clinical and full ECG view from partners ecg (#211)

* clinical and full view ECG plotting

* add msttcorefonts to docker image

* tmaps

* import order, top panel func name, remove constrained layout warning

* Pre-requirements not to have errors

* Better comment for pre_requirements

* period

Co-authored-by: Paolo Di Achille <[email protected]>

Author: StevenSong

docker/vm_boot_images/Dockerfile

@@ -33,4 +33,5 @@ RUN apt-get install python3-tk libgl1-mesa-glx libxt-dev -y
 
 # Requirements for the tensorflow project
 RUN pip3 install --upgrade pip
+RUN pip3 install -r pre_requirements.txt
 RUN pip3 install -r tensorflow-requirements.txt

docker/vm_boot_images/config/pre_requirements.txt

@@ -0,0 +1,14 @@
+# These are packages required to install the packages in tensorflow_requirements.txt.
+# pip currently fails to recognize them as necessary for installation.
+cvxopt
+cvxpy>=1.0
+cython>=0.29
+numexpr
+numpy
+osqp!=0.6.0,!=0.6.1
+pandas>=0.21
+scipy!=1.3.0,>=1.0
+scikit-learn<0.23,>=0.22.0
+pytest
+pytest-cov
+setuptools_scm

docker/vm_boot_images/config/ubuntu.sh

@@ -2,4 +2,5 @@
 
 # Other necessities
 apt-get update
-apt-get install -y wget unzip curl python-pydot python-pydot-ng graphviz
+echo "ttf-mscorefonts-installer msttcorefonts/accepted-mscorefonts-eula select true" | debconf-set-selections
+apt-get install -y wget unzip curl python-pydot python-pydot-ng graphviz ttf-mscorefonts-installer

ml4cvd/arguments.py

@@ -200,6 +200,7 @@ def parse_args():
     parser.add_argument('--inspect_model', default=False, action='store_true', help='Plot model architecture, measure inference and training speeds.')
     parser.add_argument('--inspect_show_labels', default=True, action='store_true', help='Plot model architecture with labels for each layer.')
     parser.add_argument('--alpha', default=0.5, type=float, help='Alpha transparency for t-SNE plots must in [0.0-1.0].')
+    parser.add_argument('--plot_mode', default='clinical', choices=['clinical', 'full'], help='ECG view to plot for partners ECGs.')
 
     # Training optimization options
     parser.add_argument('--num_workers', default=multiprocessing.cpu_count(), type=int, help="Number of workers to use for every tensor generator.")