"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"import matplotlib.pyplot as plt\n",
"\n",
"# Plot histogram of predictions\n",
"plt.hist(train_preds, bins=20)\n",
"plt.title('Train predictions')\n",
"plt.grid(axis='y', alpha=0.5)\n",
"plt.show()\n",
"\n",
"plt.hist(val_preds, bins=20)\n",
"plt.title('Validation predictions')\n",
"plt.grid(axis='y', alpha=0.5)\n",
"plt.show()\n",
"\n",
"plt.hist(test_preds, bins=20)\n",
"plt.title('Test predictions')\n",
"plt.grid(axis='y', alpha=0.5)\n",
"plt.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"TODO: Which data are in the most unsure middle region?"
]
},
{
"cell_type": "code",
"execution_count": 94,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"False 130\n",
"True 65\n",
"Name: Active, dtype: int64"
]
},
"execution_count": 94,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Get the test set predictions which are very confident\n",
"def is_active(p, pos_threshold=0.90, neg_threshold=0.005):\n",
" if p > pos_threshold:\n",
" return True\n",
" elif p < neg_threshold:\n",
" return False\n",
" else:\n",
" return np.nan\n",
"\n",
"\n",
"test_df['Active'] = [\n",
" is_active(p, pos_threshold=0.97, neg_threshold=0.001) for p in test_preds]\n",
"test_df[~test_df['Active'].isna()]['Active'].value_counts()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Cross-Validation Training"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"TODO: Cross-Val\n",
"\n",
"* Random split\n",
"* Least Tanimoto dist PROTAC-wise\n",
"* POI-wise: some POIs never in both splits\n",
"\n",
"TODO: Separate test set??"
]
},
{
"cell_type": "code",
"execution_count": 22,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"Global seed set to 42\n"
]
}
],
"source": [
"# Seed everything in pytorch lightning\n",
"pl.seed_everything(42)\n",
"\n",
"\n",
"def train_model(train_df, val_df, disabled_embeddings=[]):\n",
" train_ds = PROTAC_Dataset(\n",
" train_df,\n",
" protein_embeddings,\n",
" cell2embedding,\n",
" smiles2fp,\n",
" use_smote=True,\n",
" )\n",
" val_ds = PROTAC_Dataset(\n",
" val_df,\n",
" protein_embeddings,\n",
" cell2embedding,\n",
" smiles2fp,\n",
" )\n",
" logger = pl.loggers.TensorBoardLogger(\n",
" save_dir='../logs',\n",
" name='protac',\n",
" )\n",
" callbacks = [\n",
" pl.callbacks.EarlyStopping(\n",
" monitor='train_loss',\n",
" patience=10,\n",
" mode='max',\n",
" verbose=True,\n",
" ),\n",
" # pl.callbacks.ModelCheckpoint(\n",
" # monitor='val_acc',\n",
" # mode='max',\n",
" # verbose=True,\n",
" # filename='{epoch}-{val_metrics_opt_score:.4f}',\n",
" # ),\n",
" ]\n",
" # Define Trainer\n",
" trainer = pl.Trainer(\n",
" logger=logger,\n",
" callbacks=callbacks,\n",
" max_epochs=50,\n",
" enable_checkpointing=False,\n",
" )\n",
" model = PROTAC_Model(\n",
" hidden_dim=768,\n",
" smiles_emb_dim=1024,\n",
" poi_emb_dim=1024,\n",
" e3_emb_dim=1024,\n",
" cell_emb_dim=768,\n",
" batch_size=8,\n",
" learning_rate=2e-5,\n",
" train_dataset=train_ds,\n",
" val_dataset=val_ds,\n",
" disabled_embeddings=disabled_embeddings,\n",
" )\n",
" trainer.fit(model)\n",
" val_dl = DataLoader(val_ds, batch_size=32, shuffle=False)\n",
" return model, trainer, trainer.validate(model, val_dl)[0]"
]
},
{
"cell_type": "code",
"execution_count": 130,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"GPU available: True (cuda), used: True\n",
"TPU available: False, using: 0 TPU cores\n",
"IPU available: False, using: 0 IPUs\n",
"HPU available: False, using: 0 HPUs\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n",
"\n",
" | Name | Type | Params\n",
"------------------------------------------\n",
"0 | poi_emb | Linear | 787 K \n",
"1 | e3_emb | Linear | 787 K \n",
"2 | cell_emb | Linear | 590 K \n",
"3 | smiles_emb | Linear | 787 K \n",
"4 | fc1 | Linear | 2.4 M \n",
"5 | fc2 | Linear | 590 K \n",
"6 | fc3 | Linear | 769 \n",
"7 | dropout | Dropout | 0 \n",
"8 | metrics | ModuleDict | 0 \n",
"------------------------------------------\n",
"5.9 M Trainable params\n",
"0 Non-trainable params\n",
"5.9 M Total params\n",
"23.614 Total estimated model params size (MB)\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "7bad4f7bc2f641bdb5387ae2bbb95222",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Sanity Checking: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "8f412941c8884751bf5dff29683d5d4c",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Training: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "054ad46139c74d1cbaf9cb92b6ed312c",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Metric train_loss improved. New best score: 0.678\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "53c8e1d6d23c45b5aee636c8c65f780f",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "9be4ed2a0b6e42978f36defb5cfc7d4b",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "c1e90209b8624bb3aa551ee4be1ac28e",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "e1cc66975cb04cafbe498732146240c6",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "8f18056dcfc74b65994c6dd7896c3922",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "8a6a032f2c824e0097cb6a184ab01c05",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "53b46cb12bad482bbb7069027d97d7ef",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "d95edcb4257c4ce0b89901cc219511ee",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "1444787b48a04c97999f0ee6a918fdaf",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "84e125446343491cb221985ff5556838",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Monitored metric train_loss did not improve in the last 10 records. Best score: 0.678. Signaling Trainer to stop.\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "2e33e9834aa042ec8e1fdb7ad1741023",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃ Runningstage.validating ┃ ┃\n",
"┃ metric ┃ DataLoader 0 ┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│ val_acc │ 0.8623853325843811 │\n",
"│ val_f1_score │ 0.7368420958518982 │\n",
"│ val_hp_metric │ 0.8623853325843811 │\n",
"│ val_loss │ 0.3256032466888428 │\n",
"│ val_opt_score │ 1.5992274284362793 │\n",
"│ val_precision │ 0.8399999737739563 │\n",
"│ val_recall │ 0.65625 │\n",
"│ val_roc_auc │ 0.9358766078948975 │\n",
"└───────────────────────────┴───────────────────────────┘\n",
"\n"
],
"text/plain": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃\u001b[1m \u001b[0m\u001b[1m Runningstage.validating \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m┃\n",
"┃\u001b[1m \u001b[0m\u001b[1m metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│\u001b[36m \u001b[0m\u001b[36m val_acc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8623853325843811 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_f1_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.7368420958518982 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_hp_metric \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8623853325843811 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.3256032466888428 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_opt_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 1.5992274284362793 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_precision \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8399999737739563 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_recall \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.65625 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_roc_auc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.9358766078948975 \u001b[0m\u001b[35m \u001b[0m│\n",
"└───────────────────────────┴───────────────────────────┘\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"GPU available: True (cuda), used: True\n",
"TPU available: False, using: 0 TPU cores\n",
"IPU available: False, using: 0 IPUs\n",
"HPU available: False, using: 0 HPUs\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n",
"\n",
" | Name | Type | Params\n",
"------------------------------------------\n",
"0 | poi_emb | Linear | 787 K \n",
"1 | e3_emb | Linear | 787 K \n",
"2 | cell_emb | Linear | 590 K \n",
"3 | smiles_emb | Linear | 787 K \n",
"4 | fc1 | Linear | 2.4 M \n",
"5 | fc2 | Linear | 590 K \n",
"6 | fc3 | Linear | 769 \n",
"7 | dropout | Dropout | 0 \n",
"8 | metrics | ModuleDict | 0 \n",
"------------------------------------------\n",
"5.9 M Trainable params\n",
"0 Non-trainable params\n",
"5.9 M Total params\n",
"23.614 Total estimated model params size (MB)\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "e0bc584883de471faa2f05dfd7d9cbcd",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Sanity Checking: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "bfe4ea0982404048a16c4e42399565d5",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Training: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "c2f54315a0cd465db5497fa0c976c43c",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Metric train_loss improved. New best score: 0.681\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "28915d5ad1764d5e9c95479f759ae42f",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "d1e2da9f2cad45b1a5df171a37147abe",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "5def76e6dbbc4cccb184890412bf68b5",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "10f84836acf046d8a1c099f3a2560ef9",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "c899c0afd6164f3cbe91b6f886f1d842",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "b1f5699bcff045c8bfda04d618670094",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "080d55d9b2c3473da2a7e1ef8538f8af",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "bdcf21d1be6c450cb0626ba0d36897fa",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "3fe4cddc678b49bcae2d193c9a409d5f",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "8b043af809a44d289cd5e34dbe4b051a",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Monitored metric train_loss did not improve in the last 10 records. Best score: 0.681. Signaling Trainer to stop.\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "513f417ab6464668be9fe305507ba3d6",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃ Runningstage.validating ┃ ┃\n",
"┃ metric ┃ DataLoader 0 ┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│ val_acc │ 0.8440366983413696 │\n",
"│ val_f1_score │ 0.7301587462425232 │\n",
"│ val_hp_metric │ 0.8440366983413696 │\n",
"│ val_loss │ 0.4012793004512787 │\n",
"│ val_opt_score │ 1.574195384979248 │\n",
"│ val_precision │ 0.6571428775787354 │\n",
"│ val_recall │ 0.8214285969734192 │\n",
"│ val_roc_auc │ 0.9135802388191223 │\n",
"└───────────────────────────┴───────────────────────────┘\n",
"\n"
],
"text/plain": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃\u001b[1m \u001b[0m\u001b[1m Runningstage.validating \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m┃\n",
"┃\u001b[1m \u001b[0m\u001b[1m metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│\u001b[36m \u001b[0m\u001b[36m val_acc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8440366983413696 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_f1_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.7301587462425232 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_hp_metric \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8440366983413696 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.4012793004512787 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_opt_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 1.574195384979248 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_precision \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.6571428775787354 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_recall \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8214285969734192 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_roc_auc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.9135802388191223 \u001b[0m\u001b[35m \u001b[0m│\n",
"└───────────────────────────┴───────────────────────────┘\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"GPU available: True (cuda), used: True\n",
"TPU available: False, using: 0 TPU cores\n",
"IPU available: False, using: 0 IPUs\n",
"HPU available: False, using: 0 HPUs\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n",
"\n",
" | Name | Type | Params\n",
"------------------------------------------\n",
"0 | poi_emb | Linear | 787 K \n",
"1 | e3_emb | Linear | 787 K \n",
"2 | cell_emb | Linear | 590 K \n",
"3 | smiles_emb | Linear | 787 K \n",
"4 | fc1 | Linear | 2.4 M \n",
"5 | fc2 | Linear | 590 K \n",
"6 | fc3 | Linear | 769 \n",
"7 | dropout | Dropout | 0 \n",
"8 | metrics | ModuleDict | 0 \n",
"------------------------------------------\n",
"5.9 M Trainable params\n",
"0 Non-trainable params\n",
"5.9 M Total params\n",
"23.614 Total estimated model params size (MB)\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "4a2456436b8943f2af3af61adeab3a22",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Sanity Checking: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "6c507902997f4d5494533234785a3065",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Training: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "96541b344e84483ca99ff16494e3f77c",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Metric train_loss improved. New best score: 0.678\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "f12c58aed8e94865b2873ca0dbb9a5ce",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "523d9c533ff54d7dbb7b7df49e492880",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "7893269cb24a4f918b7d76d54e76aa43",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "96632302871840a6bf8b8677b39c52e4",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "0ee7de4ab69b4e7e9ebf94fa048973a1",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "6a98c7f38c074a05982dea7af9fb1e5a",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "3188161e3de34a79acea0688c8873eea",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "a2fbb1e75b444de3b72d97e71424c5a2",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "6fa8350fe63240fb8e718a1c3597b630",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "0d5ca18adae94cba9d8407ff0d988fc8",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Monitored metric train_loss did not improve in the last 10 records. Best score: 0.678. Signaling Trainer to stop.\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "827d4d9de83943628ac753c34af0ed82",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃ Runningstage.validating ┃ ┃\n",
"┃ metric ┃ DataLoader 0 ┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│ val_acc │ 0.8256880640983582 │\n",
"│ val_f1_score │ 0.6779661178588867 │\n",
"│ val_hp_metric │ 0.8256880640983582 │\n",
"│ val_loss │ 0.46647822856903076 │\n",
"│ val_opt_score │ 1.5036542415618896 │\n",
"│ val_precision │ 0.7142857313156128 │\n",
"│ val_recall │ 0.6451612710952759 │\n",
"│ val_roc_auc │ 0.8873035907745361 │\n",
"└───────────────────────────┴───────────────────────────┘\n",
"\n"
],
"text/plain": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃\u001b[1m \u001b[0m\u001b[1m Runningstage.validating \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m┃\n",
"┃\u001b[1m \u001b[0m\u001b[1m metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│\u001b[36m \u001b[0m\u001b[36m val_acc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8256880640983582 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_f1_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.6779661178588867 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_hp_metric \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8256880640983582 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.46647822856903076 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_opt_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 1.5036542415618896 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_precision \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.7142857313156128 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_recall \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.6451612710952759 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_roc_auc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8873035907745361 \u001b[0m\u001b[35m \u001b[0m│\n",
"└───────────────────────────┴───────────────────────────┘\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"GPU available: True (cuda), used: True\n",
"TPU available: False, using: 0 TPU cores\n",
"IPU available: False, using: 0 IPUs\n",
"HPU available: False, using: 0 HPUs\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n",
"\n",
" | Name | Type | Params\n",
"------------------------------------------\n",
"0 | poi_emb | Linear | 787 K \n",
"1 | e3_emb | Linear | 787 K \n",
"2 | cell_emb | Linear | 590 K \n",
"3 | smiles_emb | Linear | 787 K \n",
"4 | fc1 | Linear | 2.4 M \n",
"5 | fc2 | Linear | 590 K \n",
"6 | fc3 | Linear | 769 \n",
"7 | dropout | Dropout | 0 \n",
"8 | metrics | ModuleDict | 0 \n",
"------------------------------------------\n",
"5.9 M Trainable params\n",
"0 Non-trainable params\n",
"5.9 M Total params\n",
"23.614 Total estimated model params size (MB)\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "327f63966297483a863a97ced1835d45",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Sanity Checking: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "d234afb327b84bc3adbcfd1b1a235eb2",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Training: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "14b0ffa369c94482b776a4017ad6349e",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Metric train_loss improved. New best score: 0.679\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "3e7f13a804c44f1b8a1454b479f99d26",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "ac6a9ba1af0f44f3a990cec17ef90c23",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "79272ce2d6af40648573b8d3893384aa",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "e8cf9c4cbf2445fd976a78e815147867",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "7f585792db9242abbc25bb3e51424b77",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "760a8c615b924c3183633db97d0e39b2",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "0f0bb6ccab7a41afbe69fa4900dea4e6",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "1664e40c11f9446094e837a85c64b00e",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "24db82b937ad4eb6a9ddb5bec9ee88fd",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "adb60706c5c44d55816d34cc1e963450",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Monitored metric train_loss did not improve in the last 10 records. Best score: 0.679. Signaling Trainer to stop.\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "4c534a708f5e40d9b11a7278531856f3",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃ Runningstage.validating ┃ ┃\n",
"┃ metric ┃ DataLoader 0 ┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│ val_acc │ 0.8623853325843811 │\n",
"│ val_f1_score │ 0.7368420958518982 │\n",
"│ val_hp_metric │ 0.8623853325843811 │\n",
"│ val_loss │ 0.4182129204273224 │\n",
"│ val_opt_score │ 1.5992274284362793 │\n",
"│ val_precision │ 0.6176470518112183 │\n",
"│ val_recall │ 0.9130434989929199 │\n",
"│ val_roc_auc │ 0.9155712723731995 │\n",
"└───────────────────────────┴───────────────────────────┘\n",
"\n"
],
"text/plain": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃\u001b[1m \u001b[0m\u001b[1m Runningstage.validating \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m┃\n",
"┃\u001b[1m \u001b[0m\u001b[1m metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│\u001b[36m \u001b[0m\u001b[36m val_acc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8623853325843811 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_f1_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.7368420958518982 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_hp_metric \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8623853325843811 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.4182129204273224 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_opt_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 1.5992274284362793 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_precision \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.6176470518112183 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_recall \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.9130434989929199 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_roc_auc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.9155712723731995 \u001b[0m\u001b[35m \u001b[0m│\n",
"└───────────────────────────┴───────────────────────────┘\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"GPU available: True (cuda), used: True\n",
"TPU available: False, using: 0 TPU cores\n",
"IPU available: False, using: 0 IPUs\n",
"HPU available: False, using: 0 HPUs\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n",
"\n",
" | Name | Type | Params\n",
"------------------------------------------\n",
"0 | poi_emb | Linear | 787 K \n",
"1 | e3_emb | Linear | 787 K \n",
"2 | cell_emb | Linear | 590 K \n",
"3 | smiles_emb | Linear | 787 K \n",
"4 | fc1 | Linear | 2.4 M \n",
"5 | fc2 | Linear | 590 K \n",
"6 | fc3 | Linear | 769 \n",
"7 | dropout | Dropout | 0 \n",
"8 | metrics | ModuleDict | 0 \n",
"------------------------------------------\n",
"5.9 M Trainable params\n",
"0 Non-trainable params\n",
"5.9 M Total params\n",
"23.614 Total estimated model params size (MB)\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "b66dae7a34384a15a05aafced6d20f45",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Sanity Checking: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "ec2922e3ffc3425aad4477559ebd9c82",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Training: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "f01a30977b6840c8959a8d04a6ccd391",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Metric train_loss improved. New best score: 0.677\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "cc5e4ca4a6ed4509ac665190195a5bb5",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "c938ba47ca8d412384202d36dafc424f",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "9b462160087a47039bb22bbff87f8e31",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "95c4c6d08f6a4693b25099aa44b23937",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "90e0e49a4d4243fc8281b732adf23c35",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "d01aa404fec14656a5512e93d821b116",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "36605d552b18480ab3e27b4e7bf864d1",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "36f6b6dfd6cb4d68b711f25ab0326c2d",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "bc0069acbbbf473dbde3b0e9298e50d0",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "3cb34e255b9645f79cf79efcf87bc849",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Monitored metric train_loss did not improve in the last 10 records. Best score: 0.677. Signaling Trainer to stop.\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "60eba19b79e5474ca92358cef6569759",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃ Runningstage.validating ┃ ┃\n",
"┃ metric ┃ DataLoader 0 ┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│ val_acc │ 0.8348624110221863 │\n",
"│ val_f1_score │ 0.7272727489471436 │\n",
"│ val_hp_metric │ 0.8348624110221863 │\n",
"│ val_loss │ 0.40080317854881287 │\n",
"│ val_opt_score │ 1.5621352195739746 │\n",
"│ val_precision │ 0.6486486196517944 │\n",
"│ val_recall │ 0.8275862336158752 │\n",
"│ val_roc_auc │ 0.9105603098869324 │\n",
"└───────────────────────────┴───────────────────────────┘\n",
"\n"
],
"text/plain": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃\u001b[1m \u001b[0m\u001b[1m Runningstage.validating \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m┃\n",
"┃\u001b[1m \u001b[0m\u001b[1m metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│\u001b[36m \u001b[0m\u001b[36m val_acc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8348624110221863 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_f1_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.7272727489471436 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_hp_metric \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8348624110221863 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.40080317854881287 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_opt_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 1.5621352195739746 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_precision \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.6486486196517944 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_recall \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8275862336158752 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_roc_auc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.9105603098869324 \u001b[0m\u001b[35m \u001b[0m│\n",
"└───────────────────────────┴───────────────────────────┘\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"GPU available: True (cuda), used: True\n",
"TPU available: False, using: 0 TPU cores\n",
"IPU available: False, using: 0 IPUs\n",
"HPU available: False, using: 0 HPUs\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n",
"\n",
" | Name | Type | Params\n",
"------------------------------------------\n",
"0 | poi_emb | Linear | 787 K \n",
"1 | e3_emb | Linear | 787 K \n",
"2 | cell_emb | Linear | 590 K \n",
"3 | smiles_emb | Linear | 787 K \n",
"4 | fc1 | Linear | 2.4 M \n",
"5 | fc2 | Linear | 590 K \n",
"6 | fc3 | Linear | 769 \n",
"7 | dropout | Dropout | 0 \n",
"8 | metrics | ModuleDict | 0 \n",
"------------------------------------------\n",
"5.9 M Trainable params\n",
"0 Non-trainable params\n",
"5.9 M Total params\n",
"23.614 Total estimated model params size (MB)\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "d3736fc2b7ff42e390db860d403b3422",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Sanity Checking: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "96672514a47645b58153a27ee954dbcd",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Training: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "c41c5959d6b243e98dfdece816744c12",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Metric train_loss improved. New best score: 0.678\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "6064c50640db40d7867e0da35c46cef1",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "316aa12385634593a7fbaa68a6237aa0",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "4b1ef6614bda47c28dada91a20e32c87",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "7ae30a51bb1244dcbef64a72b126bd9b",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "16714eb2f3404282a90c70a0572d60d3",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "865878509cd74f1baeafc651352b7a70",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "f26a9a5869454698998f214fdd5e8c70",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "9bb292f1c669465d848fd7d1fcfc0ea8",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "ba3bb9a4ad414143a6b87fce0f359015",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "422e166a069445a68e0cd3d383a4ee08",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Monitored metric train_loss did not improve in the last 10 records. Best score: 0.678. Signaling Trainer to stop.\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "cd1ba74db7884e55916ce19f79400044",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃ Runningstage.validating ┃ ┃\n",
"┃ metric ┃ DataLoader 0 ┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│ val_acc │ 0.8256880640983582 │\n",
"│ val_f1_score │ 0.7076923251152039 │\n",
"│ val_hp_metric │ 0.8256880640983582 │\n",
"│ val_loss │ 0.4778938293457031 │\n",
"│ val_opt_score │ 1.533380389213562 │\n",
"│ val_precision │ 0.6216216087341309 │\n",
"│ val_recall │ 0.8214285969734192 │\n",
"│ val_roc_auc │ 0.899470865726471 │\n",
"└───────────────────────────┴───────────────────────────┘\n",
"\n"
],
"text/plain": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃\u001b[1m \u001b[0m\u001b[1m Runningstage.validating \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m┃\n",
"┃\u001b[1m \u001b[0m\u001b[1m metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│\u001b[36m \u001b[0m\u001b[36m val_acc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8256880640983582 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_f1_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.7076923251152039 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_hp_metric \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8256880640983582 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.4778938293457031 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_opt_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 1.533380389213562 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_precision \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.6216216087341309 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_recall \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8214285969734192 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_roc_auc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.899470865726471 \u001b[0m\u001b[35m \u001b[0m│\n",
"└───────────────────────────┴───────────────────────────┘\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"GPU available: True (cuda), used: True\n",
"TPU available: False, using: 0 TPU cores\n",
"IPU available: False, using: 0 IPUs\n",
"HPU available: False, using: 0 HPUs\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n",
"\n",
" | Name | Type | Params\n",
"------------------------------------------\n",
"0 | poi_emb | Linear | 787 K \n",
"1 | e3_emb | Linear | 787 K \n",
"2 | cell_emb | Linear | 590 K \n",
"3 | smiles_emb | Linear | 787 K \n",
"4 | fc1 | Linear | 2.4 M \n",
"5 | fc2 | Linear | 590 K \n",
"6 | fc3 | Linear | 769 \n",
"7 | dropout | Dropout | 0 \n",
"8 | metrics | ModuleDict | 0 \n",
"------------------------------------------\n",
"5.9 M Trainable params\n",
"0 Non-trainable params\n",
"5.9 M Total params\n",
"23.614 Total estimated model params size (MB)\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "5de1eb41278b42a6a9118e96c198afab",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Sanity Checking: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "c94aaf2fa0d14733b91f74b257edd937",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Training: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "b27e85fcb645490ba448118900ea9fb9",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Metric train_loss improved. New best score: 0.676\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "96ba2f587e45407dbc20184f2b76203c",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "130e5cee4dab470dac266c0eb4c41d54",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "4bb720631b2c4c3998d206e8552988da",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "35a889e035264d8d83888b996562a61b",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "babc66a7a39446f692388c42bed63348",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "6b35fcc6fad04c90a5b6af19a2e3b48e",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "642cd2f6b427475a97f5993eec4bfd2e",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "88830f42809742e9a4a8410a0d0db20f",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "0c98728e2a42495fab3259e4c31051dc",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "20b45ac3e75c4f159b1d316418d354cc",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Monitored metric train_loss did not improve in the last 10 records. Best score: 0.676. Signaling Trainer to stop.\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "9ec6b599ec4c49aab8246eeaa7a0680e",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃ Runningstage.validating ┃ ┃\n",
"┃ metric ┃ DataLoader 0 ┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│ val_acc │ 0.8165137767791748 │\n",
"│ val_f1_score │ 0.6969696879386902 │\n",
"│ val_hp_metric │ 0.8165137767791748 │\n",
"│ val_loss │ 0.5011997222900391 │\n",
"│ val_opt_score │ 1.5134835243225098 │\n",
"│ val_precision │ 0.6388888955116272 │\n",
"│ val_recall │ 0.7666666507720947 │\n",
"│ val_roc_auc │ 0.8742616176605225 │\n",
"└───────────────────────────┴───────────────────────────┘\n",
"\n"
],
"text/plain": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃\u001b[1m \u001b[0m\u001b[1m Runningstage.validating \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m┃\n",
"┃\u001b[1m \u001b[0m\u001b[1m metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│\u001b[36m \u001b[0m\u001b[36m val_acc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8165137767791748 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_f1_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.6969696879386902 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_hp_metric \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8165137767791748 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.5011997222900391 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_opt_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 1.5134835243225098 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_precision \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.6388888955116272 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_recall \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.7666666507720947 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_roc_auc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8742616176605225 \u001b[0m\u001b[35m \u001b[0m│\n",
"└───────────────────────────┴───────────────────────────┘\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"GPU available: True (cuda), used: True\n",
"TPU available: False, using: 0 TPU cores\n",
"IPU available: False, using: 0 IPUs\n",
"HPU available: False, using: 0 HPUs\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n",
"\n",
" | Name | Type | Params\n",
"------------------------------------------\n",
"0 | poi_emb | Linear | 787 K \n",
"1 | e3_emb | Linear | 787 K \n",
"2 | cell_emb | Linear | 590 K \n",
"3 | smiles_emb | Linear | 787 K \n",
"4 | fc1 | Linear | 2.4 M \n",
"5 | fc2 | Linear | 590 K \n",
"6 | fc3 | Linear | 769 \n",
"7 | dropout | Dropout | 0 \n",
"8 | metrics | ModuleDict | 0 \n",
"------------------------------------------\n",
"5.9 M Trainable params\n",
"0 Non-trainable params\n",
"5.9 M Total params\n",
"23.614 Total estimated model params size (MB)\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "6b66c9888c7c432ba437b770b9880269",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Sanity Checking: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "7e150636c9f34a119c42252d7d0be036",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Training: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "4d4b4d556adb44349013e9eec5270942",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Metric train_loss improved. New best score: 0.679\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "c3812ba392834b1aa498b96d88a9146d",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "746c720b0bfa4f518d203660616b403e",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "3f573dfa97934c46be4fee5793acae7c",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "b703b204cb6d44c2accf3d13cfff1e2f",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "dbedf9893f304161969b2b849fa267ad",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "4f6a4f1ec73145ecb99c1e2c113c51f6",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "33db1aaf25a449dbba205039fb9efba9",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "8e98390f334c4fc58f7400fb28a0ea06",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "c9fd3333ba8643b4a2ae086072f3487d",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "d78bc24fd3a545249f8111337f2767a2",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Monitored metric train_loss did not improve in the last 10 records. Best score: 0.679. Signaling Trainer to stop.\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "d2b152723ff842059542ac61f5fa208c",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃ Runningstage.validating ┃ ┃\n",
"┃ metric ┃ DataLoader 0 ┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│ val_acc │ 0.8148148059844971 │\n",
"│ val_f1_score │ 0.6296296119689941 │\n",
"│ val_hp_metric │ 0.8148148059844971 │\n",
"│ val_loss │ 0.4128589630126953 │\n",
"│ val_opt_score │ 1.4444444179534912 │\n",
"│ val_precision │ 0.48571428656578064 │\n",
"│ val_recall │ 0.8947368264198303 │\n",
"│ val_roc_auc │ 0.932584285736084 │\n",
"└───────────────────────────┴───────────────────────────┘\n",
"\n"
],
"text/plain": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃\u001b[1m \u001b[0m\u001b[1m Runningstage.validating \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m┃\n",
"┃\u001b[1m \u001b[0m\u001b[1m metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│\u001b[36m \u001b[0m\u001b[36m val_acc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8148148059844971 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_f1_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.6296296119689941 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_hp_metric \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8148148059844971 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.4128589630126953 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_opt_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 1.4444444179534912 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_precision \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.48571428656578064 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_recall \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8947368264198303 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_roc_auc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.932584285736084 \u001b[0m\u001b[35m \u001b[0m│\n",
"└───────────────────────────┴───────────────────────────┘\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"GPU available: True (cuda), used: True\n",
"TPU available: False, using: 0 TPU cores\n",
"IPU available: False, using: 0 IPUs\n",
"HPU available: False, using: 0 HPUs\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n",
"\n",
" | Name | Type | Params\n",
"------------------------------------------\n",
"0 | poi_emb | Linear | 787 K \n",
"1 | e3_emb | Linear | 787 K \n",
"2 | cell_emb | Linear | 590 K \n",
"3 | smiles_emb | Linear | 787 K \n",
"4 | fc1 | Linear | 2.4 M \n",
"5 | fc2 | Linear | 590 K \n",
"6 | fc3 | Linear | 769 \n",
"7 | dropout | Dropout | 0 \n",
"8 | metrics | ModuleDict | 0 \n",
"------------------------------------------\n",
"5.9 M Trainable params\n",
"0 Non-trainable params\n",
"5.9 M Total params\n",
"23.614 Total estimated model params size (MB)\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "8ebac9304e7046dcad2e58ffc71fa88e",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Sanity Checking: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "73c0044211a7435b9e0174af4d259b4e",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Training: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "fa53da9ef81d4a85b7c63f778f50656f",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Metric train_loss improved. New best score: 0.678\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "938ff958ed174cb090ccdf6359092057",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "80884a1d5479431aa2acab43072d2df7",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "429fac92c0af43b68031e1ea3252b39e",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "2bbcc4eafe8b4d8ab5eb9ed5f22a4c5a",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "986aa77e2a6e42c18d6ecd93e1eefcf9",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "3c48726787834adbbc9530381d61eaac",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "f3042e08db6c438ea66b16f1b1a0d579",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "c16a681c6aa24ef2a5e608d5bd874766",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "9f66b22ea70e401f960b730a932fda71",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "c8103c35b4b44ab0b8f3f1dac0de93c4",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Monitored metric train_loss did not improve in the last 10 records. Best score: 0.678. Signaling Trainer to stop.\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "c760034a955649088011bb70e5d0ebad",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃ Runningstage.validating ┃ ┃\n",
"┃ metric ┃ DataLoader 0 ┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│ val_acc │ 0.8703703880310059 │\n",
"│ val_f1_score │ 0.7941176295280457 │\n",
"│ val_hp_metric │ 0.8703703880310059 │\n",
"│ val_loss │ 0.3084067702293396 │\n",
"│ val_opt_score │ 1.6644880771636963 │\n",
"│ val_precision │ 0.692307710647583 │\n",
"│ val_recall │ 0.931034505367279 │\n",
"│ val_roc_auc │ 0.9456568956375122 │\n",
"└───────────────────────────┴───────────────────────────┘\n",
"\n"
],
"text/plain": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃\u001b[1m \u001b[0m\u001b[1m Runningstage.validating \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m┃\n",
"┃\u001b[1m \u001b[0m\u001b[1m metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│\u001b[36m \u001b[0m\u001b[36m val_acc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8703703880310059 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_f1_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.7941176295280457 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_hp_metric \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8703703880310059 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.3084067702293396 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_opt_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 1.6644880771636963 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_precision \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.692307710647583 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_recall \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.931034505367279 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_roc_auc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.9456568956375122 \u001b[0m\u001b[35m \u001b[0m│\n",
"└───────────────────────────┴───────────────────────────┘\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"GPU available: True (cuda), used: True\n",
"TPU available: False, using: 0 TPU cores\n",
"IPU available: False, using: 0 IPUs\n",
"HPU available: False, using: 0 HPUs\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n",
"\n",
" | Name | Type | Params\n",
"------------------------------------------\n",
"0 | poi_emb | Linear | 787 K \n",
"1 | e3_emb | Linear | 787 K \n",
"2 | cell_emb | Linear | 590 K \n",
"3 | smiles_emb | Linear | 787 K \n",
"4 | fc1 | Linear | 2.4 M \n",
"5 | fc2 | Linear | 590 K \n",
"6 | fc3 | Linear | 769 \n",
"7 | dropout | Dropout | 0 \n",
"8 | metrics | ModuleDict | 0 \n",
"------------------------------------------\n",
"5.9 M Trainable params\n",
"0 Non-trainable params\n",
"5.9 M Total params\n",
"23.614 Total estimated model params size (MB)\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "bef2da9ca4bc425cb89aaa5a0a5a7d4a",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Sanity Checking: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "a2d6d9ddf1c248f4bb9590fef4726bcc",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Training: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "e4e5db39c3514142b773133df7e081bd",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Metric train_loss improved. New best score: 0.676\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "a183ec05417444f1888b88e9f9f2e521",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "528376f5cc5847deaf32c7df090737a2",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "9c5602cc72ec477e8f7d3c77460af50e",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "08e2769e80824629bac808c4050ac22a",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "0f0ae8922ef64eac9ac192f3b9d4cb37",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "8e4bfe586a474f68ad58e7d98a3cb27b",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "cfc728dad560420685887369ec6c23e5",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "9719c095361c4b74955707fd870b9493",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "28a80e6a7cf44ef5a43bc80a23959477",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "0006c9b7d49f4ce0a9385f112bad560d",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Monitored metric train_loss did not improve in the last 10 records. Best score: 0.676. Signaling Trainer to stop.\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "5606d80cfeba4289a3c096afe26cbb22",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃ Runningstage.validating ┃ ┃\n",
"┃ metric ┃ DataLoader 0 ┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│ val_acc │ 0.7962962985038757 │\n",
"│ val_f1_score │ 0.6764705777168274 │\n",
"│ val_hp_metric │ 0.7962962985038757 │\n",
"│ val_loss │ 0.45846864581108093 │\n",
"│ val_opt_score │ 1.4727668762207031 │\n",
"│ val_precision │ 0.6216216087341309 │\n",
"│ val_recall │ 0.7419354915618896 │\n",
"│ val_roc_auc │ 0.8856305480003357 │\n",
"└───────────────────────────┴───────────────────────────┘\n",
"\n"
],
"text/plain": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃\u001b[1m \u001b[0m\u001b[1m Runningstage.validating \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m┃\n",
"┃\u001b[1m \u001b[0m\u001b[1m metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│\u001b[36m \u001b[0m\u001b[36m val_acc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.7962962985038757 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_f1_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.6764705777168274 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_hp_metric \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.7962962985038757 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.45846864581108093 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_opt_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 1.4727668762207031 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_precision \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.6216216087341309 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_recall \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.7419354915618896 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_roc_auc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8856305480003357 \u001b[0m\u001b[35m \u001b[0m│\n",
"└───────────────────────────┴───────────────────────────┘\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" val_loss | \n",
" val_acc | \n",
" val_f1_score | \n",
" val_hp_metric | \n",
" val_opt_score | \n",
" val_precision | \n",
" val_recall | \n",
" val_roc_auc | \n",
" fold | \n",
"
\n",
" \n",
" \n",
" \n",
" | mean | \n",
" 0.417120 | \n",
" 0.835304 | \n",
" 0.711396 | \n",
" 0.835304 | \n",
" 1.5467 | \n",
" 0.653788 | \n",
" 0.801927 | \n",
" 0.910050 | \n",
" 4.50000 | \n",
"
\n",
" \n",
" | std | \n",
" 0.062928 | \n",
" 0.024158 | \n",
" 0.044705 | \n",
" 0.024158 | \n",
" 0.0662 | \n",
" 0.089258 | \n",
" 0.099826 | \n",
" 0.023552 | \n",
" 3.02765 | \n",
"
\n",
" \n",
"
\n",
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# Plot aggregated results\n",
"import seaborn as sns\n",
"import matplotlib.pyplot as plt\n",
"\n",
"for n_splits in [5, 10]:\n",
" val_metrics = pd.read_csv(\n",
" f'../data/results/val_metrics_{n_splits}-splits_ablation.csv'\n",
" )\n",
"\n",
" tmp = val_metrics.copy()[['fold', 'val_acc', 'val_roc_auc',\n",
" 'val_precision', 'val_recall', 'val_f1_score']]\n",
" # Rename columns\n",
" tmp.columns = ['fold', 'Accuracy', 'ROC AUC',\n",
" 'Precision', 'Recall', 'F1 score']\n",
" # Plot aggregated mean value of all metrics in one bar plot\n",
" tmp = tmp.melt(id_vars='fold', var_name='Metric', value_name='Score')\n",
"\n",
" sns.barplot(data=tmp, x='Metric', y='Score')\n",
"\n",
" num_data = len(protac_df[protac_df['Active'].notna()])\n",
" plt.title(\n",
" f'Validation metrics ({n_splits}-fold cross-validation on {num_data} samples)')\n",
" plt.grid(axis='y', alpha=0.5)\n",
" # Set y-axis to end at 1.0\n",
" plt.ylim(0, 1.0)\n",
" # Make the y-axis as percentage\n",
" plt.gca().yaxis.set_major_formatter(\n",
" plt.matplotlib.ticker.PercentFormatter(1, decimals=0))\n",
" plt.show()"
]
},
{
"cell_type": "code",
"execution_count": 126,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# Plot aggregated results\n",
"import seaborn as sns\n",
"import matplotlib.pyplot as plt\n",
"\n",
"n_splits = 5\n",
"val_metrics = pd.read_csv(\n",
" f'../data/results/val_metrics_{n_splits}-splits_ablation.csv'\n",
")\n",
"\n",
"tmp = val_metrics.copy()[['fold', 'val_acc', 'val_roc_auc',\n",
" 'val_precision', 'val_recall', 'val_f1_score']]\n",
"# Rename columns\n",
"tmp.columns = ['fold', 'Accuracy', 'ROC AUC',\n",
" 'Precision', 'Recall', 'F1 score']\n",
"# Plot aggregated mean value of all metrics in one bar plot\n",
"tmp = tmp.melt(id_vars='fold', var_name='Metric', value_name='Score')\n",
"\n",
"sns.barplot(data=tmp, x='Metric', y='Score')\n",
"\n",
"\n",
"num_data = len(protac_df[protac_df['Active'].notna()])\n",
"plt.title(\n",
" f'Validation metrics ({n_splits}-fold cross-validation on {num_data} samples)')\n",
"plt.grid(axis='y', alpha=0.5)\n",
"# Set y-axis to end at 1.0\n",
"plt.ylim(0, 1.0)\n",
"# Make the y-axis as percentage\n",
"plt.gca().yaxis.set_major_formatter(\n",
" plt.matplotlib.ticker.PercentFormatter(1, decimals=0))\n",
"plt.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Ablation Studies"
]
},
{
"cell_type": "code",
"execution_count": 29,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"c:\\Users\\ste\\Anaconda2\\envs\\env-thesis\\Lib\\site-packages\\imblearn\\over_sampling\\_smote\\base.py:345: FutureWarning: The parameter `n_jobs` has been deprecated in 0.10 and will be removed in 0.12. You can pass an nearest neighbors estimator where `n_jobs` is already set instead.\n",
" warnings.warn(\n",
"GPU available: True (cuda), used: True\n",
"TPU available: False, using: 0 TPU cores\n",
"IPU available: False, using: 0 IPUs\n",
"HPU available: False, using: 0 HPUs\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n",
"\n",
" | Name | Type | Params\n",
"------------------------------------------\n",
"0 | smiles_emb | Linear | 787 K \n",
"1 | fc1 | Linear | 590 K \n",
"2 | fc2 | Linear | 590 K \n",
"3 | fc3 | Linear | 769 \n",
"4 | dropout | Dropout | 0 \n",
"5 | metrics | ModuleDict | 0 \n",
"------------------------------------------\n",
"2.0 M Trainable params\n",
"0 Non-trainable params\n",
"2.0 M Total params\n",
"7.877 Total estimated model params size (MB)\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "50dbc6fade944436be04b6bcd2a79e91",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Sanity Checking: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"c:\\Users\\ste\\Anaconda2\\envs\\env-thesis\\Lib\\site-packages\\pytorch_lightning\\trainer\\connectors\\data_connector.py:432: PossibleUserWarning: The dataloader, val_dataloader, does not have many workers which may be a bottleneck. Consider increasing the value of the `num_workers` argument` (try 8 which is the number of cpus on this machine) in the `DataLoader` init to improve performance.\n",
" rank_zero_warn(\n",
"c:\\Users\\ste\\Anaconda2\\envs\\env-thesis\\Lib\\site-packages\\pytorch_lightning\\trainer\\connectors\\data_connector.py:432: PossibleUserWarning: The dataloader, train_dataloader, does not have many workers which may be a bottleneck. Consider increasing the value of the `num_workers` argument` (try 8 which is the number of cpus on this machine) in the `DataLoader` init to improve performance.\n",
" rank_zero_warn(\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "c515836678af4c2786cc0857ac2a89d9",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Training: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "f5a8e06b6f3e488bbb95135d5869b821",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Metric train_loss improved. New best score: 0.679\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "cffd92ecff6b4e26ab07eff9bd96f520",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "b4206cc9119848d7bc2fc72206f67fc7",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "3911809e2a5d4562aa378cf7e19ce420",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "382ca9b39e6149c88019e09c0d2eaa0e",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "83804e7a6df449e286d1719a8fea697e",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "a05b21e45d00417e80b3d0502828b07c",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "1c450f427d444b1c88351773896ddc78",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "bcff7aadc1d44a6fa4267b4e4d8c6859",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "b9796bc1e5344695adf3103d12ad3fd7",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "720f5c893557434aa0cd1360cce94c11",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Monitored metric train_loss did not improve in the last 10 records. Best score: 0.679. Signaling Trainer to stop.\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "f2ad945fe02b472787fc02cb2608fc51",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃ Runningstage.validating ┃ ┃\n",
"┃ metric ┃ DataLoader 0 ┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│ val_acc │ 0.8894008994102478 │\n",
"│ val_f1_score │ 0.8285714387893677 │\n",
"│ val_hp_metric │ 0.8894008994102478 │\n",
"│ val_loss │ 0.22978010773658752 │\n",
"│ val_opt_score │ 1.7179722785949707 │\n",
"│ val_precision │ 0.7250000238418579 │\n",
"│ val_recall │ 0.9666666388511658 │\n",
"│ val_roc_auc │ 0.9700636863708496 │\n",
"└───────────────────────────┴───────────────────────────┘\n",
"\n"
],
"text/plain": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃\u001b[1m \u001b[0m\u001b[1m Runningstage.validating \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m┃\n",
"┃\u001b[1m \u001b[0m\u001b[1m metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│\u001b[36m \u001b[0m\u001b[36m val_acc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8894008994102478 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_f1_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8285714387893677 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_hp_metric \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8894008994102478 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.22978010773658752 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_opt_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 1.7179722785949707 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_precision \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.7250000238418579 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_recall \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.9666666388511658 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_roc_auc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.9700636863708496 \u001b[0m\u001b[35m \u001b[0m│\n",
"└───────────────────────────┴───────────────────────────┘\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/plain": [
"{'val_loss': 0.22978010773658752,\n",
" 'val_acc': 0.8894008994102478,\n",
" 'val_f1_score': 0.8285714387893677,\n",
" 'val_hp_metric': 0.8894008994102478,\n",
" 'val_opt_score': 1.7179722785949707,\n",
" 'val_precision': 0.7250000238418579,\n",
" 'val_recall': 0.9666666388511658,\n",
" 'val_roc_auc': 0.9700636863708496}"
]
},
"execution_count": 29,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Random split the data into train and test\n",
"from sklearn.model_selection import train_test_split\n",
"\n",
"train_df, test_df = train_test_split(\n",
" protac_df[protac_df['Active'].notna()], test_size=0.2, random_state=42,\n",
")\n",
"model, trainer, val_metric = train_model(\n",
" train_df, val_df, disabled_embeddings=['poi', 'e3', 'cell'],\n",
")\n",
"val_metric"
]
},
{
"cell_type": "code",
"execution_count": 30,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Fold 0, disabled embeddings: ['poi', 'e3', 'cell']\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"c:\\Users\\ste\\Anaconda2\\envs\\env-thesis\\Lib\\site-packages\\imblearn\\over_sampling\\_smote\\base.py:345: FutureWarning: The parameter `n_jobs` has been deprecated in 0.10 and will be removed in 0.12. You can pass an nearest neighbors estimator where `n_jobs` is already set instead.\n",
" warnings.warn(\n",
"GPU available: True (cuda), used: True\n",
"TPU available: False, using: 0 TPU cores\n",
"IPU available: False, using: 0 IPUs\n",
"HPU available: False, using: 0 HPUs\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n",
"\n",
" | Name | Type | Params\n",
"------------------------------------------\n",
"0 | smiles_emb | Linear | 787 K \n",
"1 | fc1 | Linear | 590 K \n",
"2 | fc2 | Linear | 590 K \n",
"3 | fc3 | Linear | 769 \n",
"4 | dropout | Dropout | 0 \n",
"5 | metrics | ModuleDict | 0 \n",
"------------------------------------------\n",
"2.0 M Trainable params\n",
"0 Non-trainable params\n",
"2.0 M Total params\n",
"7.877 Total estimated model params size (MB)\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "6d2003ac853140d4be07094f870d01e7",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Sanity Checking: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"c:\\Users\\ste\\Anaconda2\\envs\\env-thesis\\Lib\\site-packages\\pytorch_lightning\\trainer\\connectors\\data_connector.py:432: PossibleUserWarning: The dataloader, val_dataloader, does not have many workers which may be a bottleneck. Consider increasing the value of the `num_workers` argument` (try 8 which is the number of cpus on this machine) in the `DataLoader` init to improve performance.\n",
" rank_zero_warn(\n",
"c:\\Users\\ste\\Anaconda2\\envs\\env-thesis\\Lib\\site-packages\\pytorch_lightning\\trainer\\connectors\\data_connector.py:432: PossibleUserWarning: The dataloader, train_dataloader, does not have many workers which may be a bottleneck. Consider increasing the value of the `num_workers` argument` (try 8 which is the number of cpus on this machine) in the `DataLoader` init to improve performance.\n",
" rank_zero_warn(\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "c5dc5bc109154647a91eba389ddfd0d5",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Training: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "717bab699d8e469ebfab4c4d165dbea2",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Metric train_loss improved. New best score: 0.681\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "a05d22e6bf7048d2be4fc36bb0da2564",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "7b747e15ea954c39b327caffc09ff866",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "68cd758182654515ba18d86dbee435c8",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "773de468100e493fa4190a263cf2d407",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "6f83c8642a7e4f598f7ee3d47a6ce363",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "b11a9156d5dc4883978bf3680c4dd17b",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "1cf15595d32d43b5a34887cbdbc1ff60",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "5eea4f1426664a7794cb229e4fc2b697",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "854eca19c354476c8134eb6758b9fe3e",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "56219f060834490f90e0597caa0ef4df",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Monitored metric train_loss did not improve in the last 10 records. Best score: 0.681. Signaling Trainer to stop.\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "5c1cc664f09f488f92b1f98cf61fdee2",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃ Runningstage.validating ┃ ┃\n",
"┃ metric ┃ DataLoader 0 ┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│ val_acc │ 0.8256880640983582 │\n",
"│ val_f1_score │ 0.6984127163887024 │\n",
"│ val_hp_metric │ 0.8256880640983582 │\n",
"│ val_loss │ 0.43253543972969055 │\n",
"│ val_opt_score │ 1.5241007804870605 │\n",
"│ val_precision │ 0.6666666865348816 │\n",
"│ val_recall │ 0.7333333492279053 │\n",
"│ val_roc_auc │ 0.8918248414993286 │\n",
"└───────────────────────────┴───────────────────────────┘\n",
"\n"
],
"text/plain": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃\u001b[1m \u001b[0m\u001b[1m Runningstage.validating \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m┃\n",
"┃\u001b[1m \u001b[0m\u001b[1m metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│\u001b[36m \u001b[0m\u001b[36m val_acc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8256880640983582 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_f1_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.6984127163887024 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_hp_metric \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8256880640983582 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.43253543972969055 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_opt_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 1.5241007804870605 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_precision \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.6666666865348816 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_recall \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.7333333492279053 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_roc_auc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8918248414993286 \u001b[0m\u001b[35m \u001b[0m│\n",
"└───────────────────────────┴───────────────────────────┘\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Fold 1, disabled embeddings: ['poi', 'e3', 'cell']\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"c:\\Users\\ste\\Anaconda2\\envs\\env-thesis\\Lib\\site-packages\\imblearn\\over_sampling\\_smote\\base.py:345: FutureWarning: The parameter `n_jobs` has been deprecated in 0.10 and will be removed in 0.12. You can pass an nearest neighbors estimator where `n_jobs` is already set instead.\n",
" warnings.warn(\n",
"GPU available: True (cuda), used: True\n",
"TPU available: False, using: 0 TPU cores\n",
"IPU available: False, using: 0 IPUs\n",
"HPU available: False, using: 0 HPUs\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n",
"\n",
" | Name | Type | Params\n",
"------------------------------------------\n",
"0 | smiles_emb | Linear | 787 K \n",
"1 | fc1 | Linear | 590 K \n",
"2 | fc2 | Linear | 590 K \n",
"3 | fc3 | Linear | 769 \n",
"4 | dropout | Dropout | 0 \n",
"5 | metrics | ModuleDict | 0 \n",
"------------------------------------------\n",
"2.0 M Trainable params\n",
"0 Non-trainable params\n",
"2.0 M Total params\n",
"7.877 Total estimated model params size (MB)\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "d19d15ddd9b941129b8fbcf5269061de",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Sanity Checking: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"c:\\Users\\ste\\Anaconda2\\envs\\env-thesis\\Lib\\site-packages\\pytorch_lightning\\trainer\\connectors\\data_connector.py:432: PossibleUserWarning: The dataloader, val_dataloader, does not have many workers which may be a bottleneck. Consider increasing the value of the `num_workers` argument` (try 8 which is the number of cpus on this machine) in the `DataLoader` init to improve performance.\n",
" rank_zero_warn(\n",
"c:\\Users\\ste\\Anaconda2\\envs\\env-thesis\\Lib\\site-packages\\pytorch_lightning\\trainer\\connectors\\data_connector.py:432: PossibleUserWarning: The dataloader, train_dataloader, does not have many workers which may be a bottleneck. Consider increasing the value of the `num_workers` argument` (try 8 which is the number of cpus on this machine) in the `DataLoader` init to improve performance.\n",
" rank_zero_warn(\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "9a8bbd54d38a4abba80e21b1a1a05a48",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Training: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "bd431510337a400a991c561a505e09fa",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Metric train_loss improved. New best score: 0.681\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "2d819e2db2c5490f84d950ffa65c3c21",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "88a9d28bdbda4cccbc172e0ec5d72e5b",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "6147bbb29d544d51a657aeb8ce5216a8",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "62c442218b7242cc9e13d4392935d8ab",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "a9380f8439ff4f2f969a9a9020d9d454",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "628ec57e41404628be9cffb1f11a18bf",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "4e26d4f2cf4e474790d45d1011a7fcc5",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "459b040053fc4454b1d7fc60b1d0a243",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "c443ace858774c38bc911277817931cb",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "6d46e4b8f5514cc5ab33c02fd32555b2",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Monitored metric train_loss did not improve in the last 10 records. Best score: 0.681. Signaling Trainer to stop.\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "d98851e6b9a04e95bacb89e54fda1e6c",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃ Runningstage.validating ┃ ┃\n",
"┃ metric ┃ DataLoader 0 ┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│ val_acc │ 0.8302752375602722 │\n",
"│ val_f1_score │ 0.678260862827301 │\n",
"│ val_hp_metric │ 0.8302752375602722 │\n",
"│ val_loss │ 0.42626988887786865 │\n",
"│ val_opt_score │ 1.5085361003875732 │\n",
"│ val_precision │ 0.6393442749977112 │\n",
"│ val_recall │ 0.7222222089767456 │\n",
"│ val_roc_auc │ 0.878669798374176 │\n",
"└───────────────────────────┴───────────────────────────┘\n",
"\n"
],
"text/plain": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃\u001b[1m \u001b[0m\u001b[1m Runningstage.validating \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m┃\n",
"┃\u001b[1m \u001b[0m\u001b[1m metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│\u001b[36m \u001b[0m\u001b[36m val_acc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8302752375602722 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_f1_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.678260862827301 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_hp_metric \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8302752375602722 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.42626988887786865 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_opt_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 1.5085361003875732 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_precision \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.6393442749977112 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_recall \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.7222222089767456 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_roc_auc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.878669798374176 \u001b[0m\u001b[35m \u001b[0m│\n",
"└───────────────────────────┴───────────────────────────┘\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Fold 2, disabled embeddings: ['poi', 'e3', 'cell']\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"c:\\Users\\ste\\Anaconda2\\envs\\env-thesis\\Lib\\site-packages\\imblearn\\over_sampling\\_smote\\base.py:345: FutureWarning: The parameter `n_jobs` has been deprecated in 0.10 and will be removed in 0.12. You can pass an nearest neighbors estimator where `n_jobs` is already set instead.\n",
" warnings.warn(\n",
"GPU available: True (cuda), used: True\n",
"TPU available: False, using: 0 TPU cores\n",
"IPU available: False, using: 0 IPUs\n",
"HPU available: False, using: 0 HPUs\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n",
"\n",
" | Name | Type | Params\n",
"------------------------------------------\n",
"0 | smiles_emb | Linear | 787 K \n",
"1 | fc1 | Linear | 590 K \n",
"2 | fc2 | Linear | 590 K \n",
"3 | fc3 | Linear | 769 \n",
"4 | dropout | Dropout | 0 \n",
"5 | metrics | ModuleDict | 0 \n",
"------------------------------------------\n",
"2.0 M Trainable params\n",
"0 Non-trainable params\n",
"2.0 M Total params\n",
"7.877 Total estimated model params size (MB)\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "b34dbf29083d470faaa27389eec33307",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Sanity Checking: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"c:\\Users\\ste\\Anaconda2\\envs\\env-thesis\\Lib\\site-packages\\pytorch_lightning\\trainer\\connectors\\data_connector.py:432: PossibleUserWarning: The dataloader, val_dataloader, does not have many workers which may be a bottleneck. Consider increasing the value of the `num_workers` argument` (try 8 which is the number of cpus on this machine) in the `DataLoader` init to improve performance.\n",
" rank_zero_warn(\n",
"c:\\Users\\ste\\Anaconda2\\envs\\env-thesis\\Lib\\site-packages\\pytorch_lightning\\trainer\\connectors\\data_connector.py:432: PossibleUserWarning: The dataloader, train_dataloader, does not have many workers which may be a bottleneck. Consider increasing the value of the `num_workers` argument` (try 8 which is the number of cpus on this machine) in the `DataLoader` init to improve performance.\n",
" rank_zero_warn(\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "c9f0b66f5f904854b19f39f6577f78ff",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Training: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "1242edb426bd4fe7a586d40b5a03bc78",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Metric train_loss improved. New best score: 0.681\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "8ac5b3a3c37c47a6989698e3381f7f64",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "3891c114903a4a98a84ea80337f11862",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "7eb7a538328a4322bfeead1e4870af47",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "471b041184324116bae3f3dab334ab11",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "055eced85ba14e1689e93d92f8632d83",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "b17ec92dc032401db16a412cf20f1446",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "d7751a12e9484eea8314c9077178e41c",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "d1750a5a165d459e8dbd7c4b55cc9107",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "df6a7fab18224e8ead414f5f9c21116a",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "b4d0e7ee279e4a1b9a256335c0402a57",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Monitored metric train_loss did not improve in the last 10 records. Best score: 0.681. Signaling Trainer to stop.\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "38dfcd5784fe42309b022c4b3fb3643c",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃ Runningstage.validating ┃ ┃\n",
"┃ metric ┃ DataLoader 0 ┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│ val_acc │ 0.8018433451652527 │\n",
"│ val_f1_score │ 0.6814814805984497 │\n",
"│ val_hp_metric │ 0.8018433451652527 │\n",
"│ val_loss │ 0.44118818640708923 │\n",
"│ val_opt_score │ 1.4833247661590576 │\n",
"│ val_precision │ 0.5822784900665283 │\n",
"│ val_recall │ 0.8214285969734192 │\n",
"│ val_roc_auc │ 0.8986246585845947 │\n",
"└───────────────────────────┴───────────────────────────┘\n",
"\n"
],
"text/plain": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃\u001b[1m \u001b[0m\u001b[1m Runningstage.validating \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m┃\n",
"┃\u001b[1m \u001b[0m\u001b[1m metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│\u001b[36m \u001b[0m\u001b[36m val_acc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8018433451652527 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_f1_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.6814814805984497 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_hp_metric \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8018433451652527 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.44118818640708923 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_opt_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 1.4833247661590576 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_precision \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.5822784900665283 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_recall \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8214285969734192 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_roc_auc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8986246585845947 \u001b[0m\u001b[35m \u001b[0m│\n",
"└───────────────────────────┴───────────────────────────┘\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Fold 3, disabled embeddings: ['poi', 'e3', 'cell']\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"c:\\Users\\ste\\Anaconda2\\envs\\env-thesis\\Lib\\site-packages\\imblearn\\over_sampling\\_smote\\base.py:345: FutureWarning: The parameter `n_jobs` has been deprecated in 0.10 and will be removed in 0.12. You can pass an nearest neighbors estimator where `n_jobs` is already set instead.\n",
" warnings.warn(\n",
"GPU available: True (cuda), used: True\n",
"TPU available: False, using: 0 TPU cores\n",
"IPU available: False, using: 0 IPUs\n",
"HPU available: False, using: 0 HPUs\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n",
"\n",
" | Name | Type | Params\n",
"------------------------------------------\n",
"0 | smiles_emb | Linear | 787 K \n",
"1 | fc1 | Linear | 590 K \n",
"2 | fc2 | Linear | 590 K \n",
"3 | fc3 | Linear | 769 \n",
"4 | dropout | Dropout | 0 \n",
"5 | metrics | ModuleDict | 0 \n",
"------------------------------------------\n",
"2.0 M Trainable params\n",
"0 Non-trainable params\n",
"2.0 M Total params\n",
"7.877 Total estimated model params size (MB)\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "2dd06a2b45e84e8c8a993f61660ba6ad",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Sanity Checking: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"c:\\Users\\ste\\Anaconda2\\envs\\env-thesis\\Lib\\site-packages\\pytorch_lightning\\trainer\\connectors\\data_connector.py:432: PossibleUserWarning: The dataloader, val_dataloader, does not have many workers which may be a bottleneck. Consider increasing the value of the `num_workers` argument` (try 8 which is the number of cpus on this machine) in the `DataLoader` init to improve performance.\n",
" rank_zero_warn(\n",
"c:\\Users\\ste\\Anaconda2\\envs\\env-thesis\\Lib\\site-packages\\pytorch_lightning\\trainer\\connectors\\data_connector.py:432: PossibleUserWarning: The dataloader, train_dataloader, does not have many workers which may be a bottleneck. Consider increasing the value of the `num_workers` argument` (try 8 which is the number of cpus on this machine) in the `DataLoader` init to improve performance.\n",
" rank_zero_warn(\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "06c4442f48b74b4d88cf345db745a8ff",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Training: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "944d17ee72ff40a7ab71e7a8f5b233bf",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Metric train_loss improved. New best score: 0.681\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "5c90f09634e2419b90518c10de844d15",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "7439a86f21db4e2b9828e8d31689ee12",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "085cf04c66c14c419dd06b7a10d78921",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "f0760a28b97b42c1b97e9bf42de5904b",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "3e347c35a768467fa371bd4857e4c18b",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "79607567baae44258cad8d96c93185dd",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "71c052c5a3464a01b3382b77eada459b",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "a8a0518f250e44fb9cd39173b1795b9e",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "445027abe12444cc81dfd315e6d0b13f",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "f931a407d4f64fd28f53e4e89408d4fb",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Monitored metric train_loss did not improve in the last 10 records. Best score: 0.681. Signaling Trainer to stop.\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "70ea178af7b04737bb6a92552ff39f94",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃ Runningstage.validating ┃ ┃\n",
"┃ metric ┃ DataLoader 0 ┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│ val_acc │ 0.7972350120544434 │\n",
"│ val_f1_score │ 0.6271186470985413 │\n",
"│ val_hp_metric │ 0.7972350120544434 │\n",
"│ val_loss │ 0.4882689416408539 │\n",
"│ val_opt_score │ 1.4243535995483398 │\n",
"│ val_precision │ 0.5441176295280457 │\n",
"│ val_recall │ 0.7400000095367432 │\n",
"│ val_roc_auc │ 0.8674850463867188 │\n",
"└───────────────────────────┴───────────────────────────┘\n",
"\n"
],
"text/plain": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃\u001b[1m \u001b[0m\u001b[1m Runningstage.validating \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m┃\n",
"┃\u001b[1m \u001b[0m\u001b[1m metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│\u001b[36m \u001b[0m\u001b[36m val_acc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.7972350120544434 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_f1_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.6271186470985413 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_hp_metric \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.7972350120544434 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.4882689416408539 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_opt_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 1.4243535995483398 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_precision \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.5441176295280457 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_recall \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.7400000095367432 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_roc_auc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8674850463867188 \u001b[0m\u001b[35m \u001b[0m│\n",
"└───────────────────────────┴───────────────────────────┘\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Fold 4, disabled embeddings: ['poi', 'e3', 'cell']\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"c:\\Users\\ste\\Anaconda2\\envs\\env-thesis\\Lib\\site-packages\\imblearn\\over_sampling\\_smote\\base.py:345: FutureWarning: The parameter `n_jobs` has been deprecated in 0.10 and will be removed in 0.12. You can pass an nearest neighbors estimator where `n_jobs` is already set instead.\n",
" warnings.warn(\n",
"GPU available: True (cuda), used: True\n",
"TPU available: False, using: 0 TPU cores\n",
"IPU available: False, using: 0 IPUs\n",
"HPU available: False, using: 0 HPUs\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n",
"\n",
" | Name | Type | Params\n",
"------------------------------------------\n",
"0 | smiles_emb | Linear | 787 K \n",
"1 | fc1 | Linear | 590 K \n",
"2 | fc2 | Linear | 590 K \n",
"3 | fc3 | Linear | 769 \n",
"4 | dropout | Dropout | 0 \n",
"5 | metrics | ModuleDict | 0 \n",
"------------------------------------------\n",
"2.0 M Trainable params\n",
"0 Non-trainable params\n",
"2.0 M Total params\n",
"7.877 Total estimated model params size (MB)\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "887bda9f44ba4f43b5cb33fe3e850f0d",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Sanity Checking: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"c:\\Users\\ste\\Anaconda2\\envs\\env-thesis\\Lib\\site-packages\\pytorch_lightning\\trainer\\connectors\\data_connector.py:432: PossibleUserWarning: The dataloader, val_dataloader, does not have many workers which may be a bottleneck. Consider increasing the value of the `num_workers` argument` (try 8 which is the number of cpus on this machine) in the `DataLoader` init to improve performance.\n",
" rank_zero_warn(\n",
"c:\\Users\\ste\\Anaconda2\\envs\\env-thesis\\Lib\\site-packages\\pytorch_lightning\\trainer\\connectors\\data_connector.py:432: PossibleUserWarning: The dataloader, train_dataloader, does not have many workers which may be a bottleneck. Consider increasing the value of the `num_workers` argument` (try 8 which is the number of cpus on this machine) in the `DataLoader` init to improve performance.\n",
" rank_zero_warn(\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "d86c1f414c6d4bf4836c9c9df7befef1",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Training: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "13e4cea9ba2d4e2cad9fb7ffb04e734e",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Metric train_loss improved. New best score: 0.679\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "9024ec21c159467e99cf0e53100bcb08",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "135764bbe2c743868055ed007a2da585",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "c97d0e9b0a514c71b9ecabf2c4742a91",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "eccc6b5f67c447a7b67baa201a7370be",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "893ce33822b54e4e92b0c8e697d71e55",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "32bd15f2da724b45ac7500a88613306d",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "cb3c0d29d38c4197940c727bf47b315a",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "efd8ec6f393d487c96d68d1d69bb6073",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "75d471ac44a743f1bddff068c4b7c3c8",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "4218ae5c129f49c7a168c26bd92df42d",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Monitored metric train_loss did not improve in the last 10 records. Best score: 0.679. Signaling Trainer to stop.\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "a0824ee73ebd4e18bc55bf3923b1ebed",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃ Runningstage.validating ┃ ┃\n",
"┃ metric ┃ DataLoader 0 ┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│ val_acc │ 0.8018433451652527 │\n",
"│ val_f1_score │ 0.6861313581466675 │\n",
"│ val_hp_metric │ 0.8018433451652527 │\n",
"│ val_loss │ 0.46606215834617615 │\n",
"│ val_opt_score │ 1.4879746437072754 │\n",
"│ val_precision │ 0.6103895902633667 │\n",
"│ val_recall │ 0.7833333611488342 │\n",
"│ val_roc_auc │ 0.8800424337387085 │\n",
"└───────────────────────────┴───────────────────────────┘\n",
"\n"
],
"text/plain": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃\u001b[1m \u001b[0m\u001b[1m Runningstage.validating \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m┃\n",
"┃\u001b[1m \u001b[0m\u001b[1m metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│\u001b[36m \u001b[0m\u001b[36m val_acc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8018433451652527 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_f1_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.6861313581466675 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_hp_metric \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8018433451652527 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.46606215834617615 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_opt_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 1.4879746437072754 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_precision \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.6103895902633667 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_recall \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.7833333611488342 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_roc_auc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8800424337387085 \u001b[0m\u001b[35m \u001b[0m│\n",
"└───────────────────────────┴───────────────────────────┘\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"C:\\Users\\ste\\AppData\\Local\\Temp\\ipykernel_9048\\4021510138.py:42: FutureWarning: ['Uniprot', 'Smiles', 'E3 Ligase', 'InChI', 'InChI Key', 'Cell Type', 'Article DOI', 'Comments', 'Database', 'Molecular Formula', 'Target', 'PDB', 'Name', 'Assay (DC50/Dmax)', 'Target (Parsed)', 'POI Sequence', 'E3 Ligase Uniprot', 'E3 Ligase Sequence', 'Cell Line Identifier', 'disabled_embeddings'] did not aggregate successfully. If any error is raised this will raise in a future version of pandas. Drop these columns/ops to avoid this warning.\n",
" val_metrics.aggregate(['mean', 'std'])\n"
]
},
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" Compound ID | \n",
" Molecular Weight | \n",
" Heavy Atom Count | \n",
" Ring Count | \n",
" Rotatable Bond Count | \n",
" Topological Polar Surface Area | \n",
" Hydrogen Bond Acceptor Count | \n",
" Hydrogen Bond Donor Count | \n",
" Treatment Time (h) | \n",
" DC50 (nM) | \n",
" ... | \n",
" XLogP3 | \n",
" val_loss | \n",
" val_acc | \n",
" val_f1_score | \n",
" val_hp_metric | \n",
" val_opt_score | \n",
" val_precision | \n",
" val_recall | \n",
" val_roc_auc | \n",
" fold | \n",
"
\n",
" \n",
" \n",
" \n",
" | mean | \n",
" 1309.798528 | \n",
" 984.393343 | \n",
" 70.029439 | \n",
" 7.385465 | \n",
" 20.139834 | \n",
" 220.732282 | \n",
" 13.651334 | \n",
" 4.467341 | \n",
" 18.154774 | \n",
" 2396.467627 | \n",
" ... | \n",
" 5.260886 | \n",
" 0.450865 | \n",
" 0.811377 | \n",
" 0.674281 | \n",
" 0.811377 | \n",
" 1.485658 | \n",
" 0.608559 | \n",
" 0.760064 | \n",
" 0.883329 | \n",
" 2.000000 | \n",
"
\n",
" \n",
" | std | \n",
" 931.846640 | \n",
" 165.936373 | \n",
" 11.294283 | \n",
" 1.368378 | \n",
" 6.379550 | \n",
" 38.692208 | \n",
" 2.872499 | \n",
" 1.429301 | \n",
" 10.410593 | \n",
" 8579.934090 | \n",
" ... | \n",
" 2.418567 | \n",
" 0.025810 | \n",
" 0.015360 | \n",
" 0.027453 | \n",
" 0.015360 | \n",
" 0.037981 | \n",
" 0.047883 | \n",
" 0.041401 | \n",
" 0.012142 | \n",
" 1.581139 | \n",
"
\n",
" \n",
"
\n",
"
2 rows × 24 columns
\n",
"
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃ Runningstage.validating ┃ ┃\n",
"┃ metric ┃ DataLoader 0 ┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│ val_acc │ 0.8119266033172607 │\n",
"│ val_f1_score │ 0.6611570119857788 │\n",
"│ val_hp_metric │ 0.8119266033172607 │\n",
"│ val_loss │ 0.36858928203582764 │\n",
"│ val_opt_score │ 1.4730836153030396 │\n",
"│ val_precision │ 0.6557376980781555 │\n",
"│ val_recall │ 0.6666666865348816 │\n",
"│ val_roc_auc │ 0.8914557695388794 │\n",
"└───────────────────────────┴───────────────────────────┘\n",
"\n"
],
"text/plain": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃\u001b[1m \u001b[0m\u001b[1m Runningstage.validating \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m┃\n",
"┃\u001b[1m \u001b[0m\u001b[1m metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│\u001b[36m \u001b[0m\u001b[36m val_acc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8119266033172607 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_f1_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.6611570119857788 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_hp_metric \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8119266033172607 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.36858928203582764 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_opt_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 1.4730836153030396 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_precision \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.6557376980781555 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_recall \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.6666666865348816 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_roc_auc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8914557695388794 \u001b[0m\u001b[35m \u001b[0m│\n",
"└───────────────────────────┴───────────────────────────┘\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "b20cd9147ce2493d876d1704fc8d82bf",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Predicting: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/plain": [
"[{'val_loss': 0.36858928203582764,\n",
" 'val_acc': 0.8119266033172607,\n",
" 'val_f1_score': 0.6611570119857788,\n",
" 'val_hp_metric': 0.8119266033172607,\n",
" 'val_opt_score': 1.4730836153030396,\n",
" 'val_precision': 0.6557376980781555,\n",
" 'val_recall': 0.6666666865348816,\n",
" 'val_roc_auc': 0.8914557695388794}]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Oracle iteration 1\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"GPU available: True (cuda), used: True\n",
"TPU available: False, using: 0 TPU cores\n",
"IPU available: False, using: 0 IPUs\n",
"HPU available: False, using: 0 HPUs\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n",
"\n",
" | Name | Type | Params\n",
"------------------------------------------\n",
"0 | poi_emb | Linear | 787 K \n",
"1 | e3_emb | Linear | 787 K \n",
"2 | cell_emb | Linear | 590 K \n",
"3 | smiles_emb | Linear | 787 K \n",
"4 | fc1 | Linear | 2.4 M \n",
"5 | fc2 | Linear | 590 K \n",
"6 | fc3 | Linear | 769 \n",
"7 | dropout | Dropout | 0 \n",
"8 | metrics | ModuleDict | 0 \n",
"------------------------------------------\n",
"5.9 M Trainable params\n",
"0 Non-trainable params\n",
"5.9 M Total params\n",
"23.614 Total estimated model params size (MB)\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "3ed1d6be59d14ff6be46e16b266751cf",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Sanity Checking: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "98660bb9f10f4e4497cf935c2d145869",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Training: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "c8b5685f0f2643d7b0403dd210c80048",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Metric train_loss improved. New best score: 0.680\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "c3f6a0cf84f54e6aa11d85dde291108f",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "ae9f8760aa8e449886cd1658283c8de4",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "49e9339fa3634dd3937a7c00dd79a51b",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "0e36317bc6fb41c4b9d2ef87e913b95f",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"`Trainer.fit` stopped: `max_epochs=5` reached.\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "5bc6623e98a94a61b5fe3be9424f637a",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃ Runningstage.validating ┃ ┃\n",
"┃ metric ┃ DataLoader 0 ┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│ val_acc │ 0.8119266033172607 │\n",
"│ val_f1_score │ 0.7284768223762512 │\n",
"│ val_hp_metric │ 0.8119266033172607 │\n",
"│ val_loss │ 0.4363510012626648 │\n",
"│ val_opt_score │ 1.5404033660888672 │\n",
"│ val_precision │ 0.6043956279754639 │\n",
"│ val_recall │ 0.9166666865348816 │\n",
"│ val_roc_auc │ 0.8967299461364746 │\n",
"└───────────────────────────┴───────────────────────────┘\n",
"\n"
],
"text/plain": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃\u001b[1m \u001b[0m\u001b[1m Runningstage.validating \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m┃\n",
"┃\u001b[1m \u001b[0m\u001b[1m metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│\u001b[36m \u001b[0m\u001b[36m val_acc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8119266033172607 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_f1_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.7284768223762512 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_hp_metric \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8119266033172607 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.4363510012626648 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_opt_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 1.5404033660888672 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_precision \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.6043956279754639 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_recall \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.9166666865348816 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_roc_auc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8967299461364746 \u001b[0m\u001b[35m \u001b[0m│\n",
"└───────────────────────────┴───────────────────────────┘\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "c5ac531089ef4db19f115739821d2f2a",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Predicting: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/plain": [
"[{'val_loss': 0.4363510012626648,\n",
" 'val_acc': 0.8119266033172607,\n",
" 'val_f1_score': 0.7284768223762512,\n",
" 'val_hp_metric': 0.8119266033172607,\n",
" 'val_opt_score': 1.5404033660888672,\n",
" 'val_precision': 0.6043956279754639,\n",
" 'val_recall': 0.9166666865348816,\n",
" 'val_roc_auc': 0.8967299461364746}]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Oracle iteration 2\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"GPU available: True (cuda), used: True\n",
"TPU available: False, using: 0 TPU cores\n",
"IPU available: False, using: 0 IPUs\n",
"HPU available: False, using: 0 HPUs\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n",
"\n",
" | Name | Type | Params\n",
"------------------------------------------\n",
"0 | poi_emb | Linear | 787 K \n",
"1 | e3_emb | Linear | 787 K \n",
"2 | cell_emb | Linear | 590 K \n",
"3 | smiles_emb | Linear | 787 K \n",
"4 | fc1 | Linear | 2.4 M \n",
"5 | fc2 | Linear | 590 K \n",
"6 | fc3 | Linear | 769 \n",
"7 | dropout | Dropout | 0 \n",
"8 | metrics | ModuleDict | 0 \n",
"------------------------------------------\n",
"5.9 M Trainable params\n",
"0 Non-trainable params\n",
"5.9 M Total params\n",
"23.614 Total estimated model params size (MB)\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "52f2dee9fd5145bc8281ec9baf04d759",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Sanity Checking: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "52d8b6bab5c84103b77281a06c552103",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Training: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "5b64908b21c14682aef60fd5348a317b",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Metric train_loss improved. New best score: 0.678\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "a4d55544d779490d830147bd4adb9cc7",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "4f6ece7c4eb049f3910719f758a2e84d",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "b6455efe0cc94ca68cb2e16be011e738",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "ae3c2e527d504dcf94ff1937655863b8",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"`Trainer.fit` stopped: `max_epochs=5` reached.\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "9d73120c6f324920acbd2e092689c270",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃ Runningstage.validating ┃ ┃\n",
"┃ metric ┃ DataLoader 0 ┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│ val_acc │ 0.8486238718032837 │\n",
"│ val_f1_score │ 0.7441860437393188 │\n",
"│ val_hp_metric │ 0.8486238718032837 │\n",
"│ val_loss │ 0.3814685344696045 │\n",
"│ val_opt_score │ 1.5928099155426025 │\n",
"│ val_precision │ 0.695652186870575 │\n",
"│ val_recall │ 0.800000011920929 │\n",
"│ val_roc_auc │ 0.9017932415008545 │\n",
"└───────────────────────────┴───────────────────────────┘\n",
"\n"
],
"text/plain": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃\u001b[1m \u001b[0m\u001b[1m Runningstage.validating \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m┃\n",
"┃\u001b[1m \u001b[0m\u001b[1m metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│\u001b[36m \u001b[0m\u001b[36m val_acc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8486238718032837 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_f1_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.7441860437393188 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_hp_metric \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8486238718032837 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.3814685344696045 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_opt_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 1.5928099155426025 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_precision \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.695652186870575 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_recall \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.800000011920929 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_roc_auc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.9017932415008545 \u001b[0m\u001b[35m \u001b[0m│\n",
"└───────────────────────────┴───────────────────────────┘\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "05e87ec2e35d4b7c990a69c77cfd9ff2",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Predicting: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/plain": [
"[{'val_loss': 0.3814685344696045,\n",
" 'val_acc': 0.8486238718032837,\n",
" 'val_f1_score': 0.7441860437393188,\n",
" 'val_hp_metric': 0.8486238718032837,\n",
" 'val_opt_score': 1.5928099155426025,\n",
" 'val_precision': 0.695652186870575,\n",
" 'val_recall': 0.800000011920929,\n",
" 'val_roc_auc': 0.9017932415008545}]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Oracle iteration 3\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"GPU available: True (cuda), used: True\n",
"TPU available: False, using: 0 TPU cores\n",
"IPU available: False, using: 0 IPUs\n",
"HPU available: False, using: 0 HPUs\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n",
"\n",
" | Name | Type | Params\n",
"------------------------------------------\n",
"0 | poi_emb | Linear | 787 K \n",
"1 | e3_emb | Linear | 787 K \n",
"2 | cell_emb | Linear | 590 K \n",
"3 | smiles_emb | Linear | 787 K \n",
"4 | fc1 | Linear | 2.4 M \n",
"5 | fc2 | Linear | 590 K \n",
"6 | fc3 | Linear | 769 \n",
"7 | dropout | Dropout | 0 \n",
"8 | metrics | ModuleDict | 0 \n",
"------------------------------------------\n",
"5.9 M Trainable params\n",
"0 Non-trainable params\n",
"5.9 M Total params\n",
"23.614 Total estimated model params size (MB)\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "10cee0bb09e54342b5667cc310c3b0db",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Sanity Checking: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "412395eb2826440dbc6ddf6ef9cfaee6",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Training: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "6341b540c19e483a8fb161552062c49e",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Metric train_loss improved. New best score: 0.665\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "372085f35ee94e909997b58bb74cf411",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "7aa5b829a3134f41ae08bd03523dd0bf",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "469991d3dcef48a39d462d92e321f6d4",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "049a245540a041209a17c8cdf9d4c249",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"`Trainer.fit` stopped: `max_epochs=5` reached.\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "cbf64814d8a349d1bdf785e3c2003bda",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃ Runningstage.validating ┃ ┃\n",
"┃ metric ┃ DataLoader 0 ┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│ val_acc │ 0.8486238718032837 │\n",
"│ val_f1_score │ 0.7591241002082825 │\n",
"│ val_hp_metric │ 0.8486238718032837 │\n",
"│ val_loss │ 0.3775690495967865 │\n",
"│ val_opt_score │ 1.607748031616211 │\n",
"│ val_precision │ 0.6753246784210205 │\n",
"│ val_recall │ 0.8666666746139526 │\n",
"│ val_roc_auc │ 0.8989450931549072 │\n",
"└───────────────────────────┴───────────────────────────┘\n",
"\n"
],
"text/plain": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃\u001b[1m \u001b[0m\u001b[1m Runningstage.validating \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m┃\n",
"┃\u001b[1m \u001b[0m\u001b[1m metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│\u001b[36m \u001b[0m\u001b[36m val_acc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8486238718032837 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_f1_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.7591241002082825 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_hp_metric \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8486238718032837 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.3775690495967865 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_opt_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 1.607748031616211 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_precision \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.6753246784210205 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_recall \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8666666746139526 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_roc_auc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8989450931549072 \u001b[0m\u001b[35m \u001b[0m│\n",
"└───────────────────────────┴───────────────────────────┘\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "f142c9af8e7749969f88cf54ab742e86",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Predicting: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/plain": [
"[{'val_loss': 0.3775690495967865,\n",
" 'val_acc': 0.8486238718032837,\n",
" 'val_f1_score': 0.7591241002082825,\n",
" 'val_hp_metric': 0.8486238718032837,\n",
" 'val_opt_score': 1.607748031616211,\n",
" 'val_precision': 0.6753246784210205,\n",
" 'val_recall': 0.8666666746139526,\n",
" 'val_roc_auc': 0.8989450931549072}]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Oracle iteration 4\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"GPU available: True (cuda), used: True\n",
"TPU available: False, using: 0 TPU cores\n",
"IPU available: False, using: 0 IPUs\n",
"HPU available: False, using: 0 HPUs\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n",
"\n",
" | Name | Type | Params\n",
"------------------------------------------\n",
"0 | poi_emb | Linear | 787 K \n",
"1 | e3_emb | Linear | 787 K \n",
"2 | cell_emb | Linear | 590 K \n",
"3 | smiles_emb | Linear | 787 K \n",
"4 | fc1 | Linear | 2.4 M \n",
"5 | fc2 | Linear | 590 K \n",
"6 | fc3 | Linear | 769 \n",
"7 | dropout | Dropout | 0 \n",
"8 | metrics | ModuleDict | 0 \n",
"------------------------------------------\n",
"5.9 M Trainable params\n",
"0 Non-trainable params\n",
"5.9 M Total params\n",
"23.614 Total estimated model params size (MB)\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "b46c159286f94c4292a76e773b10ecbd",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Sanity Checking: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "c3ee612bf9db4755b0c0d562c79d3935",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Training: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "305decdb9ec048b4bb3e2c16bc6304dc",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Metric train_loss improved. New best score: 0.665\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "91b59e48a2d541429d9902caf07f2546",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "e170c32c520a4619a96b9c0f400a2d68",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "db12c97c83dc4a77a904325ab83dec48",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "a11d8f8aa45640aab06ce829dc760e3c",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"`Trainer.fit` stopped: `max_epochs=5` reached.\n",
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "53d56161558244359526ba0ad3a88930",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Validation: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃ Runningstage.validating ┃ ┃\n",
"┃ metric ┃ DataLoader 0 ┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│ val_acc │ 0.8348624110221863 │\n",
"│ val_f1_score │ 0.71875 │\n",
"│ val_hp_metric │ 0.8348624110221863 │\n",
"│ val_loss │ 0.38852933049201965 │\n",
"│ val_opt_score │ 1.553612470626831 │\n",
"│ val_precision │ 0.6764705777168274 │\n",
"│ val_recall │ 0.7666666507720947 │\n",
"│ val_roc_auc │ 0.8900843858718872 │\n",
"└───────────────────────────┴───────────────────────────┘\n",
"\n"
],
"text/plain": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━┓\n",
"┃\u001b[1m \u001b[0m\u001b[1m Runningstage.validating \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m┃\n",
"┃\u001b[1m \u001b[0m\u001b[1m metric \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m DataLoader 0 \u001b[0m\u001b[1m \u001b[0m┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━┩\n",
"│\u001b[36m \u001b[0m\u001b[36m val_acc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8348624110221863 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_f1_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.71875 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_hp_metric \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8348624110221863 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_loss \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.38852933049201965 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_opt_score \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 1.553612470626831 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_precision \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.6764705777168274 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_recall \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.7666666507720947 \u001b[0m\u001b[35m \u001b[0m│\n",
"│\u001b[36m \u001b[0m\u001b[36m val_roc_auc \u001b[0m\u001b[36m \u001b[0m│\u001b[35m \u001b[0m\u001b[35m 0.8900843858718872 \u001b[0m\u001b[35m \u001b[0m│\n",
"└───────────────────────────┴───────────────────────────┘\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "2a1d290c9a5940aea897f8c0412646f5",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Predicting: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/plain": [
"[{'val_loss': 0.38852933049201965,\n",
" 'val_acc': 0.8348624110221863,\n",
" 'val_f1_score': 0.71875,\n",
" 'val_hp_metric': 0.8348624110221863,\n",
" 'val_opt_score': 1.553612470626831,\n",
" 'val_precision': 0.6764705777168274,\n",
" 'val_recall': 0.7666666507720947,\n",
" 'val_roc_auc': 0.8900843858718872}]"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"def get_oracle_predictions(model, trainer, test_df):\n",
" test_ds = PROTAC_Dataset(\n",
" test_df,\n",
" protein_embeddings,\n",
" cell2embedding,\n",
" smiles2fp,\n",
" )\n",
" test_dl = DataLoader(test_ds, batch_size=32, shuffle=False)\n",
" test_preds = trainer.predict(model, test_dl)\n",
" test_preds = torch.vstack(test_preds).numpy().flatten()\n",
" test_df['Active'] = [\n",
" is_active(p, pos_threshold=0.97, neg_threshold=0.001) for p in test_preds]\n",
" return test_df\n",
"\n",
"\n",
"train_df, val_df = train_test_split(\n",
" protac_df[protac_df['Active'].notna()], test_size=0.2, random_state=42)\n",
"test_df = protac_df[protac_df['Active'].isna()]\n",
"\n",
"val_metrics = []\n",
"for i in range(5):\n",
" print(f'Oracle iteration {i}')\n",
" model, trainer, metrics = train_model(train_df, val_df)\n",
" val_metrics.append(metrics)\n",
" test_df = get_oracle_predictions(model, trainer, test_df)\n",
" # Update the training set\n",
" train_df = pd.concat([\n",
" train_df,\n",
" test_df[test_df['Active'].notna()]], axis=0).drop_duplicates()\n",
" display(metrics)"
]
},
{
"cell_type": "code",
"execution_count": 108,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "5aede109a90c428e900696220d318335",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Predicting: 0it [00:00, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"image/png": "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",
"text/plain": [
""
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"val_dl = DataLoader(val_ds, batch_size=32, shuffle=False)\n",
"val_preds = trainer.predict(model, val_dl)\n",
"val_preds = torch.vstack(val_preds).numpy().flatten()\n",
"\n",
"# Plot histogram of predictions\n",
"plt.hist(val_preds, bins=20)\n",
"plt.title('Validation predictions')\n",
"plt.grid(axis='y', alpha=0.5)\n",
"plt.show()"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.11.0"
}
},
"nbformat": 4,
"nbformat_minor": 2
}