Getting HTTP Error Code: 422 when using Inference API
I am trying to use bigscience/bloom through Inference API, but I keep receiving HTTP Error Code: 422 (unfortunately, no more informative messages).
If I use the very same request for gpt2 (just changing the model), it works, therefore I assume it is not a badly formed request.
The request is shown below:
{
"inputs" : [ "How high is Mt. Everest?" ],
"options" : { },
"parameters" : {
"max_new_tokens" : 15,
"return_full_text" : false
}
}
Am I doing something wrong? Thanks.
+1, I get this all the time
Same, Does anyone know a fix for this?
I use this code and it works for me (albeit for Bloomz-3b):
import json
import requests as r
ENDPOINT_URL=""# url of your endpoint
HF_TOKEN=""
payload samples
regular_payload = { "inputs": "translate English to German: The weather is nice today." }
parameter_payload = {
"inputs": "translate English to French: Hello my name is Philipp and I am a Technical Leader at Hugging Face",
"parameters" : {
"max_length": 40,
}
}
HTTP headers for authorization
headers= {
"Authorization": f"Bearer {HF_TOKEN}",
"Content-Type": "application/json"
}
send request
response = r.post(ENDPOINT_URL, headers=headers, json=parameter_payload)
generated_text = response.json()
print(generated_text)
For those who have the same problem, on my side it came from a change in the API, the output format had changed so the max_new_token had to be decreased. Most of the time, a 422 error comes from a problem in the parameters.