Кластеризация данных из таблицы Excel

import pandas as pd
from sklearn.preprocessing import scale
from sklearn.cluster import DBSCAN, AffinityPropagation, Birch
from sklearn.manifold import TSNE
import plotly.express as px
from sklearn.cluster import KMeans
 
df = pd.read_excel("importFORlr4.xlsx", index_col=0)
 
df_scaled = pd.DataFrame(scale(df), index=df.index, columns=df.columns)
 
tsne = TSNE(n_components=2)
 
res = pd.DataFrame(tsne.fit_transform(df_scaled),
                   columns=list("xy"), index=df_scaled.index)
res['text'] = df.index.astype(str) + " (" + df['DATE'].astype(str) + ")"
color = df.index.astype('category').codes / df.index.nunique()
fig = px.scatter(res, x='x', y='y', hover_data=['text'], color=color)
fig.write_html('res.html')
 
df_agg = df_scaled.groupby(level=0).mean()
 
n_clusters = 6  # ожидаемое число кластеров
 
def get_cluster_res(model, df, names):
    res = pd.DataFrame({'name': names}, index=df.index)
    res['cluster'] = model.fit_predict(df)
    return res.groupby('cluster')['name'].apply(list).to_dict()
 
get_cluster_res(KMeans(n_clusters=n_clusters), df_agg, names=df_agg.index)
get_cluster_res(Birch(n_clusters=n_clusters), df_agg, names=df_agg.index)