File size: 7,615 Bytes
83aabd5 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 |
#!/usr/bin/env python
# coding: utf-8
# Copyright 2021, IBM Corporation.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""
Flask API app and routes.
"""
__author__ = "Vagner Santana, Melina Alberio, Cassia Sanctos and Tiago Machado"
__copyright__ = "IBM Corporation 2024"
__credits__ = ["Vagner Santana, Melina Alberio, Cassia Sanctos, Tiago Machado"]
__license__ = "Apache 2.0"
__version__ = "0.0.1"
from flask import Flask, request, jsonify
from flask_cors import CORS, cross_origin
from flask_restful import Resource, Api, reqparse
import control.recommendation_handler as recommendation_handler
from helpers import get_credentials, authenticate_api, save_model
import config as cfg
import requests
import logging
import uuid
import json
import os
app = Flask(__name__)
# configure logging
logging.basicConfig(
filename='app.log', # Log file name
level=logging.INFO, # Log level (INFO, DEBUG, WARNING, ERROR, CRITICAL)
format='%(asctime)s - %(levelname)s - %(message)s' # Log message format
)
# access the app's logger
logger = app.logger
# create user id
id = str(uuid.uuid4())
# swagger configs
app.register_blueprint(cfg.SWAGGER_BLUEPRINT, url_prefix = cfg.SWAGGER_URL)
FRONT_LOG_FILE = 'front_log.json'
@app.route("/")
def index():
user_ip = request.remote_addr
logger.info(f'USER {user_ip} - ID {id} - started the app')
return app.send_static_file('demo/index.html')
@app.route("/recommend", methods=['GET'])
@cross_origin()
def recommend():
user_ip = request.remote_addr
hf_token, hf_url = get_credentials.get_credentials()
api_url, headers = authenticate_api.authenticate_api(hf_token, hf_url)
prompt_json = recommendation_handler.populate_json()
args = request.args
prompt = args.get("prompt")
print(prompt)
recommendation_json = recommendation_handler.recommend_prompt(prompt, prompt_json,
api_url, headers)
logger.info(f'USER - {user_ip} - ID {id} - accessed recommend route')
logger.info(f'RECOMMEND ROUTE - request: {prompt} response: {recommendation_json}')
return recommendation_json
@app.route("/get_thresholds", methods=['GET'])
@cross_origin()
def get_thresholds():
hf_token, hf_url = get_credentials.get_credentials()
api_url, headers = authenticate_api.authenticate_api(hf_token, hf_url)
prompt_json = recommendation_handler.populate_json()
model_id = 'sentence-transformers/all-minilm-l6-v2'
args = request.args
#print("args list = ", args)
prompt = args.get("prompt")
thresholds_json = recommendation_handler.get_thresholds(prompt, prompt_json, api_url,
headers, model_id)
return thresholds_json
@app.route("/recommend_local", methods=['GET'])
@cross_origin()
def recommend_local():
model_id, model_path = save_model.save_model()
prompt_json = recommendation_handler.populate_json()
args = request.args
print("args list = ", args)
prompt = args.get("prompt")
local_recommendation_json = recommendation_handler.recommend_local(prompt, prompt_json,
model_id, model_path)
return local_recommendation_json
@app.route("/log", methods=['POST'])
@cross_origin()
def log():
f_path = 'static/demo/log/'
new_data = request.get_json()
try:
with open(f_path+FRONT_LOG_FILE, 'r') as f:
existing_data = json.load(f)
except FileNotFoundError:
existing_data = []
existing_data.update(new_data)
#log_data = request.json
with open(f_path+FRONT_LOG_FILE, 'w') as f:
json.dump(existing_data, f)
return jsonify({'message': 'Data added successfully', 'data': existing_data}), 201
@app.route("/demo_inference", methods=['GET'])
@cross_origin()
def demo_inference():
args = request.args
model_id = args.get('model_id', default="meta-llama/Llama-4-Scout-17B-16E-Instruct")
temperature = args.get('temperature', default=0.5)
max_new_tokens = args.get('max_new_tokens', default=1000)
return {
'content': """
To effectively support your demands for increased resources, you'll want to gather a combination of quantitative and qualitative evidence. Here's a list of items you might consider compiling:
1. **Project backlog and pipeline:** Show the number of projects currently in the pipeline and those waiting to be started. This can help demonstrate the demand for your team's services.
2. **Project completion rate:** Calculate the percentage of projects completed on time and within budget. This can help show the efficiency of your team and the potential for scaling up without significantly impacting project quality.
3. **Client satisfaction data:** Collect feedback from clients, such as Net Promoter Score (NPS), survey responses, or testimonials. This can help demonstrate the value your team provides and the potential for acquiring new clients through word-of-mouth referrals.
4. **User engagement metrics:** Gather data on user engagement from your landing pages and UX interfaces, such as click-through rates, conversion rates, and bounce rates. This can help show the effectiveness of your designs and the potential for improved results with a larger team.
5. **Average project timeline:** Calculate the average time it takes for a project to be completed from start to finish. This can help demonstrate the need for more resources to meet increasing demand and maintain a reasonable project turnaround time.
6. **Resource utilization:** Analyze the current workload distribution among team members to identify bottlenecks and areas where additional resources could improve efficiency.
""",
'model_id':model_id,
'temperature': temperature,
'max_new_tokens': max_new_tokens
}
hf_token, _ = get_credentials.get_credentials()
prompt = args.get('prompt')
API_URL = "https://router.huggingface.co/together/v1/chat/completions"
headers = {
"Authorization": f"Bearer {hf_token}",
}
response = requests.post(
API_URL,
headers=headers,
json={
"messages": [
{
"role": "user",
"content": [
{
"type": "text",
"text": prompt
},
]
}
],
"model": model_id,
'temperature': temperature,
'max_new_tokens': max_new_tokens,
}
)
try:
response = response.json()["choices"][0]["message"]
response.update({
'model_id': model_id,
'temperature': temperature,
'max_new_tokens': max_new_tokens,
})
return response
except:
return response.text, response.status_code
if __name__=='__main__':
debug_mode = os.getenv('FLASK_DEBUG', 'True').lower() in ['true', '1', 't']
app.run(host='0.0.0.0', port='8080', debug=debug_mode) |