System requirements clarification

[rokbra]

Hi,

I recently tried out awesome UNET-defmapping with mostly up to date packages (see below), however with no luck. I wonder maybe a little clarification on originally used package names and versions would solve all the issues I encountered tinkering around and trying to make this amazing code to work in my environment.

Python 3.7.10 ? Keras 2.4.3 ? Tensorflow 2.3.0 ? sklearn: scikit-learn 0.24.2 rasterio 1.1.5 ? rkimage ? fiona 1.8.9.post2 ? cv2: opencv-python numpy-indexed 0.3.5 sentinelsat 1.0.0 zipfile.py ? glob2 0.7 matplotlib 3.4.2

I'd be very grateful for any kind of assistance.

[bragagnololu]

Hi,

Here are the versions of the libraries used in the development of the algorithm:

keras-base=2.3.1 keras-gpu=2.3.1 tensorflow=2.1.0 tensorflow-base=2.1.0 tensorflow-gpu=2.1.0 scikit-learn=0.22.2.post1 rasterio=1.1.3 fiona=1.8.13 opencv=4.2.0 numpy=1.18.1 numpy-indexed sentinelhub=3.0.2 sentinelsat=0.13

zipfile.py: https://github.com/python/cpython/blob/3.9/Lib/zipfile.py / https://docs.python.org/3/library/zipfile.html

Let me know if any problem occur!

[rokbra]

I highly appreciate your quick response. I tested the algorithm on new environment with described libraries and came to conclusion that I simply lacked a proper GPU... And so I tested it out on another machine, and even with up to date libraries everything worked perfectly after some tinkering: model_UNET_def.py line 58: model = tf.keras.Model(inputs = inputs, outputs = conv10) and increase of swap memory (linux) from 2GB to ~10GB - because kernel kept crashing due to numpy zeros function

Thank you and I hope you will get the best possible evaluations for such amazing work. :)