Update app.py

app.py

@@ -1,6 +1,6 @@
-import asyncio
+#import asyncio
 
-asyncio.set_event_loop_policy(asyncio.DefaultEventLoopPolicy())
+#asyncio.set_event_loop_policy(asyncio.DefaultEventLoopPolicy())
 
 from pathlib import Path
 import io
@@ -77,7 +77,7 @@ DARK_MODE_HEAD = """<script>
     const url = new URL(window.location.href);
     if (!url.searchParams.has("__theme")) {
         url.searchParams.set("__theme", "dark");
-        window.history.replaceState(null, "", url);
+        window.location.replace(url);
     }
 })();
 </script>"""
@@ -363,7 +363,6 @@ def build_code(analysis, x, y, group, subject, within, dv, selected_vars, covars
         if model_family == "OLS (ANOVA / ANCOVA)":
             return f'''import statsmodels.formula.api as smf
 from statsmodels.stats.anova import anova_lm
-
 d = df[{selected!r}].dropna().copy()
 model = smf.ols(formula={f"Q({outcome!r}) ~ {rhs}"!r}, data=d).fit()
 result = model.summary2().tables[1]
@@ -373,7 +372,6 @@ anova = anova_lm(model, typ=2)'''
             fit = "logit" if model_family.startswith("Logit") else "probit"
             return f'''import statsmodels.formula.api as smf
 from scipy.stats import chi2
-
 d = df[{selected!r}].dropna().copy()
 d["__outcome"] = pd.Categorical(d[{outcome!r}]).codes
 if d["__outcome"].nunique() != 2:
@@ -387,7 +385,6 @@ overall = pd.DataFrame({{"LR chi2": [model.llr], "df": [model.df_model], "p-valu
 import patsy
 from scipy.stats import chi2
 from statsmodels.miscmodels.ordinal_model import OrderedModel
-
 d = df[{selected!r}].dropna().copy()
 d["__outcome"] = pd.Categorical(d[{outcome!r}].astype(str), categories={levels!r}, ordered=True).codes
 if (d["__outcome"] < 0).any() or d["__outcome"].nunique() < 2:
@@ -412,7 +409,6 @@ overall = pd.DataFrame({{"LR chi2": [lr_chi2], "df": [X.shape[1]], "p-value": [c
         return f'''import statsmodels.api as sm
 import statsmodels.formula.api as smf
 from scipy.stats import chi2
-
 d = df[{selected!r}].dropna().copy()
 {response}family = {family}
 model = smf.glm(formula={f"{glm_outcome} ~ {rhs}"!r}, data=d, family=family).fit()