davidberenstein1957 HF staff commited on
Commit
e525bd5
1 Parent(s): 5ff9afc

fix: update example

Browse files
Files changed (1) hide show
  1. app.py +31 -21
app.py CHANGED
@@ -50,7 +50,7 @@ def load_examples():
50
 
51
 
52
  # Create Gradio examples
53
- examples = load_examples()
54
 
55
 
56
  def process_fields(fields):
@@ -112,34 +112,44 @@ from gradio_client import Client
112
  import argilla as rg
113
 
114
  # Initialize Argilla client
115
- client = rg.Argilla(
 
116
  api_key=os.environ["ARGILLA_API_KEY"], api_url=os.environ["ARGILLA_API_URL"]
117
  )
118
 
119
  # Load the dataset
120
- dataset = client.datasets(name="my_dataset", workspace="my_workspace")
121
-
122
- # Prepare example data
123
- example_field = dataset.settings.fields["my_input_field"].serialize()
124
- example_question = dataset.settings.questions["my_question_to_predict"].serialize()
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
125
 
 
126
  payload = {
127
- "records": [next(dataset.records()).to_dict()],
128
- "fields": [example_field],
129
- "question": example_question,
 
 
130
  }
131
 
132
- # Use gradio client to process the data
133
- client = Client("davidberenstein1957/distilabel-argilla-labeller")
134
-
135
- result = client.predict(
136
- records=json.dumps(payload["records"]),
137
- example_records=json.dumps(payload["example_records"]),
138
- fields=json.dumps(payload["fields"]),
139
- question=json.dumps(payload["question"]),
140
- api_name="/predict"
141
- )
142
-
143
  ```
144
  """
145
 
 
50
 
51
 
52
  # Create Gradio examples
53
+ examples = load_examples()[:1]
54
 
55
 
56
  def process_fields(fields):
 
112
  import argilla as rg
113
 
114
  # Initialize Argilla client
115
+ gradio_client = Client("davidberenstein1957/distilabel-argilla-labeller")
116
+ argilla_client = rg.Argilla(
117
  api_key=os.environ["ARGILLA_API_KEY"], api_url=os.environ["ARGILLA_API_URL"]
118
  )
119
 
120
  # Load the dataset
121
+ dataset = argilla_client.datasets(name="my_dataset", workspace="my_workspace")
122
+
123
+ # Get the field and question
124
+ field = dataset.settings.fields["text"]
125
+ question = dataset.settings.questions["sentiment"]
126
+
127
+ # Get completed and pending records
128
+ completed_records_filter = rg.Filter(("status", "==", "completed"))
129
+ pending_records_filter = rg.Filter(("status", "==", "pending"))
130
+ example_records = list(
131
+ dataset.records(
132
+ query=rg.Query(filter=completed_records_filter),
133
+ limit=5,
134
+ )
135
+ )
136
+ some_pending_records = list(
137
+ dataset.records(
138
+ query=rg.Query(filter=pending_records_filter),
139
+ limit=5,
140
+ )
141
+ )
142
 
143
+ # Process the records
144
  payload = {
145
+ "records": [record.to_dict() for record in some_pending_records],
146
+ "fields": [field.serialize()],
147
+ "question": question.serialize(),
148
+ "example_records": [record.to_dict() for record in example_records],
149
+ "api_name": "/predict",
150
  }
151
 
152
+ response = gradio_client.predict(**payload)
 
 
 
 
 
 
 
 
 
 
153
  ```
154
  """
155