### Author:  [@monty](https://discourse.julialang.org/u/monty)
#### Post date: [December 8, 2018, 12:42am UTC](https://discourse.julialang.org/t/converting-pyobject-to-a-julia-array/18445/1 "2018-12-08T00:42:24Z")
How to convert PyObject into a Julia array
```julia
import PyCall
tf = PyCall.pyimport("tensorflow")
o = tf[:random_normal]((3,3))
PyObject
```
---
### Author:  [@kristoffer.carlsson](https://discourse.julialang.org/u/kristoffer.carlsson)
#### Post date: [December 8, 2018, 12:56am UTC](https://discourse.julialang.org/t/converting-pyobject-to-a-julia-array/18445/2 "2018-12-08T00:56:56Z")
Friendly nudge: [PSA: how to quote code with backticks](https://discourse.julialang.org/t/psa-how-to-quote-code-with-backticks/7530)
---
### Author:  [@tkf](https://discourse.julialang.org/u/tkf)
#### Post date: [December 8, 2018, 1:22am UTC](https://discourse.julialang.org/t/converting-pyobject-to-a-julia-array/18445/3 "2018-12-08T01:22:28Z")
Convert it to a numpy array [python - Convert a tensor to numpy array in Tensorflow? - Stack Overflow](https://stackoverflow.com/questions/34097281/how-can-i-convert-a-tensor-into-a-numpy-array-in-tensorflow)
PyCall then automatically converts it to a Julia’s `Array` (provided that element type can be converted).
---
### Author:  [@monty](https://discourse.julialang.org/u/monty)
#### Post date: [December 8, 2018, 1:51am UTC](https://discourse.julialang.org/t/converting-pyobject-to-a-julia-array/18445/4 "2018-12-08T01:51:18Z")
it does not seem to work:
```julia
import PyCall
tf = PyCall.pyimport("tensorflow")
o = tf[:random_normal]((3,3))[:eval]
PyObject
```
---
### Author:  [@kristoffer.carlsson](https://discourse.julialang.org/u/kristoffer.carlsson)
#### Post date: [December 8, 2018, 1:56am UTC](https://discourse.julialang.org/t/converting-pyobject-to-a-julia-array/18445/5 "2018-12-08T01:56:13Z")
Don’t you need to execute the `eval` function as well?
---
### Author:  [@monty](https://discourse.julialang.org/u/monty)
#### Post date: [December 8, 2018, 3:36pm UTC](https://discourse.julialang.org/t/converting-pyobject-to-a-julia-array/18445/6 "2018-12-08T15:36:59Z")
actually this is what works:
```julia
o = tf[:random_normal]((3,3))[:numpy]()
3×3 Array{Float32,2}:
-1.35462 -0.0406312 -0.993727
-0.8862 1.15086 -1.85081
0.461948 -1.41717 -0.993305
```
Thank you very much!
---
### Author:  [@wookyoung](https://discourse.julialang.org/u/wookyoung)
#### Post date: [December 9, 2018, 1:36am UTC](https://discourse.julialang.org/t/converting-pyobject-to-a-julia-array/18445/7 "2018-12-09T01:36:46Z")
I found the answer from the issue on the TensorFlow-Examples.
[https://github.com/aymericdamien/TensorFlow-Examples/issues/40#issuecomment-405892316](https://github.com/aymericdamien/TensorFlow-Examples/issues/40#issuecomment-405892316)
so using PyCall in that way.
```julia
using PyCall
tf = pyimport("tensorflow")
o = tf[:random_normal]((3,3))
@pyimport tensorflow.python.keras.backend as K
sess = K.get_session()
array = sess[:run](o)
```
and by adding `Base.getproperty` to `PyObject`.
```julia
Base.getproperty(o::PyObject, name::Symbol) = (:o == name ? getfield : getindex)(o, name)
array = sess.run(o)
```
now it gets more python-like code.