Plot adaptation for github vision results.ipynb reading

epfl-ada · Dec 20, 2024 · 0707de5 · 0707de5
1 parent 0b89a70
commit 0707de5
Show file tree

Hide file tree

Showing 9 changed files with 1,003 additions and 3,626 deletions.
diff --git a/img/evolution_norwegian_influenced_names.png b/img/evolution_norwegian_influenced_names.png
diff --git a/img/evolution_norwegian_names.png b/img/evolution_norwegian_names.png
diff --git a/img/norwegian_influenced_name_per_year.png b/img/norwegian_influenced_name_per_year.png
diff --git a/img/norwegian_names_trend_per_decade.png b/img/norwegian_names_trend_per_decade.png
diff --git a/img/show_influence_amplitude_by_movie_genre.png b/img/show_influence_amplitude_by_movie_genre.png
diff --git a/img/show_top10_genre_occurence.png b/img/show_top10_genre_occurence.png
diff --git a/results_MS3.ipynb → results.ipynb b/results_MS3.ipynb → results.ipynb
diff --git a/src/models/trend_by_genres.py b/src/models/trend_by_genres.py
@@ -1,5 +1,6 @@
 import pandas as pd
 import plotly.express as px
+import matplotlib.pyplot as plt
 
 def clean_valid_names(global_names, names_to_remove=["M", "DOCTOR"]):
     """
@@ -309,15 +310,14 @@ def plot_top_genres(genre_influence, metric):
     """
     Create a bar chart for the top N genres and save it as an HTML file.
     """
-    fig = px.bar(
-        genre_influence,
-        x='Genres',
-        y=metric,
-        title='Top 10 Most Influential Movie Genres on Names',
-        labels={'Mean Difference': 'Total Influence Score'},
-        template='plotly_white'
-    )
-    fig.show()
+    fig, ax = plt.subplots()
+    ax.bar(genre_influence['Genres'], genre_influence[metric], color='skyblue')
+    ax.set_title('Top 10 Most Influential Movie Genres on Names')
+    ax.set_xlabel('Genres')
+    ax.set_ylabel('Total Influence Score')
+    plt.xticks(rotation=45, ha='right')
+    plt.tight_layout()
+    plt.show()
 
 def get_top_names_by_genre(exploded):
     """

diff --git a/src/utils/time_analysis.py b/src/utils/time_analysis.py
@@ -1,5 +1,5 @@
 import pandas as pd
-import plotly.express as px
+import matplotlib.pyplot as plt
 
 def show_influence_by_eras():
     prophet = pd.read_csv("data/clean/influenced_names_prophet.csv")
@@ -11,12 +11,11 @@ def show_influence_by_eras():
         labels=["60s", "70s", "80s", "90s", "00s"]
     )
     eras = influenced_prophet.groupby("Era", observed=False).size().reset_index(name="Count")
-    fig = px.bar(eras, x="Era", y="Count", title="Number of influenced names per Era")
-    fig.update_layout(
-        xaxis_title='Era',
-        yaxis_title='Count',
-        template="plotly_white",
-        width=800
-    )
-    fig.show()
+    fig, ax = plt.subplots()
+    ax.bar(eras['Era'], eras['Count'])
+    ax.set_title('Number of influenced names per Era')
+    ax.set_xlabel('Era')
+    ax.set_ylabel('Count')
+    fig.tight_layout()
+    plt.show()