medium': 0.8978622327790974, 'Weak': 0.9717868338557993}\n",
+      "Val - Reg Loss: 1.7904, Cls Loss: 1.6392\n",
+      "Val - Correlation: 0.5908, Accuracy: 0.6260\n",
+      "Val - Per-class F1: {'Tight': 0.5, 'Medium': 0.5, 'Weak': 0.7407407407407407}\n",
+      "Epoch 48/50:\n",
+      "Train - Reg Loss: 0.2892, Cls Loss: 0.1683\n",
+      "Train - Correlation: 0.9337, Accuracy: 0.9382\n",
+      "Train - Per-class F1: {'Tight': 0.8939051918735892, 'Medium': 0.8978622327790974, 'Weak': 0.9717868338557993}\n",
+      "Val - Reg Loss: 1.7904, Cls Loss: 1.6392\n",
+      "Val - Correlation: 0.5908, Accuracy: 0.6260\n",
+      "Val - Per-class F1: {'Tight': 0.5, 'Medium': 0.5, 'Weak': 0.7407407407407407}\n",
+      "Epoch 49/50:\n",
+      "Train - Reg Loss: 0.2674, Cls Loss: 0.1425\n",
+      "Train - Correlation: 0.9312, Accuracy: 0.9403\n",
+      "Train - Per-class F1: {'Tight': 0.9016018306636155, 'Medium': 0.8994082840236687, 'Weak': 0.9724655819774718}\n",
+      "Val - Reg Loss: 1.8287, Cls Loss: 1.6693\n",
+      "Val - Correlation: 0.5864, Accuracy: 0.6316\n",
+      "Val - Per-class F1: {'Tight': 0.5052631578947369, 'Medium': 0.4485981308411215, 'Weak': 0.7554479418886199}\n",
+      "Epoch 49/50:\n",
+      "Train - Reg Loss: 0.2674, Cls Loss: 0.1425\n",
+      "Train - Correlation: 0.9312, Accuracy: 0.9403\n",
+      "Train - Per-class F1: {'Tight': 0.9016018306636155, 'Medium': 0.8994082840236687, 'Weak': 0.9724655819774718}\n",
+      "Val - Reg Loss: 1.8287, Cls Loss: 1.6693\n",
+      "Val - Correlation: 0.5864, Accuracy: 0.6316\n",
+      "Val - Per-class F1: {'Tight': 0.5052631578947369, 'Medium': 0.4485981308411215, 'Weak': 0.7554479418886199}\n",
+      "Epoch 50/50:\n",
+      "Train - Reg Loss: 0.2770, Cls Loss: 0.1414\n",
+      "Train - Correlation: 0.9386, Accuracy: 0.9403\n",
+      "Train - Per-class F1: {'Tight': 0.8959276018099548, 'Medium': 0.9025270758122743, 'Weak': 0.9719975108898569}\n",
+      "Val - Reg Loss: 1.6301, Cls Loss: 1.8242\n",
+      "Val - Correlation: 0.5823, Accuracy: 0.6454\n",
+      "Val - Per-class F1: {'Tight': 0.48, 'Medium': 0.5, 'Weak': 0.7688442211055276}\n",
+      "Epoch 50/50:\n",
+      "Train - Reg Loss: 0.2770, Cls Loss: 0.1414\n",
+      "Train - Correlation: 0.9386, Accuracy: 0.9403\n",
+      "Train - Per-class F1: {'Tight': 0.8959276018099548, 'Medium': 0.9025270758122743, 'Weak': 0.9719975108898569}\n",
+      "Val - Reg Loss: 1.6301, Cls Loss: 1.8242\n",
+      "Val - Correlation: 0.5823, Accuracy: 0.6454\n",
+      "Val - Per-class F1: {'Tight': 0.48, 'Medium': 0.5, 'Weak': 0.7688442211055276}\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 1000x800 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 1000x800 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\n",
+      "Training complete! Best model from epoch 21 with correlation: 0.6109\n",
+      "Model saved as: best_model.pt\n",
+      "Correlation plots saved as: best_model_val_correlation.png, best_model_train_correlation.png\n"
+     ]
+    }
+   ],
+   "source": [
+    "model = ImprovedBindingPredictor(\n",
+    "    esm_dim=1280,\n",
+    "    smiles_dim=768,\n",
+    "    hidden_dim=512,\n",
+    "    n_heads=8,\n",
+    "    n_layers=3,\n",
+    "    dropout=0.1\n",
+    ").to(device)\n",
+    "\n",
+    "criterion_reg = nn.MSELoss()\n",
+    "criterion_cls = nn.CrossEntropyLoss()\n",
+    "\n",
+    "# Optimizer\n",
+    "optimizer = torch.optim.AdamW(model.parameters(), lr=1e-4, weight_decay=0.01)\n",
+    "\n",
+    "# Track best model\n",
+    "best_val_correlation = -1.0\n",
+    "best_epoch = 0\n",
+    "best_train_results = None\n",
+    "best_val_results = None\n",
+    "\n",
+    "num_epochs = 50\n",
+    "for epoch in range(num_epochs):\n",
+    "    train_results = train_epoch(model, train_loader, optimizer, criterion_reg, criterion_cls, device)\n",
+    "    val_results = validate_epoch(model, val_loader, criterion_reg, criterion_cls, device)\n",
+    "    \n",
+    "    print(f'Epoch {epoch+1}/{num_epochs}:')\n",
+    "    print(f'Train - Reg Loss: {train_results[\"reg_loss\"]:.4f}, Cls Loss: {train_results[\"cls_loss\"]:.4f}')\n",
+    "    print(f'Train - Correlation: {train_results[\"correlation\"]:.4f}, Accuracy: {train_results[\"accuracy\"]:.4f}')\n",
+    "    print('Train - Per-class F1:', train_results[\"per_class_f1\"])\n",
+    "    print(f'Val - Reg Loss: {val_results[\"reg_loss\"]:.4f}, Cls Loss: {val_results[\"cls_loss\"]:.4f}')\n",
+    "    print(f'Val - Correlation: {val_results[\"correlation\"]:.4f}, Accuracy: {val_results[\"accuracy\"]:.4f}')\n",
+    "    print('Val - Per-class F1:', val_results[\"per_class_f1\"])\n",
+    "    \n",
+    "    # Save best model based on validation correlation\n",
+    "    if val_results[\"correlation\"] > best_val_correlation:\n",
+    "        best_val_correlation = val_results[\"correlation\"]\n",
+    "        best_epoch = epoch + 1\n",
+    "        best_train_results = train_results\n",
+    "        best_val_results = val_results\n",
+    "        \n",
+    "        torch.save({\n",
+    "            'epoch': epoch + 1,\n",
+    "            'model_state_dict': model.state_dict(),\n",
+    "            'optimizer_state_dict': optimizer.state_dict(),\n",
+    "            'val_correlation': val_results[\"correlation\"],\n",
+    "            'val_accuracy': val_results[\"accuracy\"],\n",
+    "            'val_f1': val_results[\"f1_score\"],\n",
+    "            'train_correlation': train_results[\"correlation\"],\n",
+    "        }, 'best_model.pt')\n",
+    "        print(f'✓ Saved best model at epoch {epoch+1} with validation correlation: {val_results[\"correlation\"]:.4f}')\n",
+    "        \n",
+    "        # Save correlation plots for best model\n",
+    "        plt.figure(figsize=(10, 8))\n",
+    "        plt.scatter(val_results['true_values'], val_results['reg_predictions'], alpha=0.5)\n",
+    "        val_true = np.array(val_results['true_values'])\n",
+    "        val_pred = np.array(val_results['reg_predictions'])\n",
+    "        min_val = min(val_true.min(), val_pred.min())\n",
+    "        max_val = max(val_true.max(), val_pred.max())\n",
+    "        plt.plot([min_val, max_val], [min_val, max_val], 'r--', label='Perfect Correlation')\n",
+    "        plt.axvline(x=7.5, color='g', linestyle='--', alpha=0.5)\n",
+    "        plt.axvline(x=6.0, color='b', linestyle='--', alpha=0.5)\n",
+    "        plt.xlabel('True Affinity (-log scale)')\n",
+    "        plt.ylabel('Predicted Affinity (-log scale)')\n",
+    "        plt.title(f'Best Model Validation Set (Epoch {epoch+1})\\nSpearman Correlation: {val_results[\"correlation\"]:.3f}')\n",
+    "        plt.grid(True, alpha=0.3)\n",
+    "        plt.tight_layout()\n",
+    "        plt.savefig('best_model_val_correlation.png', dpi=300, bbox_inches='tight')\n",
+    "        plt.close()\n",
+    "        \n",
+    "        plt.figure(figsize=(10, 8))\n",
+    "        plt.scatter(train_results['true_values'], train_results['reg_predictions'], alpha=0.5)\n",
+    "        train_true = np.array(train_results['true_values'])\n",
+    "        train_pred = np.array(train_results['reg_predictions'])\n",
+    "        min_val = min(train_true.min(), train_pred.min())\n",
+    "        max_val = max(train_true.max(), train_pred.max())\n",
+    "        plt.plot([min_val, max_val], [min_val, max_val], 'r--', label='Perfect Correlation')\n",
+    "        plt.axvline(x=7.5, color='g', linestyle='--', alpha=0.5)\n",
+    "        plt.axvline(x=6.0, color='b', linestyle='--', alpha=0.5)\n",
+    "        plt.xlabel('True Affinity (-log scale)')\n",
+    "        plt.ylabel('Predicted Affinity (-log scale)')\n",
+    "        plt.title(f'Best Model Training Set (Epoch {epoch+1})\\nSpearman Correlation: {train_results[\"correlation\"]:.3f}')\n",
+    "        plt.grid(True, alpha=0.3)\n",
+    "        plt.tight_layout()\n",
+    "        plt.savefig('best_model_train_correlation.png', dpi=300, bbox_inches='tight')\n",
+    "        plt.close()\n",
+    "        \n",
+    "        print(f'✓ Saved correlation plots: best_model_val_correlation.png, best_model_train_correlation.png')\n",
+    "    \n",
+    "    if epoch == num_epochs - 1:\n",
+    "        plot_correlation_with_classes(\n",
+    "            val_results['reg_predictions'], \n",
+    "            val_results['true_values'],\n",
+    "            epoch + 1, \n",
+    "            \"Validation\"\n",
+    "        )\n",
+    "        plot_correlation_with_classes(\n",
+    "            train_results['reg_predictions'], \n",
+    "            train_results['true_values'],\n",
+    "            epoch + 1, \n",
+    "            \"Training\"\n",
+    "        )\n",
+    "\n",
+    "print(f'\\nTraining complete! Best model from epoch {best_epoch} with correlation: {best_val_correlation:.4f}')\n",
+    "print(f'Model saved as: best_model.pt')\n",
+    "print(f'Correlation plots saved as: best_model_val_correlation.png, best_model_train_correlation.png')\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "tr2d2-pep",
+   "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": "3.10.18"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}