.. | ||
__init__.py | ||
activation_benchmark.py | ||
attention_benchmark.py | ||
base_benchmark.py | ||
conv_benchmark.py | ||
core_benchmark.py | ||
merge_benchmark.py | ||
normalization_benchmark.py | ||
pooling_benchmark.py | ||
README.md | ||
regularization_benchmark.py | ||
reshaping_benchmark.py | ||
rnn_benchmark.py |
Benchmark the layer performance
This directory contains benchmarks to compare the performance of
keras.layers.XXX
and tf.keras.layers.XXX
. We compare the performance of
both the forward pass and train step (forward & backward pass).
To run the benchmark, use the command below and change the flags according to your target:
python3 -m benchmarks.layer_benchmark.conv_benchmark \
--benchmark_name=benchmark_conv2D \
--num_samples=2048 \
--batch_size=256 \
--jit_compile=True