You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I'm getting the processing status "failed". Checking the OpenAI message revealed its missing the method, url, and body in the jsonl. In other words, the format isn't what OpenAI expects. Any idea what I'm doing wrong? Help would be appreciated. Thanks in advance.
The text was updated successfully, but these errors were encountered:
{"custom_id":"b43f1dca-dac0-4ece-b6cf-3aae7b433ac4","params":{"model":"gpt-3.5-turbo","messages":[...],"max_tokens":1000,"temperature":0.0,"tools":[{"type":"function","function":{"name":"ClassModel","description":"Correctly extracted ClassModel with all the required parameters with correct types","parameters":{"properties":{"relevant":{"title":"Relevant","type":"integer"},"confidence":{"title":"Confidence","type":"integer"},"relevant_section":{"anyOf":[{"type":"string"},{"type":"null"}],"title":"Relevant Section"},"justification":{"title":"Justification","type":"string"}},"required":["confidence","justification","relevant","relevant_section"],"type":"object"}}}],"tool_choice":{"type":"function","function":{"name":"ClassModel"}}}}
I am creating a batch for gpt-3.5-turbo:
chunks = pd.read_csv(input_path)
def create_batch(chunks):
for index, row in chunks.iterrows():
chunk_text = row["chunk_text"]
yield [
{
"role": "system",
"content": system_prompt,
},
{
"role": "user",
"content": prompt.format(chunk=chunk_text, examples=examples),
},
]
batch_job = BatchJob.create_from_messages(
messages_batch=messages,
model=model,
temperature=temp,
file_path=output_path,
response_model=ClassModel
)
but when i want to upload it using
$instructor batch create-from-file --file-path /my/path/example_batchname.jsonl
I'm getting the processing status "failed". Checking the OpenAI message revealed its missing the method, url, and body in the jsonl. In other words, the format isn't what OpenAI expects. Any idea what I'm doing wrong? Help would be appreciated. Thanks in advance.
The text was updated successfully, but these errors were encountered: