558694 (2) [Avatar] Offline
#1
I have no idea what this error is:


> library(keras)
> imdb <- dataset_imdb(num_words = 10000)
/Users/Karsa/.virtualenvs/r-tensorflow/lib/python2.7/site-packages/h5py/__init__.py:36: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.
from ._conv import register_converters as _register_converters
Using TensorFlow backend.

*** caught illegal operation ***
address 0x1148b9a1b, cause 'illegal opcode'

Traceback:
1: .Call(`_reticulate_py_module_import`, module, convert)
2: py_module_import(module, convert = convert)
3: import(module)
4: doTryCatch(return(expr), name, parentenv, handler)
5: tryCatchOne(expr, names, parentenv, handlers[[1L]])
6: tryCatchList(expr, classes, parentenv, handlers)
7: tryCatch(import(module), error = clear_error_handler())
8: py_resolve_module_proxy(x)
9: `$.python.builtin.module`(keras, datasets)
10: keras$datasets
11: dataset_imdb(num_words = 10000)

Possible actions:
1: abort (with core dump, if enabled)
2: normal R exit
3: exit R without saving workspace
4: exit R saving workspace
Selection:
256385 (46) [Avatar] Offline
#2
This is very likely due to the most recent version of TensorFlow (v1.7) building compiled against CPU instructions that don't work on older CPU models. You can actually install an older version of TensorFlow that doesn't have this issue as follows:

library(keras)
install_keras(tensorflow = "1.4")
558694 (2) [Avatar] Offline
#3
You are absolutely right, this fixed the problem. Thanks a million!