def Predict():
record = 1
if askyesno('Record Mode', 'Yes to Record else No to Import Wavfile'):
message = predict.pred(0)
else:
Path = askopenfilename()
message = predict.pred(1, Path=Path)
showinfo('Result', message)
def test_one_as_train(ckpt_path):
predict_param = {
'model_name': 'Seq2seq_12',
'ckpt_path': ckpt_path,
'cangtou': '',
'keywords': '',
'test_set': '',
'eval_set': 'resource/dataset/test_10k_1k.txt',
'use_planning': False,
'bleu_eval': False,
'poem_type': 'poem7',
'train_mode': 'kw2poem',
'note': '',
'as_train': True,
# 'pred_soft': pred_soft,
# 'template': template,
'hard_rhyme': True,
'hard_tone': False,
# 'w1': w1,
# 'w2': w2,
}
save_file = pred(predict_param)
lv_rate, yun_rate = get_rate(save_file, 'result')
lm = get_lm(save_file)
return save_file, lv_rate, yun_rate, lm
def translate(self): word = self.LineEdit1.toPlainText() if not word: return results = pred(source_str = word,infer_model=infer_model,infer_sess=infer_sess) print(results) self.LineEdit2.setText(results)
def upload_file():
if request.method == 'POST':
# check if the post request has the file part
if 'file' not in request.files:
#flash('No file part')
return redirect(request.url)
file = request.files['file']
# if user does not select file, browser also
# submit a empty part without filename
if file.filename == '':
return redirect(request.url)
if file and allowed_file(file.filename):
filename = secure_filename(file.filename)
file.save(os.path.join(app.config['UPLOAD_FOLDER'], filename))
result = predict.pred(
os.path.join(app.config['UPLOAD_FOLDER'], filename))
return jsonify(result)
#return json.dumps({'status':'success'})
return
def uploads():
if request.method == "POST":
#file = request.files['file']
uploaded_files = request.files.getlist("file[]")
weight_path2 = os.path.join("static", "Classify_model.h5")
filenames = {
"bird": [],
"book": [],
"butterfly": [],
"cattle": [],
"chicken": [],
"elephant": [],
"horse": [],
"phone": [],
"sheep": [],
"shoes": [],
"spider": [],
"squirrel": [],
"watch": []
}
for file in uploaded_files:
if file and is_allowed_file(file.filename):
basepath = os.path.dirname(__file__)
filename = secure_filename(file.filename)
file.save(os.path.join(app.config['UPLOAD_FOLDER'], filename))
img_path = os.path.join(app.config['UPLOAD_FOLDER'], filename)
class_index, class_name = pred(img_path, weight_path2)
filenames[class_name].append(filename)
# print(class_index,class_name)
result = show_json(filename, class_name)
#data.append(result)
#res_json = json.dumps({"status": "200", "msg": "success","data":data})
return render_template(
'upload_more_ok.html',
filenames=filenames,
)
return render_template("upload_more.html", path="./images/test.jpg")
def cinc():
# print("scheduler")
global tcount,tyawn,tear,tearate,ta,tb
tdata[tcount]=[tyawn,tear,average(ta),average(tb)]
#print(pred(tdata[tcount]))
x=pred(tdata[tcount])
if(x[0]=="Drowsy"):
alarm.playalarm()
#print(len(ta),len(tb))
# with open('data2.csv', 'a', newline='') as file:
# writer = csv.writer(file)
# # writer.writerow(["Yawn", "Eye Blink", "Left Eye","Right Eye"])
# writer.writerow(tdata[tcount])
tyawn=0
tear=0
ta=[]
tb=[]
tearate=0
tcount+=1
# print(tdata)
return
def __call__(self, img):
ori_im = img.copy()
dt_boxes, elapse = self.text_detector(img)
print("dt_boxes num : {}, elapse : {}".format(len(dt_boxes), elapse))
if dt_boxes is None:
return None, None
img_crop_list = []
dt_boxes = sorted_boxes(dt_boxes)
for bno in range(len(dt_boxes)):
tmp_box = copy.deepcopy(dt_boxes[bno])
img_crop = self.get_rotate_crop_image(ori_im, tmp_box)
img_crop_list.append(img_crop)
# rec_res, elapse = self.text_recognizer(img_crop_list)
rec_res = []
elapse = 0
for img in img_crop_list:
txt, score, time_u = pred(img)
rec_res.append([txt, score])
elapse += time_u
print("rec_res num : {}, elapse : {}".format(len(rec_res), elapse))
# self.print_draw_crop_rec_res(img_crop_list, rec_res)
return dt_boxes, rec_res
def predictStuff(qns, file):
return pred(file, qns)
def get(self):
s = parser.parse_args()['query']
ans = pred(s)
print(ans, s)
return 'Prediction: ' + str(ans)
def results():
s = request.args['data']
ans = pred(s)
return render_template("result.html", data=ans)
def find(self, instance):
img, result = pred(self.path.text)
predictor.outmenu.setData(img1 = self.path.text, result1 = f"{result}")
predictor.screenmanager.current = "OutMenu"
from emg import (
Preprocess, BuildCNN , plot_history, train,Preprocess_all
)
from predict import (
pred
)
import numpy as np
def bond(x_train,x_train1,y_train,y_train1):
x_train = np.concatenate([x_train,x_train1], axis=0)
y_train = np.concatenate([y_train,y_train1], axis=0)
for l in [x_train, y_train]:
np.random.seed(1)
np.random.shuffle(l)
return x_train,y_train
if __name__ == '__main__':
#x_train, x_val, y_train, y_val =Preprocess_all()
x_train, x_val, y_train, y_val =Preprocess(0,0)
x_train1, x_val1, y_train1, y_val1=Preprocess(1,0.5)
x_train , y_train = bond(x_train,x_train1,y_train,y_train1)
model = BuildCNN()
hist = train(model,x_train, x_val1, y_train, y_val1)
#hist = train(model,x_train, x_val, y_train, y_val)
plot_history(hist)
pred(x_val1,y_val1)
def sentiment():
vars = request.args.items()
googpred=pred(getGoogleData('ya29.eQCnabTuxqmxPBwAAAB2AZe6wAdMBSKEvMO7GjvKRuSQsfpiJz2KrnWFA_KnCg','my-user-agent/1.0'))
print googpred
fbpred=pred(getFBMSG('a10cf4775b52e5412537f495696b3a15'))
print fbpred
def chance_fraud():
return predict.pred()
def predict_result(self):
return pred(
([self.age, self.sex, self.cigs, self.chol, self.bp, self.glucose
]), 'true_seasup.pkl')
print(X_train.shape, y_train.shape)
print(X_val.shape, y_val.shape)
classnum = [[548, 540, 392, 542, 256, 532, 375, 514, 231],
[5472, 13750, 1331, 2573, 1122, 4572, 981, 3363, 776]]
# Train!
model = net.netconv()
logger = keras.callbacks.CSVLogger("logs/train.log")
cp = keras.callbacks.ModelCheckpoint("./cp/weights.{epoch:02d}.hdf5")
X_full = np.concatenate((X_train, X_val))
y_full = np.concatenate((y_train, y_val))
print(X_full.shape, y_full.shape)
model.fit(X_full,
y_full,
batch_size=32,
epochs=25,
validation_split=.5,
callbacks=[logger, cp])
print("Predicting")
p = predict.pred(X, model)
i_p = predict.toImg(p, y.shape)
i_gt = predict.toImg(y, y.shape, palette=i_p.getpalette()[1:])
print(list(i_p.getdata())[100:130])
print(list(i_gt.getdata())[100:130])
print(i_p.getpalette())
print(i_gt.getpalette())
i_p.save("pred.png")
i_gt.save("ground.png")
predict.compare(i_p, i_gt).save("comp.png")
