{
"cells": [
{
"cell_type": "markdown",
"id": "cbc8016f",
"metadata": {},
"source": [
"# GenerateModel\n",
"This tutorial shows how to define and generate a keras model for use with\n",
"TMVA.\n",
"\n",
"\n",
"\n",
"\n",
"**Author:** TMVA Team \n",
"This notebook tutorial was automatically generated with ROOTBOOK-izer from the macro found in the ROOT repository on Tuesday, May 19, 2026 at 08:21 PM."
]
},
{
"cell_type": "code",
"execution_count": 1,
"id": "cd05be6c",
"metadata": {
"collapsed": false,
"execution": {
"iopub.execute_input": "2026-05-19T20:21:09.821895Z",
"iopub.status.busy": "2026-05-19T20:21:09.821772Z",
"iopub.status.idle": "2026-05-19T20:21:12.101831Z",
"shell.execute_reply": "2026-05-19T20:21:12.101258Z"
}
},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"I0000 00:00:1779222070.011287 568646 port.cc:153] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.\n",
"I0000 00:00:1779222070.012539 568646 cudart_stub.cc:31] Could not find cuda drivers on your machine, GPU will not be used.\n",
"I0000 00:00:1779222070.050194 568646 cpu_feature_guard.cc:227] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\n",
"To enable the following instructions: AVX2 AVX512F AVX512_VNNI AVX512_BF16 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
"I0000 00:00:1779222071.034433 568646 port.cc:153] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.\n",
"I0000 00:00:1779222071.035133 568646 cudart_stub.cc:31] Could not find cuda drivers on your machine, GPU will not be used.\n"
]
}
],
"source": [
"from tensorflow.keras.models import Sequential\n",
"from tensorflow.keras.layers import Dense, Activation\n",
"from tensorflow.keras.regularizers import l2\n",
"from tensorflow.keras.optimizers import SGD\n",
"from tensorflow.keras.utils import plot_model"
]
},
{
"cell_type": "markdown",
"id": "765de834",
"metadata": {},
"source": [
"Setup the model here"
]
},
{
"cell_type": "code",
"execution_count": 2,
"id": "b7603253",
"metadata": {
"collapsed": false,
"execution": {
"iopub.execute_input": "2026-05-19T20:21:12.103888Z",
"iopub.status.busy": "2026-05-19T20:21:12.103622Z",
"iopub.status.idle": "2026-05-19T20:21:12.107781Z",
"shell.execute_reply": "2026-05-19T20:21:12.107320Z"
}
},
"outputs": [],
"source": [
"num_input_nodes = 4\n",
"num_output_nodes = 2\n",
"num_hidden_layers = 1\n",
"nodes_hidden_layer = 64\n",
"l2_val = 1e-5\n",
"\n",
"model = Sequential()"
]
},
{
"cell_type": "markdown",
"id": "22fbee19",
"metadata": {},
"source": [
"Hidden layer 1\n",
"NOTE: Number of input nodes need to be defined in this layer"
]
},
{
"cell_type": "code",
"execution_count": 3,
"id": "1a0fcffe",
"metadata": {
"collapsed": false,
"execution": {
"iopub.execute_input": "2026-05-19T20:21:12.108984Z",
"iopub.status.busy": "2026-05-19T20:21:12.108862Z",
"iopub.status.idle": "2026-05-19T20:21:12.146501Z",
"shell.execute_reply": "2026-05-19T20:21:12.145993Z"
}
},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"/py-venv/ROOT-CI/lib64/python3.12/site-packages/keras/src/layers/core/dense.py:107: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
" super().__init__(activity_regularizer=activity_regularizer, **kwargs)\n",
"E0000 00:00:1779222072.112280 568646 cuda_platform.cc:52] failed call to cuInit: INTERNAL: CUDA error: Failed call to cuInit: UNKNOWN ERROR (303)\n"
]
}
],
"source": [
"model.add(Dense(nodes_hidden_layer, activation='relu', kernel_regularizer=l2(l2_val), input_dim=num_input_nodes))"
]
},
{
"cell_type": "markdown",
"id": "bd6422c3",
"metadata": {},
"source": [
"Hidden layer 2 to num_hidden_layers\n",
"NOTE: Here, you can do what you want"
]
},
{
"cell_type": "code",
"execution_count": 4,
"id": "f5e5f127",
"metadata": {
"collapsed": false,
"execution": {
"iopub.execute_input": "2026-05-19T20:21:12.147830Z",
"iopub.status.busy": "2026-05-19T20:21:12.147706Z",
"iopub.status.idle": "2026-05-19T20:21:12.149845Z",
"shell.execute_reply": "2026-05-19T20:21:12.149401Z"
}
},
"outputs": [],
"source": [
"for k in range(num_hidden_layers-1):\n",
" model.add(Dense(nodes_hidden_layer, activation='relu', kernel_regularizer=l2(l2_val)))"
]
},
{
"cell_type": "markdown",
"id": "350cb7d7",
"metadata": {},
"source": [
"Output layer\n",
"NOTE: Use following output types for the different tasks\n",
"Binary classification: 2 output nodes with 'softmax' activation\n",
"Regression: 1 output with any activation ('linear' recommended)\n",
"Multiclass classification: (number of classes) output nodes with 'softmax' activation"
]
},
{
"cell_type": "code",
"execution_count": 5,
"id": "4a373b0c",
"metadata": {
"collapsed": false,
"execution": {
"iopub.execute_input": "2026-05-19T20:21:12.150967Z",
"iopub.status.busy": "2026-05-19T20:21:12.150849Z",
"iopub.status.idle": "2026-05-19T20:21:12.157977Z",
"shell.execute_reply": "2026-05-19T20:21:12.157517Z"
}
},
"outputs": [],
"source": [
"model.add(Dense(num_output_nodes, activation='softmax'))"
]
},
{
"cell_type": "markdown",
"id": "ff116734",
"metadata": {},
"source": [
"Compile model\n",
"NOTE: Use following settings for the different tasks\n",
"Any classification: 'categorical_crossentropy' is recommended loss function\n",
"Regression: 'mean_squared_error' is recommended loss function"
]
},
{
"cell_type": "code",
"execution_count": 6,
"id": "800faa7b",
"metadata": {
"collapsed": false,
"execution": {
"iopub.execute_input": "2026-05-19T20:21:12.159434Z",
"iopub.status.busy": "2026-05-19T20:21:12.159317Z",
"iopub.status.idle": "2026-05-19T20:21:12.165770Z",
"shell.execute_reply": "2026-05-19T20:21:12.165319Z"
}
},
"outputs": [],
"source": [
"model.compile(loss='categorical_crossentropy', optimizer=SGD(learning_rate=0.01), weighted_metrics=['accuracy',])"
]
},
{
"cell_type": "markdown",
"id": "15a7731a",
"metadata": {},
"source": [
"Save model"
]
},
{
"cell_type": "code",
"execution_count": 7,
"id": "8240cd9f",
"metadata": {
"collapsed": false,
"execution": {
"iopub.execute_input": "2026-05-19T20:21:12.166969Z",
"iopub.status.busy": "2026-05-19T20:21:12.166848Z",
"iopub.status.idle": "2026-05-19T20:21:12.190210Z",
"shell.execute_reply": "2026-05-19T20:21:12.189743Z"
}
},
"outputs": [],
"source": [
"model.save('model.keras')"
]
},
{
"cell_type": "markdown",
"id": "cae29aec",
"metadata": {},
"source": [
"Additional information about the model\n",
"NOTE: This is not needed to run the model"
]
},
{
"cell_type": "markdown",
"id": "8545b43d",
"metadata": {},
"source": [
"Print summary"
]
},
{
"cell_type": "code",
"execution_count": 8,
"id": "b85e1132",
"metadata": {
"collapsed": false,
"execution": {
"iopub.execute_input": "2026-05-19T20:21:12.191583Z",
"iopub.status.busy": "2026-05-19T20:21:12.191459Z",
"iopub.status.idle": "2026-05-19T20:21:12.201383Z",
"shell.execute_reply": "2026-05-19T20:21:12.200936Z"
}
},
"outputs": [
{
"data": {
"text/html": [
"Model: \"sequential\"\n",
"\n"
],
"text/plain": [
"\u001b[1mModel: \"sequential\"\u001b[0m\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
"┃ Layer (type) ┃ Output Shape ┃ Param # ┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
"│ dense (Dense) │ (None, 64) │ 320 │\n",
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
"│ dense_1 (Dense) │ (None, 2) │ 130 │\n",
"└─────────────────────────────────┴────────────────────────┴───────────────┘\n",
"\n"
],
"text/plain": [
"┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
"┃\u001b[1m \u001b[0m\u001b[1mLayer (type) \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
"┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
"│ dense (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m320\u001b[0m │\n",
"├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
"│ dense_1 (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m2\u001b[0m) │ \u001b[38;5;34m130\u001b[0m │\n",
"└─────────────────────────────────┴────────────────────────┴───────────────┘\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
" Total params: 450 (1.76 KB)\n",
"\n"
],
"text/plain": [
"\u001b[1m Total params: \u001b[0m\u001b[38;5;34m450\u001b[0m (1.76 KB)\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
" Trainable params: 450 (1.76 KB)\n",
"\n"
],
"text/plain": [
"\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m450\u001b[0m (1.76 KB)\n"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
" Non-trainable params: 0 (0.00 B)\n",
"\n"
],
"text/plain": [
"\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"model.summary()"
]
},
{
"cell_type": "markdown",
"id": "7677eed6",
"metadata": {},
"source": [
"Visualize model as graph"
]
},
{
"cell_type": "code",
"execution_count": 9,
"id": "c409cd83",
"metadata": {
"collapsed": false,
"execution": {
"iopub.execute_input": "2026-05-19T20:21:12.202614Z",
"iopub.status.busy": "2026-05-19T20:21:12.202477Z",
"iopub.status.idle": "2026-05-19T20:21:12.204916Z",
"shell.execute_reply": "2026-05-19T20:21:12.204494Z"
}
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"You must install pydot (`pip install pydot`) for `plot_model` to work.\n"
]
}
],
"source": [
"try:\n",
" plot_model(model, to_file='model.png', show_shapes=True)\n",
"except:\n",
" print('[INFO] Failed to make model plot')"
]
}
],
"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": "3.12.12"
}
},
"nbformat": 4,
"nbformat_minor": 5
}