#keras question - what's the difference between model(test) and model.predict(test)?
@fchollet #TensorFlow
model() is like a layer call -- in-memory, single-batch, differentiable. Its input and output are tensors. Meanwhile, model.predict() is batched, not differentiable, and may be distributed. Its input may be a Dataset. Its output is a NumPy array.
-
-
Roughly, `predict()` is a `for` loop over the data provided, doing efficient (maybe distributed) inference for each batch, and packing the results into a NumPy array.
-
Is predict() batched when input contains numpy arrays instead of http://tf.data .Dataset?
- Show replies
New conversation -
-
-
If I followed correctly, it means model.predict() can be used with multi processing where model() cannot ?
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.