Sam Passaglia commited on
Commit
65d65b7
1 Parent(s): 94492dd

update column name

Browse files
Files changed (1) hide show
  1. app.py +16 -6
app.py CHANGED
@@ -68,13 +68,15 @@ def get_stats():
68
 
69
  # if reading != '<OTHER>'
70
 
71
- df = pd.DataFrame({"heteronym": heteronyms, "accuracy": accuracy, "readings": readings})
 
 
72
 
73
  df = df[df["readings"].str.contains("、")]
74
 
75
  df["readings"] = df["readings"].str.replace("<OTHER>", "Other")
76
 
77
- df = df.rename(columns={"readings": "readings (test corr./total)"})
78
 
79
  df = df.sort_values("accuracy", ascending=False, ignore_index=True)
80
 
@@ -113,7 +115,7 @@ def furigana_to_spacy(text_with_furigana):
113
  }
114
 
115
 
116
- st.title("Yomikata: Disambiguate Japanese Heteronyms with a BERT model")
117
 
118
  # Input text box
119
  st.markdown("Input a Japanese sentence:")
@@ -137,10 +139,14 @@ label_colors = {
137
  reading: colors[i % len(colors)]
138
  for i, reading in enumerate(set([item["label"] for item in spacy_dict["ents"]]))
139
  }
140
- html = spacy.displacy.render(spacy_dict, style="ent", manual=True, options={"colors": label_colors})
 
 
141
 
142
  if len(spacy_dict["ents"]) > 0:
143
- st.markdown("**Yomikata** found and disambiguated the following heteronyms:")
 
 
144
  st.write(
145
  f"{add_border(html)}",
146
  unsafe_allow_html=True,
@@ -184,11 +190,15 @@ if st.button("🎲 Randomize the input sentence"):
184
  global_accuracy, stats_df = get_stats()
185
 
186
  st.subheader(
187
- f"{len(stats_df)} heteronyms supported, with a global accuracy of {global_accuracy:.0%}"
188
  )
189
 
190
  st.dataframe(stats_df)
191
 
 
 
 
 
192
  # Hide the footer
193
  hide_streamlit_style = """
194
  <style>
68
 
69
  # if reading != '<OTHER>'
70
 
71
+ df = pd.DataFrame(
72
+ {"heteronym": heteronyms, "accuracy": accuracy, "readings": readings}
73
+ )
74
 
75
  df = df[df["readings"].str.contains("、")]
76
 
77
  df["readings"] = df["readings"].str.replace("<OTHER>", "Other")
78
 
79
+ df = df.rename(columns={"readings": "readings (correct/total)"})
80
 
81
  df = df.sort_values("accuracy", ascending=False, ignore_index=True)
82
 
115
  }
116
 
117
 
118
+ st.title("Yomikata: Disambiguate Japanese Heteronyms")
119
 
120
  # Input text box
121
  st.markdown("Input a Japanese sentence:")
139
  reading: colors[i % len(colors)]
140
  for i, reading in enumerate(set([item["label"] for item in spacy_dict["ents"]]))
141
  }
142
+ html = spacy.displacy.render(
143
+ spacy_dict, style="ent", manual=True, options={"colors": label_colors}
144
+ )
145
 
146
  if len(spacy_dict["ents"]) > 0:
147
+ st.markdown(
148
+ "**Yomikata** disambiguated the following words with multiple readings:"
149
+ )
150
  st.write(
151
  f"{add_border(html)}",
152
  unsafe_allow_html=True,
190
  global_accuracy, stats_df = get_stats()
191
 
192
  st.subheader(
193
+ f"**Yomikata** supports {len(stats_df)} heteronyms, with a global accuracy of {global_accuracy:.0%}!"
194
  )
195
 
196
  st.dataframe(stats_df)
197
 
198
+ st.subheader(
199
+ "Check out **Yomikata** on [GitHub](https://github.com/passaglia/yomikata) today!"
200
+ )
201
+
202
  # Hide the footer
203
  hide_streamlit_style = """
204
  <style>