Spaces:
Runtime error
Runtime error
fix: revert to old compression method for parquet file & now show 10 first emails subjects
Browse files- app.py +4 -4
- email_200.parquet +2 -2
app.py
CHANGED
@@ -17,9 +17,8 @@ title = "Confidential forensics tool with ChatGPT"
|
|
17 |
examples = ["Who is Phillip Allen?", "What the project in Austin is about?", "Give me more details about the real estate project"]
|
18 |
|
19 |
index = GPTSimpleVectorIndex.load_from_disk('email.json')
|
20 |
-
|
21 |
-
dat_fr = pd.DataFrame({"Documents loaded": ["email.json", "email_200.parquet"]})
|
22 |
df = pd.read_parquet("email_200.parquet")
|
|
|
23 |
|
24 |
def search_emails(opt, message, n=3):
|
25 |
"Outputs the top n emails that match the most the pattern"
|
@@ -34,7 +33,7 @@ def search_emails(opt, message, n=3):
|
|
34 |
message_tmp = df.sort_values('similarities', ascending=False).head(n)
|
35 |
message_tmp = [(row.file, row.body, row.similarities) for index, row in message_tmp.iterrows()]
|
36 |
for msg in message_tmp:
|
37 |
-
message += f"{msg[0]}\nContent: {msg[1].strip()}\
|
38 |
except Exception as e:
|
39 |
message = "An error occured when handling your query, please try again."
|
40 |
print(e)
|
@@ -78,9 +77,10 @@ with gr.Blocks(title=title) as demo:
|
|
78 |
# """ + title + """
|
79 |
""")
|
80 |
dat = gr.Dataframe(
|
81 |
-
value=
|
82 |
max_cols=1,
|
83 |
max_rows=4,
|
|
|
84 |
overflow_row_behaviour="paginate",
|
85 |
)
|
86 |
btn_upload = gr.UploadButton("Upload a new document...", file_types=["text"])
|
|
|
17 |
examples = ["Who is Phillip Allen?", "What the project in Austin is about?", "Give me more details about the real estate project"]
|
18 |
|
19 |
index = GPTSimpleVectorIndex.load_from_disk('email.json')
|
|
|
|
|
20 |
df = pd.read_parquet("email_200.parquet")
|
21 |
+
df2 = df["subject"].replace([None], "<no title>").head(10).to_frame()
|
22 |
|
23 |
def search_emails(opt, message, n=3):
|
24 |
"Outputs the top n emails that match the most the pattern"
|
|
|
33 |
message_tmp = df.sort_values('similarities', ascending=False).head(n)
|
34 |
message_tmp = [(row.file, row.body, row.similarities) for index, row in message_tmp.iterrows()]
|
35 |
for msg in message_tmp:
|
36 |
+
message += f"Mail ID: {msg[0]}\nContent: {msg[1].strip()}\nSimilarity score: {msg[2]}\n\n"
|
37 |
except Exception as e:
|
38 |
message = "An error occured when handling your query, please try again."
|
39 |
print(e)
|
|
|
77 |
# """ + title + """
|
78 |
""")
|
79 |
dat = gr.Dataframe(
|
80 |
+
value=df2,
|
81 |
max_cols=1,
|
82 |
max_rows=4,
|
83 |
+
title="Documents loaded",
|
84 |
overflow_row_behaviour="paginate",
|
85 |
)
|
86 |
btn_upload = gr.UploadButton("Upload a new document...", file_types=["text"])
|
email_200.parquet
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:689f185923970ee7a8da61eb63ae2116d75af0a473dd2735c0848bc87b505135
|
3 |
+
size 8670196
|