Update yaml dec6

.jupyter/lab/workspaces/auto-z-87ad.jupyterlab-workspace

@@ -1 +1 @@
-{"data":{"layout-restorer:data":{"main":{"dock":{"type":"tab-area","currentIndex":1,"widgets":["terminal:1","notebook:analysis.ipynb"]},"current":"notebook:analysis.ipynb"},"down":{"size":0,"widgets":[]},"left":{"collapsed":false,"visible":true,"current":"filebrowser","widgets":["filebrowser","running-sessions","@jupyterlab/toc:plugin","extensionmanager.main-view"],"widgetStates":{"jp-running-sessions":{"sizes":[0.16666666666666666,0.16666666666666666,0.16666666666666666,0.16666666666666666,0.16666666666666666,0.16666666666666666],"expansionStates":[false,false,false,false,false,false]},"extensionmanager.main-view":{"sizes":[0.5,0.5],"expansionStates":[false,false]}}},"right":{"collapsed":true,"visible":true,"widgets":["jp-property-inspector","debugger-sidebar"],"widgetStates":{"jp-debugger-sidebar":{"sizes":[0.2,0.2,0.2,0.2,0.2],"expansionStates":[false,false,false,false,false]}}},"relativeSizes":[0.2134124336766542,0.7865875663233458,0],"top":{"simpleVisibility":true}},"docmanager:recents":{"opened":[{"path":"","contentType":"directory","root":"/home/jovyan"},{"path":"analysis.ipynb","contentType":"notebook","factory":"Notebook","root":"/home/jovyan"}],"closed":[]},"terminal:1":{"data":{"name":"1"}},"file-browser-filebrowser:cwd":{"path":""},"notebook:analysis.ipynb":{"data":{"path":"analysis.ipynb","factory":"Notebook"}}},"metadata":{"id":"auto-Z"}}
+{"data":{"layout-restorer:data":{"main":{"dock":{"type":"tab-area","currentIndex":1,"widgets":["terminal:1","notebook:analysis.ipynb"]},"current":"notebook:analysis.ipynb"},"down":{"size":0,"widgets":[]},"left":{"collapsed":false,"visible":true,"current":"filebrowser","widgets":["filebrowser","running-sessions","@jupyterlab/toc:plugin","extensionmanager.main-view"],"widgetStates":{"jp-running-sessions":{"sizes":[0.16666666666666666,0.16666666666666666,0.16666666666666666,0.16666666666666666,0.16666666666666666,0.16666666666666666],"expansionStates":[false,false,false,false,false,false]},"extensionmanager.main-view":{"sizes":[0.5,0.5],"expansionStates":[false,false]}}},"right":{"collapsed":true,"visible":true,"widgets":["jp-property-inspector","debugger-sidebar"],"widgetStates":{"jp-debugger-sidebar":{"sizes":[0.2,0.2,0.2,0.2,0.2],"expansionStates":[false,false,false,false,false]}}},"relativeSizes":[0.21248397684897452,0.7875160231510255,0],"top":{"simpleVisibility":true}},"docmanager:recents":{"opened":[{"path":"","contentType":"directory","root":"/home/jovyan"},{"path":"analysis.ipynb","contentType":"notebook","factory":"Notebook","root":"/home/jovyan"},{"path":"data_cleaning_script.py","contentType":"file","factory":"Editor","root":"/home/jovyan"},{"path":"data_validation_script.py","contentType":"file","factory":"Editor","root":"/home/jovyan"}],"closed":[{"path":"data_validation_script.py","contentType":"file","factory":"Editor","root":"/home/jovyan"},{"path":"data_cleaning_script.py","contentType":"file","factory":"Editor","root":"/home/jovyan"}]},"terminal:1":{"data":{"name":"1"}},"file-browser-filebrowser:cwd":{"path":"data"},"notebook:analysis.ipynb":{"data":{"path":"analysis.ipynb","factory":"Notebook"}}},"metadata":{"id":"auto-Z"}}

Dockerfile

@@ -7,4 +7,4 @@ RUN mamba update --quiet --file /tmp/conda-linux-64.lock \
     && fix-permissions "${CONDA_DIR}" \
     && fix-permissions "/home/${NB_USER}" 
 
-RUN pip install deepchecks==0.18.1
+RUN pip install deepchecks==0.18.1 altair_ally>=0.1.1

EDA_script

@@ -0,0 +1,28 @@
+import pandas as pd
+import numpy as np
+import 
+import click
+
+@click.command()
+@click.option('--input', type=str)
+
+
+def main(input): 
+
+combined_df = pd.read_csv(input)
+
+train_df, test_df = train_test_split(combined_df, test_size=0.3, random_state=123)
+
+numerical_chart = aly.dist(train_df, color='label')
+numerical_chart.save('results/numerical_chart.png', format='png')
+
+categorical_chart = aly.dist(train_df, dtype = 'category', color = 'label')
+categorical_chart.save('results/categorical_chart.png', format='png')
+
+train_df.to_csv('data/train_df.csv', index=False)
+test_df.to_csv('data/test_df.csv', index=False)
+
+
+# call main function 
+if __name__ == "__main__":
+    main() # pass any command line args to main here

environment.yaml → environment.yml

@@ -13,5 +13,3 @@ dependencies:
   - scipy=1.14.1
   - scikit-learn=1.2.2
   - pandera=0.20.4
-  - pip:
-    - altair_ally>=0.1.1