Gettings started - Install Keras and TensorFlow for R
You can install the Keras for R package from CRAN as follows:
You can install the Keras for R package from CRAN as follows:
from sklearn import datasetsfrom sklearn.model_selection import train_test_splitimport numpy as npfrom tensorflow.keras.models import Sequentialfrom tensorflow.keras.layers import Dense
2026-03-17 14:31:55.243056: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
2026-03-17 14:31:55.243078: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered
2026-03-17 14:31:55.244012: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered
2026-03-17 14:31:55.249183: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
2026-03-17 14:31:55.873505: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
Load the Iris dataset
# Load iris datasetiris = datasets.load_iris()X = iris.data[:, :4] # We only take the first four features.y = iris.target# Print the first 5 rows of the iris dataprint(X[:5])# Print the first 5 rows of the iris targetprint(y[:5])
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
Build one layer neural network with 10 nodes to classify the type of iris flowers using the iris dataset
model = Sequential([ Dense(10, activation='relu', input_shape=(4,)), Dense(3, activation='softmax')])
/home/esuess/.virtualenvs/r-keras-gpu/lib/python3.10/site-packages/keras/src/layers/core/dense.py:95: 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.
super().__init__(activity_regularizer=activity_regularizer, **kwargs)
2026-03-17 14:31:56.375287: I external/local_xla/xla/stream_executor/cuda/cuda_executor.cc:901] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355
2026-03-17 14:31:56.375464: I external/local_xla/xla/stream_executor/cuda/cuda_executor.cc:901] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355
2026-03-17 14:31:56.407785: I external/local_xla/xla/stream_executor/cuda/cuda_executor.cc:901] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355
2026-03-17 14:31:56.407954: I external/local_xla/xla/stream_executor/cuda/cuda_executor.cc:901] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355
2026-03-17 14:31:56.408080: I external/local_xla/xla/stream_executor/cuda/cuda_executor.cc:901] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355
2026-03-17 14:31:56.408196: I external/local_xla/xla/stream_executor/cuda/cuda_executor.cc:901] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355
2026-03-17 14:31:56.523436: I external/local_xla/xla/stream_executor/cuda/cuda_executor.cc:901] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355
2026-03-17 14:31:56.523622: I external/local_xla/xla/stream_executor/cuda/cuda_executor.cc:901] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355
2026-03-17 14:31:56.523745: I external/local_xla/xla/stream_executor/cuda/cuda_executor.cc:901] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355
2026-03-17 14:31:56.523857: I external/local_xla/xla/stream_executor/cuda/cuda_executor.cc:901] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355
2026-03-17 14:31:56.523966: I external/local_xla/xla/stream_executor/cuda/cuda_executor.cc:901] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355
2026-03-17 14:31:56.524076: I external/local_xla/xla/stream_executor/cuda/cuda_executor.cc:901] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355
2026-03-17 14:31:56.533502: I external/local_xla/xla/stream_executor/cuda/cuda_executor.cc:901] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355
2026-03-17 14:31:56.534344: I external/local_xla/xla/stream_executor/cuda/cuda_executor.cc:901] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355
2026-03-17 14:31:56.534469: I external/local_xla/xla/stream_executor/cuda/cuda_executor.cc:901] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355
2026-03-17 14:31:56.534584: I external/local_xla/xla/stream_executor/cuda/cuda_executor.cc:901] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355
2026-03-17 14:31:56.534702: I external/local_xla/xla/stream_executor/cuda/cuda_executor.cc:901] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355
2026-03-17 14:31:56.534784: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1929] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 4567 MB memory: -> device: 0, name: NVIDIA GeForce GTX 1080, pci bus id: 0000:01:00.0, compute capability: 6.1
2026-03-17 14:31:56.535064: I external/local_xla/xla/stream_executor/cuda/cuda_executor.cc:901] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355
2026-03-17 14:31:56.535146: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1929] Created device /job:localhost/replica:0/task:0/device:GPU:1 with 3570 MB memory: -> device: 1, name: NVIDIA P106-100, pci bus id: 0000:06:00.0, compute capability: 6.1
2026-03-17 14:31:56.735347: I external/local_tsl/tsl/platform/default/subprocess.cc:304] Start cannot spawn child process: No such file or directory
2026-03-17 14:31:57.533650: I external/local_xla/xla/service/service.cc:168] XLA service 0x564e8dc3e3b0 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices:
2026-03-17 14:31:57.533675: I external/local_xla/xla/service/service.cc:176] StreamExecutor device (0): NVIDIA GeForce GTX 1080, Compute Capability 6.1
2026-03-17 14:31:57.533680: I external/local_xla/xla/service/service.cc:176] StreamExecutor device (1): NVIDIA P106-100, Compute Capability 6.1
2026-03-17 14:31:57.547739: I tensorflow/compiler/mlir/tensorflow/utils/dump_mlir_util.cc:269] disabling MLIR crash reproducer, set env var `MLIR_CRASH_REPRODUCER_DIRECTORY` to enable.
2026-03-17 14:31:57.616624: I external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:454] Loaded cuDNN version 8904
WARNING: All log messages before absl::InitializeLog() is called are written to STDERR
I0000 00:00:1773783118.042740 371572 device_compiler.h:186] Compiled cluster using XLA! This line is logged at most once for the lifetime of the process.
<keras.src.callbacks.history.History at 0x7f4fa07451b0>
from sklearn.metrics import confusion_matriximport matplotlib.pyplot as pltcm = confusion_matrix(y_test, np.argmax(predictions, axis=1))plt.imshow(cm, cmap='binary')plt.show()
Summary
In this post, we learned how to build a simple neural network using Keras and TensorFlow for R. We used the iris dataset to classify the type of iris flowers. We built a one-layer neural network with 10 nodes and used the ReLU activation function. We compiled the model using the Adam optimizer and sparse categorical cross-entropy loss function. We trained the model for 100 epochs and evaluated the model using the test data. Finally, we made predictions using the model and printed the predicted classes.
