lewtun HF staff commited on
Commit
a353f77
1 Parent(s): 3a53495

Explain submission format

Browse files
app.py CHANGED
@@ -49,14 +49,40 @@ def http_post(
49
  with st.form(key="form"):
50
  # Flush local repo
51
  shutil.rmtree(LOCAL_REPO, ignore_errors=True)
52
- uploaded_file = st.file_uploader("Upload a submission.json file", type=["json"])
53
 
54
  if uploaded_file is not None:
55
  data = str(uploaded_file.read(), "utf-8")
56
  json_data = json.loads(data)
57
 
58
- with st.expander("Example submission"):
59
- with open("submission.json", "r") as f:
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
60
  example_submission = json.load(f)
61
  st.json(example_submission)
62
 
@@ -66,7 +92,7 @@ with st.form(key="form"):
66
  help="You can generate an access token via your Hub settings. See the [docs](https://huggingface.co/docs/hub/security#user-access-tokens) for more details",
67
  )
68
 
69
- submit_button = st.form_submit_button("Submit")
70
 
71
  if submit_button:
72
  validate_submission(json_data)
 
49
  with st.form(key="form"):
50
  # Flush local repo
51
  shutil.rmtree(LOCAL_REPO, ignore_errors=True)
52
+ uploaded_file = st.file_uploader("Upload submission.json file", type=["json"])
53
 
54
  if uploaded_file is not None:
55
  data = str(uploaded_file.read(), "utf-8")
56
  json_data = json.loads(data)
57
 
58
+ with st.expander("Submission format"):
59
+ st.markdown(
60
+ """
61
+ Please follow this JSON format for your `submission.json` file:
62
+
63
+ ```json
64
+ {
65
+ "submission_name": "An identifying name of your system",
66
+ "param_count": 123, # The number of parameters your system has.
67
+ "description": "An optional brief description of the system that will be shown on the results page",
68
+ "tasks":
69
+ {
70
+ "dataset_identifier": {
71
+ "values": ["output1", "output2", "..."], # A list of system outputs.
72
+ # Optionally, you can add the keys which are part of an example to ensure that there is no shuffling mistakes.
73
+ "keys": ["key-0", "key-1", ...]
74
+ }
75
+ }
76
+ }
77
+ ```
78
+ In this case, `dataset_identifier` is the identifier of the dataset
79
+ followed by an identifier of the set the outputs were created from, for
80
+ example `_validation` or `_test`. For example, the `mlsum_de` test set
81
+ would have the identifier `mlsum_de_test`. the Please see the sample
82
+ submission below:
83
+ """
84
+ )
85
+ with open("sample-submission.json", "r") as f:
86
  example_submission = json.load(f)
87
  st.json(example_submission)
88
 
 
92
  help="You can generate an access token via your Hub settings. See the [docs](https://huggingface.co/docs/hub/security#user-access-tokens) for more details",
93
  )
94
 
95
+ submit_button = st.form_submit_button("Make Submission")
96
 
97
  if submit_button:
98
  validate_submission(json_data)
submission.json → sample-submission.json RENAMED
@@ -1,7 +1,7 @@
1
  {
2
  "submission_name": "T5-base",
3
  "param_count": 222903552,
4
- "description": "",
5
  "tasks": {
6
  "mlsum_de_test": {
7
  "values": [
@@ -34,38 +34,6 @@
34
  "mlsum_de-challenge_test_covid-0",
35
  "mlsum_de-challenge_test_covid-1"
36
  ]
37
- },
38
- "mlsum_es_test": {
39
- "values": [
40
- "El PSOE y el PNV aprovechan la moción de censura que convirtió a Sánchez en presidente del Gobierno",
41
- "El desfile de las Fuerzas Armadas estará presidido por los Reyes de Espa ⁇ a"
42
- ],
43
- "keys": [
44
- "mlsum_es-test-0",
45
- "mlsum_es-test-1"
46
- ]
47
- },
48
- "mlsum_es_challenge_train_sample": {
49
- "values": [
50
- "La red gaditana de Gibraltar emplea 60.000 euros por alijo para cobrar un lanchero de las organizaciones del narcotráfico",
51
- "La Junta de Andaluc ⁇ a encarga el Centro de Estudios y Experimentación en Obras P ⁇ blicas para analizar la capacidad máxima del r ⁇ o"
52
- ]
53
- },
54
- "mlsum_es_challenge_validation_sample": {
55
- "values": [
56
- "Un nuevo estudio retrasa el horario de la primera clase de la jornada y el aumento de las calificaciones fue del 4,5%",
57
- "La hija de una toledana, Laura Sanz Nombela, acaecida en una panader ⁇ a, se encuentran en el hospital de la capital francesa"
58
- ]
59
- },
60
- "mlsum_es_challenge_test_covid": {
61
- "values": [
62
- "El 75% de los pacientes en lista de espera quir ⁇ rgica en grandes hospitales esperan intervenirse y pidiendo un aplazamiento",
63
- "La empresa Viralgen, que apuesta por el coronavirus, apuesta por el ant ⁇ geno para proteger frente al coronavirus"
64
- ],
65
- "keys": [
66
- "mlsum_es-challenge_test_covid-0",
67
- "mlsum_es-challenge_test_covid-1"
68
- ]
69
  }
70
  }
71
  }
 
1
  {
2
  "submission_name": "T5-base",
3
  "param_count": 222903552,
4
+ "description": "Sample submission with model predictions across multiple datasets",
5
  "tasks": {
6
  "mlsum_de_test": {
7
  "values": [
 
34
  "mlsum_de-challenge_test_covid-0",
35
  "mlsum_de-challenge_test_covid-1"
36
  ]
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
37
  }
38
  }
39
  }