From a11f39e7fcd68c8ac49e3e8f4cd2eb468846c80e Mon Sep 17 00:00:00 2001
From: jvivian <jtvivian@gmail.com>
Date: Sun, 9 Jun 2024 19:17:31 -0700
Subject: [PATCH] HotFix for factor analysis

---
 .../streamlit/pages/0_Dynamic_Factor_Model.py      | 14 ++++++++++++--
 covid19_drdfm/streamlit/pages/1_Factor_Analysis.py |  8 +++++---
 2 files changed, 17 insertions(+), 5 deletions(-)

diff --git a/covid19_drdfm/streamlit/pages/0_Dynamic_Factor_Model.py b/covid19_drdfm/streamlit/pages/0_Dynamic_Factor_Model.py
index 4752982..9912832 100644
--- a/covid19_drdfm/streamlit/pages/0_Dynamic_Factor_Model.py
+++ b/covid19_drdfm/streamlit/pages/0_Dynamic_Factor_Model.py
@@ -152,7 +152,7 @@ def additional_params():
     if not out_dir:
         st.warning("Specify output directory (in sidebar) to continue")
         st.stop()
-    return global_multiplier, out_dir
+    return global_multiplier, Path(out_dir)
 
 
 def run_model(ad, out_dir, batch, global_multiplier) -> ModelRunner:
@@ -175,4 +175,14 @@ def run_model(ad, out_dir, batch, global_multiplier) -> ModelRunner:
 batch = None if ad.obs.empty else ad.obs.columns[0]
 dfm = run_model(ad, out_dir, batch, global_multiplier)
 st.balloons()
-st.stop()
+
+filt_paths = [subdir / "factors.csv" for subdir in out_dir.iterdir() if (subdir / "factors.csv").exists()]
+dfs = [pd.read_csv(x) for x in filt_paths]
+try:
+    filt_df = pd.concat([x for x in dfs if ~x.empty]).set_index("Time")
+    filt_df.to_csv(out_dir / "factors.csv")
+    st.dataframe(filt_df)
+    dfm.ad.write(out_dir / "data.h5ad")
+    st.balloons()
+except ValueError:
+    st.error(f"No runs succeeded!! Check failures.txt in {out_dir}")
diff --git a/covid19_drdfm/streamlit/pages/1_Factor_Analysis.py b/covid19_drdfm/streamlit/pages/1_Factor_Analysis.py
index 2101e55..f569161 100644
--- a/covid19_drdfm/streamlit/pages/1_Factor_Analysis.py
+++ b/covid19_drdfm/streamlit/pages/1_Factor_Analysis.py
@@ -32,7 +32,7 @@ def normalize(df):
 center_title("Factor Analysis")
 
 # Read in data
-raw = get_df()
+# raw = get_df()
 # TEST_DIR = Path('covid19_drdfm/data/example-output/')
 
 # Parameter for results
@@ -72,8 +72,10 @@ def normalize(df):
 
 
 # Normalize original data for state / valid variables
-new = normalize(raw[raw.State == state][cols])
-# st.dataframe(new)
+ad = ann.read_h5ad(res_dir / "data.h5ad")
+new = ad.to_df().reset_index()
+new["State"] = ad.obs["State"].to_list()
+
 
 # Normalize factors and add to new dataframe
 if st.sidebar.checkbox("Invert Factor"):