search

Deepshift / DeepCreamPy

deeppomf's DeepCreamPy + some updates
GNU Affero General Public License v3.0
424 stars 74 forks source link

Updating dependencies #5

Open naphteine opened 1 year ago

naphteine commented 1 year ago

I am planning to slowly start working on updating dependencies. They're too old and it's kind of hard to work on this project right now. Does anyone have any issue with this? Thank you.

alexanderj1988 commented 1 year ago

highly appreciated. just found this repository and wanted to test it (mac m1) which seems to be impossible to work with tensorflow 1

(if you need someone to test it on mac just notify me

naphteine commented 1 year ago

highly appreciated. just found this repository and wanted to test it (mac m1) which seems to be impossible to work with tensorflow 1

(if you need someone to test it on mac just notify me

Can you try v2.3.0 branch? It kind of working on Mermaid image but fails on some points.

mermaid_censored 1

alexanderj1988 commented 1 year ago

the library shiboken2>=5.13.2 doesn't seem to be satisfiable. conda environment python 3.9 arm64

decensoring i get the same results

regards

mermaid_censored 0

Edit:

i get the following output while loading the model

/Users/xxx/miniforge3/envs/pai/lib/python3.9/site-packages/tensorflow/python/keras/engine/base_layer_v1.py:1697: UserWarning: `layer.apply` is deprecated and will be removed in a future version. Please use `layer.__call__` method instead.
  warnings.warn('`layer.apply` is deprecated and '
2023-07-12 18:15:05.521587: I tensorflow/compiler/mlir/mlir_graph_optimization_pass.cc:375] MLIR V1 optimization pass is not enabled
2023-07-12 18:15:07.592851: W tensorflow/core/common_runtime/graph_constructor.cc:1553] Importing a graph with a lower producer version 38 into an existing graph with producer version 1482. Shape inference will have run different parts of the graph with different producer versions.
load model finished

Other images dont work entierly. It seems the detections seem to work

Decensoring the image ./decensor_input/1.png

Found 9 censored regions in this image!
Found 9 masked regions
1 out of 9 regions decensored.
2 out of 9 regions decensored.
3 out of 9 regions decensored.
4 out of 9 regions decensored.
5 out of 9 regions decensored.
6 out of 9 regions decensored.
7 out of 9 regions decensored.
8 out of 9 regions decensored.
9 out of 9 regions decensored.
current image finished
Decensored image saved to ./decensor_output/1 0.png!

The decensoring itself seems to fail

1

naphteine commented 1 year ago

Thanks @alexanderj1988. I will look into it when I'm free