Machine learning 如何部署没有预测属性的模型？_Machine Learning_Deployment

Machine learning 如何部署没有预测属性的模型？

machine-learning deployment

Machine learning 如何部署没有预测属性的模型？,machine-learning,deployment,Machine Learning,Deployment,如何使用flask部署它，我的输出不会显示在窗口中，尽管部署时会显示主页并获取输入。您的函数“model（user）”必须返回一些内容例如： app.py from flask import Flask, jsonify, request,render_template import pickle app = Flask(__name__,template_folder='template') # load model model = pickle.load(open("model.

如何使用flask部署它，我的输出不会显示在窗口中，尽管部署时会显示主页并获取输入。

您的函数“model（user）”必须返回一些内容例如：

app.py
from flask import Flask, jsonify, request,render_template
import pickle
app = Flask(__name__,template_folder='template')
# load model
model = pickle.load(open("model.pkl",'rb'))
# app

@app.route('/')
def home():
    return render_template('recommendation.html')
# routes
@app.route('/api', methods=['POST'])

def predict():
    result=request.form
    query_user_name=result["user name"]
    user_input = {'query':query_user_name}
    output_data=model(query_user_name)
    print(output_data)
    # send back to browser
    output ={output_data}
    return f'<html><body><h1>{output_data}</h1><form action="/"><button type="submit">back </button> </form></body></html>'

if __name__ == '__main__':
    app.run(debug=True)

def model(user):
    recommended_list=[]
    top_list=[]
    x = data.iloc[data.loc[data.Users == user].index[0],2:]
    similar = np.array([(data.iloc[i,0],weight_factor(x,data.iloc[i, 2:])) for i in range(0,data.shape[0],1)])
    index= np.argsort( similar[:,1] )
    index=index[::-1]
    similar=similar[index] 
    neighbours = similar[similar[:,1].astype(float) > 0.6]  #Taking threshold as 0.6
    for i in range(0,len(neighbours),1):
        for j in range(2,len(data.columns),1):
            if data.iloc[data.loc[data.Users == neighbours[i][0]].index[0],j]==1 and data.iloc[data.loc[data.Users == user].index[0],j]==0:
               recommended_list.append(data.columns[j])
    if (len(neighbours)>10):
       for i in range(0,10,1):  #Top 10 neighbours
           top_list.append(neighbours[i][0])
    else:
       for i in range(len(neighbours)):
            top_list.append(neighbours[i][0])
    if user in top_list: #Remove the user of which we are asked to find neighbours,each user is always strongly correlated with itself and its of no use to us.
       top_list.remove(user)
    print(" ")
    print("Top users similar to this user are:")
    print(" ")
    for i in range(0,len(top_list),1):
        print(top_list[i])
    print(" ")
    print("Users similar to this user liked these products too:")
    print(" ")
    recommended_array=np.unique(np.array(recommended_list))
    for i in range(0,len(recommended_array),1):
        print(recommended_array[i])

def model(user):
    # code 
    return somethings