search

broadinstitute / neural-profiling

1 stars 2 forks source link

Archived Wikis #7

Open michaelbornholdt opened 3 years ago

michaelbornholdt commented 3 years ago

If using Docker, ignore the below and just pull the docker image. See other Wiki page. Otherwise, follow the below instructions.

Step by Step

  1. Read the DeepProfiler Wiki
  2. Do not use conda. Install and use virtual env and pip instead.
  3. Make sure your pip is up to date.
  4. I am not 100% sure, but python 3.6 seems to be the safer option than 3.8.X since you will be running an old tensorflow (TF)
  5. git clone the DeepProfiler (DP) repo and then pip install -e
  6. Manually add pip install tqdm (or, better, add it to the requirement.txt)
  7. Manually change the version of tensorflow-gpu, h5py, efficientnet and keras efficientnet == 1.1.0; keras == 2.2.5; tensorflow-gpu == 1.15.2 and h5py == 2.10.0
  8. Check if your pip list is the same as mine, you should be in a good spot then
  9. Change your config to { "train": { "sampling": "cache_size": 10 }}
  10. Run the demo to see if you have succeeded.

Full pip list

absl-py              0.12.0
astor                0.8.1
attrs                21.2.0
beautifulsoup4       4.9.3
cached-property      1.5.2
certifi              2020.12.5
chardet              4.0.0
click                7.1.2
comet-ml             3.9.1
configobj            5.0.6
cycler               0.10.0
decorator            4.4.2
deepprofiler         0.1.0     /home/ubuntu/work/DeepProfiler
dulwich              0.20.21
efficientnet         1.1.0
everett              1.0.3
gast                 0.2.2
google-pasta         0.2.0
GPy                  1.9.9
GPyOpt               1.2.6
grpcio               1.37.1
h5py                 2.10.0
idna                 2.10
imageio              2.9.0
importlib-metadata   4.0.1
joblib               1.0.1
jsonschema           3.2.0
Keras                2.2.5
Keras-Applications   1.0.8
Keras-Preprocessing  1.1.2
kiwisolver           1.3.1
lxml                 4.6.3
Markdown             3.3.4
matplotlib           3.4.1
networkx             2.5.1
numpy                1.20.2
nvidia-ml-py3        7.352.0
opt-einsum           3.3.0
pandas               1.2.4
paramz               0.9.5
Pillow               8.2.0
pip                  21.1.1
protobuf             3.16.0
pyparsing            2.4.7
pyrsistent           0.17.3
python-dateutil      2.8.1
pytz                 2021.1
PyWavelets           1.1.1
PyYAML               5.4.1
requests             2.25.1
requests-toolbelt    0.9.1
scikit-image         0.18.1
scikit-learn         0.24.2
scipy                1.6.3
setuptools           56.1.0
six                  1.16.0
soupsieve            2.2.1
tensorboard          1.15.0
tensorflow-estimator 1.15.1
tensorflow-gpu       1.15.2
termcolor            1.1.0
threadpoolctl        2.1.0
tifffile             2021.4.8
tqdm                 4.60.0
typing-extensions    3.10.0.0
urllib3              1.26.4
websocket-client     0.59.0
Werkzeug             1.0.1
wheel                0.36.2
wrapt                1.12.1
wurlitzer            2.1.0
zipp                 3.4.1

Other comments

When running the demo data from the Wiki: https://github.com/cytomining/DeepProfiler/wiki/2.-Project-structure You need to add cache size to the config! { "train": { "sampling": "cache_size": 10 }}

Confusing things about installing DP

  1. Explain what TF is needed for what version! Link to the TF sites
  2. TQDM is still missing
  3. The example config has missing cache size and it has a checkpoint which creates an error! need to set checkpoint to None

Installing DP2

  1. Same problems as above.
  2. Nothing otherwise.