mirror of https://github.com/sean1832/GPT-Brain
You cannot select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
72 lines
2.0 KiB
Python
72 lines
2.0 KiB
Python
import openai
|
|
import numpy as np
|
|
import requests
|
|
import sseclient
|
|
|
|
|
|
# this function compare similarity between two vectors.
|
|
# The higher value the dot product have, the more alike between these vectors
|
|
def similarity(v1, v2):
|
|
return np.dot(v1, v2)
|
|
|
|
|
|
# return a list of vectors
|
|
def embedding(content, engine='text-embedding-ada-002'):
|
|
response = openai.Embedding.create(input=content, engine=engine)
|
|
vector = response['data'][0]['embedding']
|
|
return vector
|
|
|
|
|
|
def search_chunks(text, data, count=1):
|
|
vector = embedding(text)
|
|
points = []
|
|
|
|
for item in data:
|
|
# compare search terms with brain-data
|
|
point = similarity(vector, item['vector'])
|
|
points.append({
|
|
'content': item['content'],
|
|
'point': point
|
|
})
|
|
# sort points base on descendant order
|
|
ordered = sorted(points, key=lambda d: d['point'], reverse=True)
|
|
|
|
return ordered[0:count]
|
|
|
|
|
|
def gpt3(prompt, model, params):
|
|
response = openai.Completion.create(
|
|
model=model,
|
|
prompt=prompt,
|
|
temperature=params.temp,
|
|
max_tokens=params.max_tokens,
|
|
top_p=params.top_p,
|
|
frequency_penalty=params.frequency_penalty,
|
|
presence_penalty=params.present_penalty
|
|
)
|
|
text = response['choices'][0]['text'].strip()
|
|
return text
|
|
|
|
|
|
def gpt3_stream(API_KEY, prompt, model, params):
|
|
url = 'https://api.openai.com/v1/completions'
|
|
headers = {
|
|
'Accept': 'text/event-stream',
|
|
'Authorization': 'Bearer ' + API_KEY
|
|
}
|
|
body = {
|
|
'model': model,
|
|
'prompt': prompt,
|
|
'max_tokens': params.max_tokens,
|
|
'temperature': params.temp,
|
|
'top_p': params.top_p,
|
|
'frequency_penalty': params.frequency_penalty,
|
|
'presence_penalty': params.present_penalty,
|
|
'stream': True,
|
|
}
|
|
|
|
req = requests.post(url, stream=True, headers=headers, json=body)
|
|
client = sseclient.SSEClient(req)
|
|
return client
|
|
# print(json.loads(event.data)['choices'][0]['text'], end='', flush=True)
|