Standardize data paths (#34)

* Add new directory structure and adjust the notebook

* Adjust the paths in the snakemake workflow

.gitignore

@@ -1,8 +1,8 @@
 # Private files
 private/
-data/
 # Snakemake-generated files
 generated/
+.snakemake/
 
 .DS_Store
 src/.DS_Store

data/.gitignore

@@ -0,0 +1,12 @@
+*.csv
+*.tsv
+*.yml
+*.yaml
+*.pdf
+*.npy
+*.npz
+*.zip
+*.png
+*.pdf
+*.jpg
+*.gif

data/main/README.md

@@ -0,0 +1,19 @@
+# Main data set
+
+This is the main data set from which we load the data.
+
+Structure:
+
+data/main
+├── deconvolved.csv
+├── README.md
+└── var_dates.yaml
+
+Checksums:
+```
+$ sha256sum data/main/deconvolved.csv 
+b8dbd93634a6310e13f639eff5402b40d47ed484a905d450b4e0a518d18d0f04  data/main/deconvolved.csv
+
+$ sha256sum data/main/var_dates.yaml 
+a6464599af6859d9542862e081766d31baffd7bc36792a0c671501142f9eda66  data/main/var_dates.yaml
+```

examples/frequentist_notebook_jax.py

@@ -24,6 +24,7 @@
 import matplotlib.ticker as ticker
 import pandas as pd
 import yaml
+from pathlib import Path
 
 from covvfit import plot, preprocess
 from covvfit import quasimultinomial as qm
@@ -37,13 +38,9 @@
 # We start by loading the data:
 
 # +
-_dir_switch = False  # Change this to True or False, depending on the laptop you are on
-if _dir_switch:
-    DATA_PATH = "../../LolliPop/lollipop_covvfit/deconvolved.csv"
-    VAR_DATES_PATH = "../../LolliPop/lollipop_covvfit/var_dates.yaml"
-else:
-    DATA_PATH = "../new_data/deconvolved.csv"
-    VAR_DATES_PATH = "../new_data/var_dates.yaml"
+DATA_DIR = Path("../data/main/")
+DATA_PATH = DATA_DIR / "deconvolved.csv"
+VAR_DATES_PATH = DATA_DIR / "var_dates.yaml"
 
 
 data = pd.read_csv(DATA_PATH, sep="\t")
@@ -550,4 +547,3 @@ def get_upper(ys):
 # Plot individual overdispersions
 for i, ax in enumerate(axs[:, 2]):
     plot_predictions(ax, i, **obtain_predictions(mcmc_indivi))
-# -

pyproject.toml

@@ -20,6 +20,7 @@ numpy = "==1.24.3"
 seaborn = "^0.13.2"
 numpyro = "^0.14.0"
 subplots-from-axsize = "^0.1.9"
+pango-aliasor = "^0.3.0"
 
 
 [tool.poetry.group.dev]

workflows/compare_clinical/.gitignore

@@ -0,0 +1,2 @@
+results/
+.snakemake/

workflows/compare_clinical/README.md

@@ -0,0 +1,11 @@
+# Comparison with clinical data
+
+To run the workflow, ensure that you have the `data/main` directory set up with the data.
+Note that the workflow should be run from this directory:
+
+Usage:
+
+```bash
+$ snakemake -s compare_clinical.smk --configfile CONFIG_NAME.yaml
+```
+

workflows/compare_clinical/config_ba1.yaml

@@ -1,6 +1,6 @@
 api_url: "https://lapis.cov-spectrum.org/open/v2/sample/aggregated"
 country: "Switzerland"
-wastewater_data_path: "../../../LolliPop/lollipop_covvfit/deconvolved.csv"
+wastewater_data_path: "../../data/main/deconvolved.csv"
 
 run_name: "config_ba1"
 

workflows/compare_clinical/config_ba1ba2.yaml

@@ -1,6 +1,6 @@
 api_url: "https://lapis.cov-spectrum.org/open/v2/sample/aggregated"
 country: "Switzerland"
-wastewater_data_path: "../../../LolliPop/lollipop_covvfit/deconvolved.csv"
+wastewater_data_path: "../../data/main/deconvolved.csv"
 
 run_name: "config_ba1ba2"
 

workflows/compare_clinical/config_ba4ba5.yaml

@@ -1,6 +1,6 @@
 api_url: "https://lapis.cov-spectrum.org/open/v2/sample/aggregated"
 country: "Switzerland"
-wastewater_data_path: "../../../LolliPop/lollipop_covvfit/deconvolved.csv"
+wastewater_data_path: "../../data/main/deconvolved.csv"
 
 run_name: "config_ba4ba5"
 

workflows/compare_clinical/config_ba4ba5_2.yaml

@@ -1,6 +1,6 @@
 api_url: "https://lapis.cov-spectrum.org/open/v2/sample/aggregated"
 country: "Switzerland"
-wastewater_data_path: "../../../LolliPop/lollipop_covvfit/deconvolved.csv"
+wastewater_data_path: "../../data/main/deconvolved.csv"
 
 run_name: "config_ba4ba5_2"