[BUG]: represent not working with facenet

[lemnet]

Before You Report a Bug, Please Confirm You Have Done The Following...

• [X] I have updated to the latest version of the packages.
• [X] I have searched for both existing issues and closed issues and found none that matched my issue.

DeepFace's version

0.0.92

Python version

3.12

Operating System

Ubuntu 24.04

Dependencies

tensorflow 2.16.1 keras 3.4.0

Reproducible example

embedding = DeepFace.represent(img_path = file,model_name = 'Facenet')

Relevant Log Output

error: libdevice not found at ./libdevice.10.bc 2024-06-25 17:45:44.871399: E tensorflow/compiler/mlir/tools/kernel_gen/tf_framework_c_interface.cc:207] INTERNAL: Generating device code failed.

[...]

tensorflow.python.framework.errors_impl.UnknownError: Exception encountered when calling layer 'Bottleneck_BatchNorm' (type BatchNormalization).

{{function_node wrappedRsqrtdevice/job:localhost/replica:0/task:0/device:GPU:0}} JIT compilation failed. [Op:Rsqrt] name:

Call arguments received by layer 'Bottleneck_BatchNorm' (type BatchNormalization): • inputs=tf.Tensor(shape=(1, 128), dtype=float32) • training=False • mask=None

Expected Result

No response

What happened instead?

No response

Additional Info

works with default model embedding = DeepFace.represent(img_path = file)

[lemnet]

My bad... deepface was not the issue ! export XLA_FLAGS=--xla_gpu_cuda_data_dir=/usr/lib/cuda solved my issue