phc/conferences-llm-crawler
6
2
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

fix: added another example without link

Browse Source
This commit is contained in:
aziis98
2024-02-01 19:23:43 +01:00
parent 0afc1860df
commit 50b6a30647
1 changed files with 33 additions and 13 deletions
Download Patch File Download Diff File Expand all files Collapse all files
main.py
+33 -13
View File
@@ -15,7 +15,7 @@ HTML_EXAMPLE_1 = r"""<p><a href="http://www.crm.sns.it/event/507/" target="_blan
OUTPUT_EXAMPLE_1 = json.dumps({
"title": "Statistical and Computational Aspects of Dynamics",
"url": "http://www.crm.sns.it/event/507/",
"description": "Organized by Buddhima Kasun Fernando Akurugodage (Centro di ricerca matematica Ennio De Giorgi – SNS), Paolo Giulietti, and Tanja Isabelle Schindler (Universität Wien, Austria). Location: Centro De Giorgi - SNS, Pisa.",
"description": "Organized by Buddhima Kasun Fernando Akurugodage (Centro di ricerca matematica Ennio De Giorgi – SNS), Paolo Giulietti, and Tanja Isabelle Schindler (Universität Wien, Austria).\n\nLocation: Centro De Giorgi – SNS, Pisa",
"startDate": "2022-12-13",
"endDate": "2022-12-16"
})
@@ -27,12 +27,24 @@ HTML_EXAMPLE_2 = r"""<p><em><a href="http://pagine.dm.unipi.it/berselli/meeting2
OUTPUT_EXAMPLE_2 = json.dumps({
"title": "Workshop on Variational problems, PDEs and applications",
"url": "http://pagine.dm.unipi.it/berselli/meeting2020/",
"description": "Organized by Luigi Berselli, Giuseppe Buttazzo, Matteo Novaga, and Andrea Malchiodi (Scuola Normale Superiore, Pisa). Location: Department of Mathematics, Pisa.",
"description": "Organized by Luigi Berselli, Giuseppe Buttazzo, Matteo Novaga, and Andrea Malchiodi (Scuola Normale Superiore, Pisa).\n\nLocation: Department of Mathematics, Pisa",
"startDate": "2020-01-17",
"endDate": "2020-01-18",
})
HTML_EXAMPLE_3 = r"""<p>Geometric Representation Theory. ICM Satellite Conference<br/>Organized by Tomoyuki Arakawa (RIMS, Kyoto, Japan), Joel Kamnitzer (University of Toronto, Japan), Hiraku Nakajima (Kavli IPMU, Japan), Markus Reineke (Ruhr-Universität Bochum), Francesco Sala, and Vera Serganova (University of California Berkeley, USA). Online. June 27 – July 2, 2022.</p>"""
OUTPUT_EXAMPLE_3 = json.dumps({
"title": "Geometric Representation Theory. ICM Satellite Conference",
"url": None,
"description": "Organized by Tomoyuki Arakawa (RIMS, Kyoto, Japan), Joel Kamnitzer (University of Toronto, Japan), Hiraku Nakajima (Kavli IPMU, Japan), Markus Reineke (Ruhr-Universität Bochum), Francesco Sala, and Vera Serganova (University of California Berkeley, USA).\n\nLocation: Online",
"startDate": "2022-06-27",
"endDate": "2022-07-02",
})
def translate_to_json(conference_html: str) -> str:
llm_answer = llm.create_chat_completion(
max_tokens=None,
@@ -41,17 +53,25 @@ def translate_to_json(conference_html: str) -> str:
"role": "system",
"content": "You are an assistant helping a developer converting raw text data to JSON. Be precise in formatting the output and only output valid JSON using the specificied fields, without including any additional fields or comments",
},
{"role": "user", "content": "INPUT:"},
{"role": "user", "content": HTML_EXAMPLE_1 },
{"role": "user", "content": "OUTPUT JSON:"},
{"role": "user", "content": OUTPUT_EXAMPLE_1},
{"role": "user", "content": "INPUT:"},
{"role": "user", "content": HTML_EXAMPLE_2 },
{"role": "user", "content": "OUTPUT JSON:"},
{"role": "user", "content": OUTPUT_EXAMPLE_2},
{"role": "user", "content": "INPUT:"},
{"role": "user", "content": conference_html },
{"role": "user", "content": "OUTPUT JSON:"},
# Example 1
{ "role": "user", "content": "INPUT:" },
{ "role": "user", "content": HTML_EXAMPLE_1 },
{ "role": "user", "content": "OUTPUT JSON:" },
{ "role": "user", "content": OUTPUT_EXAMPLE_1 },
# Example 2
{ "role": "user", "content": "INPUT:" },
{ "role": "user", "content": HTML_EXAMPLE_2 },
{ "role": "user", "content": "OUTPUT JSON:" },
{ "role": "user", "content": OUTPUT_EXAMPLE_2 },
# Example 3
{ "role": "user", "content": "INPUT:" },
{ "role": "user", "content": HTML_EXAMPLE_3 },
{ "role": "user", "content": "OUTPUT JSON:" },
{ "role": "user", "content": OUTPUT_EXAMPLE_3 },
# Actual item to process
{ "role": "user", "content": "INPUT:" },
{ "role": "user", "content": conference_html },
{ "role": "user", "content": "OUTPUT JSON:" },
],
)