google-deepmind / google-deepmind/kinetics-i3d
what are input_arrays and output_arrays for the model
- Dominant language
- Python
- Stars
- 1.8k
- Forks
- 467
- PR merge metrics
- No merged PRs in 30d
Description
hi. iam trying to export the model into tensorflow lite but i don't know what are the input_arrays and output arrays required for the conversion method
converter = tf.lite.TFLiteConverter.from_frozen_graph(
graph_def_file, input_arrays, output_arrays)
one more issue how can i save the model from a module as tflite. I am getting a graph error all the time
m = hub.Module("https://tfhub.dev/deepmind/i3d-kinetics-400/1")
INPUT_SIZE = (None, 224, 224, 3)
model = tf.keras.Sequential([
hub.KerasLayer(m,output_shape=[400], dtype=tf.float32, trainable=False),
tf.keras.layers.Dropout(rate=0.2)
])
Contributor guide
Assessment
This issue has not been assessed yet.