Kuautli commited on
Commit
edb7d72
verified
1 Parent(s): 97e7318

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +19 -9
app.py CHANGED
@@ -7,6 +7,16 @@ from dotenv import load_dotenv
7
  from flask import Flask, render_template, request
8
  import logging
9
 
 
 
 
 
 
 
 
 
 
 
10
  if os.getenv("HUGGINGFACE_HUB_CACHE") is None:
11
  load_dotenv()
12
 
@@ -34,23 +44,23 @@ def index():
34
  sentiment_count = None
35
 
36
  current_directory = os.getcwd()
37
- print("Iniciando procesamiento...")
38
 
39
  if request.method == "POST":
40
  url = request.form["url"]
41
  if url:
42
- print("Obteniendo datos")
43
  video_details = clustering.get_youtube_video_details(url, api_key)
44
  comments_df = clustering.get_youtube_comments(api_key, url)
45
- print("Generando embeddings")
46
  comments_df = clustering.add_normalized_embeddings_to_dataframe(
47
  comments_df, "comment"
48
  )
49
- print("Procesamiento de los datos")
50
  comments_df["published_at"] = pd.to_datetime(
51
  comments_df["published_at"]
52
  ).dt.date
53
- print("Clasificaci贸n de los sentimientos")
54
  comments_df = clustering.classify_sentiment_df(comments_df)
55
  comments_df.to_pickle(
56
  "./data/Comentarios-Youtube/comments_df.pkl"
@@ -66,14 +76,14 @@ def index():
66
  umap_df, min_eps, max_eps = clustering.transform_embeddings(
67
  comments_df, embeddings_col="embeddings"
68
  )
69
- print("Generaci贸n de wordcloud")
70
  image_path = os.path.join("static", "wordcloud.png")
71
  clustering.plot_wordcloud(comments_df, text_column="comment", output_filename=image_path)
72
 
73
  total = comments_df.shape[0]
74
- print("Evaluaci贸n de m茅tricas")
75
  min_items_by_cluster = clustering.determine_min_items_by_cluster(total)
76
- print("Modelado")
77
  (
78
  cluster_assignments,
79
  cluster_counts,
@@ -86,7 +96,7 @@ def index():
86
  threshold_values=[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9],
87
  embeddings_col="embeddings"
88
  )
89
- print("Creaci贸n de gr谩fico de Sankey")
90
  labels, source, target, values, comments = clustering.build_sankey_data(
91
  cluster_assignments,
92
  cluster_counts,
 
7
  from flask import Flask, render_template, request
8
  import logging
9
 
10
+ logging.basicConfig(
11
+ level=logging.DEBUG,
12
+ format='%(asctime)s - %(levelname)s - %(message)s',
13
+ datefmt='%Y-%m-%d %H:%M:%S'
14
+ )
15
+
16
+ def log_message(message):
17
+ """"""
18
+ logging.info(message)
19
+
20
  if os.getenv("HUGGINGFACE_HUB_CACHE") is None:
21
  load_dotenv()
22
 
 
44
  sentiment_count = None
45
 
46
  current_directory = os.getcwd()
47
+ log_message("Iniciando procesamiento...")
48
 
49
  if request.method == "POST":
50
  url = request.form["url"]
51
  if url:
52
+ log_message("Obteniendo datos de Youtube")
53
  video_details = clustering.get_youtube_video_details(url, api_key)
54
  comments_df = clustering.get_youtube_comments(api_key, url)
55
+ log_message("Generando embeddings")
56
  comments_df = clustering.add_normalized_embeddings_to_dataframe(
57
  comments_df, "comment"
58
  )
59
+ log_message("Procesamiento de los datos")
60
  comments_df["published_at"] = pd.to_datetime(
61
  comments_df["published_at"]
62
  ).dt.date
63
+ log_message("Clasificaci贸n de los sentimientos")
64
  comments_df = clustering.classify_sentiment_df(comments_df)
65
  comments_df.to_pickle(
66
  "./data/Comentarios-Youtube/comments_df.pkl"
 
76
  umap_df, min_eps, max_eps = clustering.transform_embeddings(
77
  comments_df, embeddings_col="embeddings"
78
  )
79
+ log_message("Generaci贸n de wordcloud")
80
  image_path = os.path.join("static", "wordcloud.png")
81
  clustering.plot_wordcloud(comments_df, text_column="comment", output_filename=image_path)
82
 
83
  total = comments_df.shape[0]
84
+
85
  min_items_by_cluster = clustering.determine_min_items_by_cluster(total)
86
+ log_message("Modelado y generaci贸n de m茅tricas")
87
  (
88
  cluster_assignments,
89
  cluster_counts,
 
96
  threshold_values=[0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9],
97
  embeddings_col="embeddings"
98
  )
99
+ log_message("Creaci贸n de gr谩fico de Sankey")
100
  labels, source, target, values, comments = clustering.build_sankey_data(
101
  cluster_assignments,
102
  cluster_counts,