Sionna modules (such as maping, utils, etc.) are not defined

[HarryAla]

I am very new with sionna and started to work with it. Once I try the tutorial codes, it return me errors such as "sionna is not defined", or "utils module is no a defined attribute for sionna".
I am not sure if anyone may help?
I installed sionna in Ubuntu 22.04, and use python 3.10. I installed TensorFlow, etc. Everything works inside VirtualBox as a virtual machine.

[merlinND]

Hello @HarryAla,

Is it possible that the Python version that you are using to run the tutorial is a different version than the one you used to install Sionna?
To avoid confusion, I like to always explicitly use the python3 binary rather than the various aliases for modules (pip, jupyterlab, etc), since they might implicitly use another Python version:

$ python3 -m pip install sionna
$ python3
>>> import sionna
>>> print(sionna.__file__)
>>> exit()
$ python3 -m jupyterlab .  # For example

[HarryAla]

Hi Merlin, I really appreciate your e-mail. To be honest, I tried it and after the following command, it gave me lots of unexpected warnings / notices / errors. It is too confusing. I installed Ubuntu 22.04 on VirtualBox, then "sudo apt-get update" & "sudo apt-get upgrade". Then, "sudo apt-get install pip". Then, "python3 -m pip install sionna". Then, the followings after typing "import sionna" inside "python3" command line. Very respectfully, Harry ```

> import sionna

2024-03-21 16:45:13.004835: I external/local_tsl/tsl/cuda/cudart_stub.cc:31] Could not find cuda drivers on your machine, GPU will not be used. 2024-03-21 16:45:13.120287: 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 2024-03-21 16:45:13.120399: 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 2024-03-21 16:45:13.123884: 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 2024-03-21 16:45:13.140550: I external/local_tsl/tsl/cuda/cudart_stub.cc:31] Could not find cuda drivers on your machine, GPU will not be used. 2024-03-21 16:45:13.140851: 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, in other operations, rebuild TensorFlow with the appropriate compiler flags. 2024-03-21 16:45:14.490891: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT jit_llvm_init(): your CPU does not support the `fma` instruction set, shutting down the LLVM backend... Traceback (most recent call last): File "/home/harry/.local/lib/python3.10/site-packages/mitsuba/__init__.py", line 107, in __getattribute__ _import('mitsuba.mitsuba_' + variant + '_ext'), File "/usr/lib/python3.10/importlib/__init__.py", line 126, in import_module return _bootstrap._gcd_import(name[level:], package, level) File "<frozen importlib._bootstrap>", line 1050, in _gcd_import File "<frozen importlib._bootstrap>", line 1027, in _find_and_load File "<frozen importlib._bootstrap>", line 1006, in _find_and_load_unlocked File "<frozen importlib._bootstrap>", line 674, in _load_unlocked File "<frozen importlib._bootstrap>", line 571, in module_from_spec File "<frozen importlib._bootstrap_external>", line 1176, in create_module File "<frozen importlib._bootstrap>", line 241, in _call_with_frames_removed ImportError: jit_init_thread_state(): the LLVM backend is inactive because the LLVM shared library ("libLLVM.so") could not be found! Set the DRJIT_LIBLLVM_PATH environment variable to specify its path. During handling of the above exception, another exception occurred: Traceback (most recent call last): File "<stdin>", line 1, in <module> File "/home/harry/.local/lib/python3.10/site-packages/sionna/__init__.py", line 19, in <module> from . import rt File "/home/harry/.local/lib/python3.10/site-packages/sionna/rt/__init__.py", line 24, in <module> mi.set_variant('llvm_ad_rgb') File "/home/harry/.local/lib/python3.10/site-packages/mitsuba/__init__.py", line 317, in set_variant _import('mitsuba.ad.integrators') File "/usr/lib/python3.10/importlib/__init__.py", line 126, in import_module return _bootstrap._gcd_import(name[level:], package, level) File "/home/harry/.local/lib/python3.10/site-packages/mitsuba/python/ad/__init__.py", line 2, in <module> from .integrators import * File "/home/harry/.local/lib/python3.10/site-packages/mitsuba/python/ad/integrators/__init__.py", line 25, in <module> importlib.import_module('mitsuba.ad.integrators.' + name) File "/usr/lib/python3.10/importlib/__init__.py", line 126, in import_module return _bootstrap._gcd_import(name[level:], package, level) File "/home/harry/.local/lib/python3.10/site-packages/mitsuba/python/ad/integrators/prb_basic.py", line 6, in <module> from .common import RBIntegrator File "/home/harry/.local/lib/python3.10/site-packages/mitsuba/python/ad/integrators/common.py", line 8, in <module> class ADIntegrator(mi.CppADIntegrator): File "/home/harry/.local/lib/python3.10/site-packages/mitsuba/__init__.py", line 253, in __getattribute__ result = module.__getattribute__(key) File "/home/harry/.local/lib/python3.10/site-packages/mitsuba/__init__.py", line 115, in __getattribute__ raise AttributeError(e) AttributeError: jit_init_thread_state(): the LLVM backend is inactive because the LLVM shared library ("libLLVM.so") could not be found! Set the DRJIT_LIBLLVM_PATH environment variable to specify its path. ```

[merlinND]

Great, I think that you are almost there. Assuming that you are not trying to use a GPU, it should start working on the CPU if you further install LLVM: sudo apt install llvm

[HarryAla]

Merlin, I appreciate your helpful email. Now, after importing sionna using "import sionna" inside python3 environment, for the attached simple piece of code, it returns the following error message, that *it does not identify the "mapping" attribute of sionna*. Very respectfully, Harry *The error message*: ```py /usr/lib/python3/dist-packages/scipy/__init__.py:146: UserWarning: A NumPy version >=1.17.3 and <1.25.0 is required for this version of SciPy (detected version 1.26.4 warnings.warn(f"A NumPy version >={np_minversion} and <{np_maxversion}" /usr/local/lib/python3.10/dist-packages/matplotlib/projections/__init__.py:63: UserWarning: Unable to import Axes3D. This may be due to multiple versions of Matplotlib being installed (e.g. as a system package and as a pip package). As a result, the 3D projection is not available. warnings.warn("Unable to import Axes3D. This may be due to multiple versions of " Traceback (most recent call last): File "/home/halasti/Sionna_Test2.py", line 34, in <module> constellation = sn.mapping.Constellation("qam", NUM_BITS_PER_SYMBOL) AttributeError: module 'sionna' has no attribute 'mapping' ```

[merlinND]

Hello @HarryAla,

Could you please try the following "isolated" setup? I've just tried in on my machine and it worked for me. The goal is to avoid any interference from existing installations.

$ mkdir sionna-test
$ cd sionna-test
$ python3 -m venv .venv
$ source ./.venv/bin/activate
$ python -m pip install sionna
$ python
>>> import sionna as sn
>>> constellation = sn.mapping.Constellation("qam", 8)

[musahusseini]

Hello:

I am having a similar issue with sionna module error. I have been trying to install sionna simulation environment on a virtual Linux system (Ubuntu 22.04) and I keep running into this problem:

ModuleNotFoundError Traceback (most recent call last)
in <cell line: 2>()
1 import tensorflow as tf
----> 2 import sionna as sn

Below is the code snippet I am trying to run to model RF communication:

import tensorflow as tf
import sionna as sn

Set up the Sionna simulation environment
sn.set_backend(“tensorflow”)

Define the parameters for the RF communication model using tensorflow for data training:
num_samples = 1000

num_channels = 2

input_shape = (num_samples, num_channels)

output_shape = (num_samples,)

Create the input and output tensors for training data
inputs = tf.random.normal(input_shape)

outputs = tf.random.normal(output_shape)

Define the RF communication model using Sionna layers
model = sn.Sequential([

sn.Dense(64, activation=“relu”, input_shape=input_shape),

sn.Dense(64, activation=“relu”),

sn.Dense(1)

])

Compile the model with an optimizer and loss function
model.compile(optimizer=“adam”, loss=“mse”)

Train the model on the training data
model.fit(inputs, outputs, epochs=10)

Generate predictions using the trained model
predictions = model.predict(inputs)

print(predictions)

[jhoydis]

Hi @musahusseini,

I looks as if you are using Sionna as if it was Keras. There is no such thing as sionna.set_backend or sionna.Sequential.

[musahusseini]

Thank you, @jhoydis. I went ahead and updated the code to Keras and was able to capture the patterns in the RF signal data (I still need to more layers to increase the model’s ability to learn complex patterns).

# Define the parameters for the RF communication model
num_samples = 1000
num_channels = 2
input_shape = (num_channels,)  # Keras expects shape excluding the batch size
output_shape = (num_samples,)

# Create the input and output tensors for training data
inputs = tf.random.normal((num_samples, num_channels))
outputs = tf.random.normal((num_samples, 1))  # Match the final output shape

# Define the RF communication model using Keras layers
model = tf.keras.Sequential([
    tf.keras.layers.Dense(64, activation="relu", input_shape=input_shape),
    tf.keras.layers.Dense(64, activation="relu"),
    tf.keras.layers.Dense(1)
])

# Compile the model with an optimizer and loss function
model.compile(optimizer="adam", loss="mse")

# Train the model on the training data
model.fit(inputs, outputs, epochs=10)

# Generate predictions using the trained model
predictions = model.predict(inputs)

print(predictions)

[musahusseini]

Hello again:

Is it possible to install sionna.set_backend and sionna.Sequential on Google Colab? I was able to run the simulation using Keras, but not Sionna.

[jhoydis]

Hi @musahusseini,

There is no such thing as sionna.set_backend or sionna.Sequential. These are all Keras functionalities.