def test_predict(self):
tests = [("<?php echo 1+1; ?>", 1), ('<?php system("ls"); ?>', 0)]
p = predict.Predictor("./model/model.joblib")
tmp_file = "/tmp/tmp.php"
for t in tests:
with open(tmp_file, "w") as f:
f.write(t[0])
result = p.predict(tmp_file)
print(result)
self.assertEqual(result, t[1])
def initPredictor():
global predictor
max_gram = 5
tokenizer = predict.Tokenizer()
tokens = []
tokens = tokens + tokenizer.tokenize_file("data/bee_movie.txt")
tokens = tokens + tokenizer.tokenize_file("data/moby_dick.txt")
tokens = tokens + tokenizer.tokenize_file("data/the_iliad.txt")
predictor = predict.Predictor(tokens, tokenizer, max_gram)
def add_predictor(self, predictor_name):
if predictor_name == setting.PREDICTOR_DEFAULT:
self.predictor = predict.Predictor()
elif predictor_name == setting.PREDICTOR_SIMPLE:
self.predictor = predict.SimplePredictor()
elif predictor_name == setting.PREDICTOR_Q:
self.predictor = predict.QPredictor()
self.predictor.init(self.switch_num)
elif predictor_name == setting.PREDICTOR_DQN:
self.predictor = predict.DQNPredictor()
self.predictor.init(self.switch_num)
else:
raise NameError('Error. No such predictor. Exit')
return
def _actualrun(self):
self.pickle_and_call('DPPP {}'.format(self.ddecal2))
self.fix_h5('prephase.h5')
self.pickle_and_call('DPPP {}'.format(self.acal2))
# self.pickle_and_call('DPPP {}'.format(self.ddecal_diag))
# self.fix_h5('prephase2.h5', True)
# self.pickle_and_call('DPPP {}'.format(self.acal_diag))
self._init_losoto()
self.pickle_and_call(self.losoto_p)
# self.pickle_and_call(self.losoto_a)
self.pickle_and_call('DPPP {}'.format(self.ddecal_pu))
self.fix_folders()
predictor = pr.Predictor(self.ms, self.predict_path, self.fpath, self.pset_loc)
predictor.initialize()
predictor.execute()
os.mkdir('{}/losoto'.format(self.ms))
def main():
max_gram = 3
tokenizer = predict.Tokenizer()
tokens = []
tokens = tokens + tokenizer.tokenize_file("data/bee_movie.txt")
tokens = tokens + tokenizer.tokenize_file("data/moby_dick.txt")
predictor = predict.Predictor(tokens, tokenizer, max_gram)
while True:
user_input = input("~> ")
if user_input == ":q":
return
predicted = predictor.get_prediction(user_input)
print(json.dumps(predicted.most_common(10)))
def setModel(self, modelFilePath):
self.model = predict.Predictor(modelFilePath)
import predict
import determine_position
import numpy as np
def convert_frame(a, b, c, T):
x = np.array([[a, b, c, 1]]).T
y = (T @ x).squeeze()
return y[0], y[1], y[2]
# load the trained model
# TODO determine path
predictor = predict.Predictor(ckpt_path='20000.pth')
# let the robot touch one edge
# 1st point on edge
N = 3 # take 3 points along an edge and take average
a1, b1, c1 = [0] * N, [0] * N, [0] * N
a2, b2, c2 = [0] * N, [0] * N, [0] * N
for i in range(N):
depth_map0 = from_robot() # TODO (400,) ndarray, some assumed API
a1[i], b1[i], c1[i], a2[i], b2[i], c2[i] = predictor.predict(depth_map0)
a1[i], b1[i], c1[i], a2[i], b2[i], c2[i] = a1[i].item(), b1[i].item(), c1[i].item(), a2[i].item(), b2[i].item(), c2[i].item()
a1 = sum(a1) / N
b1 = sum(b1) / N
c1 = sum(c1) / N
a2 = sum(a2) / N
b2 = sum(b2) / N
c2 = sum(c2) / N
#!/usr/bin/env python
import time
import cv2
import predict
import numpy as np
import sys
import rospy
CLASSES = [
'forward', 'backward', 'left', 'right', 'stop', 'fleft', 'fright', 'bleft',
'bright'
]
predictor = predict.Predictor(sys.argv[1], sys.argv[2])
from sensor_msgs.msg import Image, CameraInfo
from cv_bridge import CvBridge, CvBridgeError
class Exp:
def __init__(self):
self.output = None
self.ori = None
self.blur = None
self.mask = None
self.data_methods = ['blur', 'mask', 'ori', 'output']
self.classes = [
'forward', 'backward', 'left', 'right', 'stop', 'fleft', 'fright',
'bleft', 'bright'
]
import socketio
import predict
sio = socketio.Client()
pr = predict.Predictor()
@sio.on("connect")
def on_connect():
print("I\'m Connected!")
@sio.on("message")
def on_message(data):
print("I received a message: " + str(data))
@sio.on("disconnect")
def on_disconnect():
print("I\'m Disonnected!")
if __name__ == "__main__":
sio.connect("http://joshuayuan.me:8080")
pr.start(sio)
# sio.emit("pi:server", {"msg": "actually new message!"})
import numpy as np
import pandas as pd
import preprocess as prep
import predict as tr
STK = pd.read_csv("GOOGL.csv")
if len(STK) == 0:
print("no csv data")
exit()
dataPrep = prep.DataPrep("quandl", STK)
features, labels = dataPrep.stk2ind()
RF = tr.Predictor(features, labels, "rf")
RF.test()
