I amusing this method of query (answers)
openai.Answer.create(
search_model="ada",
model="curie",
question="which puppy is happy?",
documents=["Puppy A is happy.", "Puppy B is sad."],
examples_context="In 2017, U.S. life expectancy was 78.6 years.",
examples=[["What is human life expectancy in the United States?","78 years."]],
max_tokens=5,
stop=["\n", "<|endoftext|>"],
)
But instead of documents, I am putting a fileID
but here is my problem which I do not think I am in the wrong.
Successful Query
openai.Answer.create(
search_model="ada",
model="ada",
question="Which Puppy is happy?",
file="file-XQmVUQ0Gxi6anBIdioCcSuaf",
examples_context="In 2017, U.S. life expectancy was 78.6 years.",
examples=[["What is human life expectancy in the United States?","78 years."]],
max_tokens=5,
stop=["\n", "<|endoftext|>"],
)
<OpenAIObject answer at 0x27de7e38af0> JSON: {
"answers": [
"Puppy is happy."
],
"completion": "cmpl-3UtyWZ4FpQI82P8WeglAeuV7xQ8aT",
"file": "file-XQmVUQ0Gxi6anBIdioCcSuaf",
"model": "ada:2020-05-03",
"object": "answer",
"search_model": "ada:2020-05-03",
"selected_documents": [
{
"document": 0,
"object": "search_result",
"score": -33.798,
"text": "Research Projects Dementia Detection via Video Detect diseases such as dementia using iPhone Camera which required detection using smaller models using CNNs and Object detectors. Trained on 200GB of image data to detect gazes which is used in detecting dementia. Implemented in Android and iPhone devices as an App. (Python, Tensorflow, Keras, OpenCV, Docker and XML)"
}
]
}
Unsuccessful Query
openai.Answer.create(
search_model="ada",
model="ada",
question="What is his phone number?",
file="file-XQmVUQ0Gxi6anBIdioCcSuaf",
examples_context="In 2017, U.S. life expectancy was 78.6 years.",
examples=[["What is human life expectancy in the United States?","78 years."]],
max_tokens=5,
stop=["\n", "<|endoftext|>"],
)
---------------------------------------------------------------------------
InvalidRequestError Traceback (most recent call last)
<ipython-input-46-2d757d9afef1> in <module>
7 examples=[["What is human life expectancy in the United States?","78 years."]],
8 max_tokens=5,
----> 9 stop=["\n", "<|endoftext|>"],
10 )
D:\Anaconda\envs\Tensorflow2\lib\site-packages\openai\api_resources\answer.py in create(cls, **params)
12 def create(cls, **params):
13 instance = cls()
---> 14 return instance.request("post", cls.get_url("answers"), params)
D:\Anaconda\envs\Tensorflow2\lib\site-packages\openai\openai_object.py in request(self, method, url, params, headers, stream, plain_old_data)
242 )
243 response, stream, api_key = requestor.request(
--> 244 method, url, params, headers, stream=stream
245 )
246
D:\Anaconda\envs\Tensorflow2\lib\site-packages\openai\api_requestor.py in request(self, method, url, params, headers, stream)
130 method.lower(), url, params, headers, stream=stream
131 )
--> 132 resp = self.interpret_response(rbody, rcode, rheaders, stream=stream)
133 return resp, stream, my_api_key
134
D:\Anaconda\envs\Tensorflow2\lib\site-packages\openai\api_requestor.py in interpret_response(self, rbody, rcode, rheaders, stream)
356 )
357 else:
--> 358 return self.interpret_response_line(rbody, rcode, rheaders, stream)
359
360 def interpret_response_line(self, rbody, rcode, rheaders, stream=False):
D:\Anaconda\envs\Tensorflow2\lib\site-packages\openai\api_requestor.py in interpret_response_line(self, rbody, rcode, rheaders, stream)
376 if stream_error or not 200 <= rcode < 300:
377 raise self.handle_error_response(
--> 378 rbody, rcode, resp.data, rheaders, stream_error=stream_error
379 )
380
InvalidRequestError: No similar documents were found in file with ID 'file-XQmVUQ0Gxi6anBIdioCcSuaf'.Please upload more documents or adjust your query.
Document not found… I just used the document earlier for the “which puppy is happy”.
The document itself is a proper JSONL file which I have double checked it.
The format in the JSONL file is {“text”:“text inside here”}
