Skip to content

zhangguichuan/BiLSTM-CRF

BranchesTags
 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
data
data
 
 
BiLSTMCRF.py
BiLSTMCRF.py
 
 
CRF.py
CRF.py
 
 
README.md
README.md
 
 
main.ipynb
main.ipynb
 
 

Repository files navigation

Loading data

import pandas as pd
import numpy as np
myDf=pd.read_csv("data/test.csv")

Preprocessing

myDf["text"]=myDf["text"].apply(lambda x:x+" <end>")
myDf["tag"]=myDf["tag"].apply(lambda x:x+" END")
myDf[:1]
text tag
0 将 军 百 战 死 <end> B I B I S END 
myDf.dropna(inplace=True)

Transforming data to one-hot embedding to generalize X

wordIndexDict={"<pad>":0}
wi=1
for row in myDf["text"].values.tolist():
    if type(row)==float:
        print(row)
        break
    for word in row.split(" "):
        if word not in wordIndexDict:
            wordIndexDict[word]=wi
            wi+=1
vocabSize=wi
maxLen=max(len(row) for row in myDf["text"].values.tolist())
sequenceLengths=[len(row) for row in myDf["text"].values.tolist()]
myDf["text"]=myDf["text"].apply(lambda x:[wordIndexDict[word] for word in x.split()])
import tensorflow as tf
X=tf.keras.preprocessing.sequence.pad_sequences(myDf["text"],
                                                value=wordIndexDict["<pad>"],
                                                padding='post',
                                                maxlen=maxLen)
X
array([[ 1,  2,  3,  4,  5,  6,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,
         0,  0,  0],
       [ 1,  2,  7,  8,  9,  4, 10,  6,  0,  0,  0,  0,  0,  0,  0,  0,
         0,  0,  0],
       [ 1,  1,  2,  8,  9,  4, 11,  6,  0,  0,  0,  0,  0,  0,  0,  0,
         0,  0,  0],
       [ 2,  4,  1,  2,  6,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,
         0,  0,  0],
       [ 1,  2,  4,  1,  2,  6,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,
         0,  0,  0],
       [ 1,  2,  4,  1,  2,  6,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,
         0,  0,  0]])

Generalizing Y

import tqdm
import re

myDf["tag"]=myDf["tag"].apply(lambda x:re.sub("\-\S+","",x))

tagIndexDict = {"PAD": 0}
ti = 1
for row in tqdm.tqdm(myDf["tag"].values.tolist()):
    for tag in row.split(" "):
        if tag not in tagIndexDict:
            tagIndexDict[tag] = ti
            ti += 1
tagSum = len(list(tagIndexDict.keys()))
myDf["tag"] = myDf["tag"].apply(lambda x:x.split()+["PAD" for i in range(maxLen-len(x.split()))])
myDf["tag"] = myDf["tag"].apply(lambda x:[tagIndexDict[tagItem] for tagItem in x])
# myDf["tag"] = myDf["tag"].apply(lambda x: [[0 if tagI != tagIndexDict[tagItem] else 1
#                                             for tagI in range(len(tagIndexDict))]
#                                             for tagItem in x])
y=np.array(myDf["tag"].values.tolist())
100%|██████████| 6/6 [00:00<?, ?it/s]

y.shape # it is OK whether y is one-hot embedding or not
(6, 19)

Generalizing Model

from BiLSTMCRF import MyBiLSTMCRF
myModel=MyBiLSTMCRF(vocabSize,maxLen, tagIndexDict,tagSum,sequenceLengths)
myModel.myBiLSTMCRF.summary()
Model: "sequential_9"
_________________________________________________________________
Layer (type)                 Output Shape              Param #
=================================================================
embedding_9 (Embedding)      (None, 19, 100)           1200
_________________________________________________________________
bidirectional_18 (Bidirectio (None, 19, 5)             4240
_________________________________________________________________
bidirectional_19 (Bidirectio (None, 19, 5)             440
_________________________________________________________________
crf_layer (CRF)              (None, 19)                65
=================================================================
Total params: 5,945
Trainable params: 5,945
Non-trainable params: 0
_________________________________________________________________

training model

history=myModel.fit(X,y,epochs=1500)
.9719
Epoch 1263/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.9693
Epoch 1264/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.9667
Epoch 1265/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.9641
Epoch 1266/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.9616
Epoch 1267/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.9590
Epoch 1268/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.9565
Epoch 1269/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.9539
Epoch 1270/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.9514
Epoch 1271/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.9489
Epoch 1272/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.9463
Epoch 1273/1500
6/6 [==============================] - 0s 8ms/sample - loss: 1.9438
Epoch 1274/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.9413
Epoch 1275/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.9388
Epoch 1276/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.9363
Epoch 1277/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.9338
Epoch 1278/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.9313
Epoch 1279/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.9288
Epoch 1280/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.9264
Epoch 1281/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.9239
Epoch 1282/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.9214
Epoch 1283/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.9190
Epoch 1284/1500
6/6 [==============================] - 0s 8ms/sample - loss: 1.9165
Epoch 1285/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.9141
Epoch 1286/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.9116
Epoch 1287/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.9092
Epoch 1288/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.9068
Epoch 1289/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.9043
Epoch 1290/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.9019
Epoch 1291/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.8995
Epoch 1292/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.8971
Epoch 1293/1500
6/6 [==============================] - 0s 8ms/sample - loss: 1.8947
Epoch 1294/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.8923
Epoch 1295/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.8899
Epoch 1296/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.8875
Epoch 1297/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.8852
Epoch 1298/1500
6/6 [==============================] - 0s 8ms/sample - loss: 1.8828
Epoch 1299/1500
6/6 [==============================] - 0s 9ms/sample - loss: 1.8804
Epoch 1300/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.8781
Epoch 1301/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.8757
Epoch 1302/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.8734
Epoch 1303/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.8710
Epoch 1304/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.8687
Epoch 1305/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.8663
Epoch 1306/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.8640
Epoch 1307/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.8617
Epoch 1308/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.8594
Epoch 1309/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.8571
Epoch 1310/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.8547
Epoch 1311/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.8525
Epoch 1312/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.8502
Epoch 1313/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.8479
Epoch 1314/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.8456
Epoch 1315/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.8433
Epoch 1316/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.8410
Epoch 1317/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.8388
Epoch 1318/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.8365
Epoch 1319/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.8342
Epoch 1320/1500
6/6 [==============================] - 0s 8ms/sample - loss: 1.8320
Epoch 1321/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.8297
Epoch 1322/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.8275
Epoch 1323/1500
6/6 [==============================] - 0s 8ms/sample - loss: 1.8252
Epoch 1324/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.8230
Epoch 1325/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.8208
Epoch 1326/1500
6/6 [==============================] - 0s 8ms/sample - loss: 1.8185
Epoch 1327/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.8163
Epoch 1328/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.8141
Epoch 1329/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.8119
Epoch 1330/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.8097
Epoch 1331/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.8075
Epoch 1332/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.8053
Epoch 1333/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.8031
Epoch 1334/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.8009
Epoch 1335/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.7987
Epoch 1336/1500
6/6 [==============================] - 0s 8ms/sample - loss: 1.7965
Epoch 1337/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.7944
Epoch 1338/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.7922
Epoch 1339/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.7900
Epoch 1340/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.7879
Epoch 1341/1500
6/6 [==============================] - 0s 8ms/sample - loss: 1.7857
Epoch 1342/1500
6/6 [==============================] - 0s 8ms/sample - loss: 1.7836
Epoch 1343/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.7814
Epoch 1344/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.7793
Epoch 1345/1500
6/6 [==============================] - 0s 9ms/sample - loss: 1.7771
Epoch 1346/1500
6/6 [==============================] - 0s 8ms/sample - loss: 1.7750
Epoch 1347/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.7729
Epoch 1348/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.7707
Epoch 1349/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.7686
Epoch 1350/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.7665
Epoch 1351/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.7644
Epoch 1352/1500
6/6 [==============================] - 0s 8ms/sample - loss: 1.7623
Epoch 1353/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.7602
Epoch 1354/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.7581
Epoch 1355/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.7560
Epoch 1356/1500
6/6 [==============================] - 0s 8ms/sample - loss: 1.7539
Epoch 1357/1500
6/6 [==============================] - 0s 8ms/sample - loss: 1.7518
Epoch 1358/1500
6/6 [==============================] - 0s 8ms/sample - loss: 1.7497
Epoch 1359/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.7476
Epoch 1360/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.7456
Epoch 1361/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.7435
Epoch 1362/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.7414
Epoch 1363/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.7394
Epoch 1364/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.7373
Epoch 1365/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.7353
Epoch 1366/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.7332
Epoch 1367/1500
6/6 [==============================] - 0s 8ms/sample - loss: 1.7312
Epoch 1368/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.7291
Epoch 1369/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.7271
Epoch 1370/1500
6/6 [==============================] - 0s 8ms/sample - loss: 1.7251
Epoch 1371/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.7230
Epoch 1372/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.7210
Epoch 1373/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.7190
Epoch 1374/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.7169
Epoch 1375/1500
6/6 [==============================] - 0s 9ms/sample - loss: 1.7149
Epoch 1376/1500
6/6 [==============================] - 0s 8ms/sample - loss: 1.7129
Epoch 1377/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.7109
Epoch 1378/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.7089
Epoch 1379/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.7069
Epoch 1380/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.7049
Epoch 1381/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.7029
Epoch 1382/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.7009
Epoch 1383/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.6989
Epoch 1384/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.6969
Epoch 1385/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.6950
Epoch 1386/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.6930
Epoch 1387/1500
6/6 [==============================] - 0s 8ms/sample - loss: 1.6910
Epoch 1388/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.6891
Epoch 1389/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.6871
Epoch 1390/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.6851
Epoch 1391/1500
6/6 [==============================] - 0s 8ms/sample - loss: 1.6832
Epoch 1392/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.6812
Epoch 1393/1500
6/6 [==============================] - 0s 11ms/sample - loss: 1.6793
Epoch 1394/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.6773
Epoch 1395/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.6754
Epoch 1396/1500
6/6 [==============================] - 0s 8ms/sample - loss: 1.6735
Epoch 1397/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.6715
Epoch 1398/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.6696
Epoch 1399/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.6677
Epoch 1400/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.6657
Epoch 1401/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.6638
Epoch 1402/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.6619
Epoch 1403/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.6600
Epoch 1404/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.6581
Epoch 1405/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.6562
Epoch 1406/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.6543
Epoch 1407/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.6524
Epoch 1408/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.6505
Epoch 1409/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.6486
Epoch 1410/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.6467
Epoch 1411/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.6448
Epoch 1412/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.6429
Epoch 1413/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.6410
Epoch 1414/1500
6/6 [==============================] - 0s 8ms/sample - loss: 1.6392
Epoch 1415/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.6373
Epoch 1416/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.6354
Epoch 1417/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.6336
Epoch 1418/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.6317
Epoch 1419/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.6298
Epoch 1420/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.6280
Epoch 1421/1500
6/6 [==============================] - 0s 11ms/sample - loss: 1.6261
Epoch 1422/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.6243
Epoch 1423/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.6224
Epoch 1424/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.6206
Epoch 1425/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.6187
Epoch 1426/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.6169
Epoch 1427/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.6150
Epoch 1428/1500
6/6 [==============================] - 0s 8ms/sample - loss: 1.6132
Epoch 1429/1500
6/6 [==============================] - 0s 8ms/sample - loss: 1.6114
Epoch 1430/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.6096
Epoch 1431/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.6077
Epoch 1432/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.6059
Epoch 1433/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.6041
Epoch 1434/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.6023
Epoch 1435/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.6005
Epoch 1436/1500
6/6 [==============================] - 0s 8ms/sample - loss: 1.5987
Epoch 1437/1500
6/6 [==============================] - 0s 8ms/sample - loss: 1.5969
Epoch 1438/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.5951
Epoch 1439/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.5933
Epoch 1440/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.5915
Epoch 1441/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.5897
Epoch 1442/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.5879
Epoch 1443/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.5861
Epoch 1444/1500
6/6 [==============================] - 0s 8ms/sample - loss: 1.5843
Epoch 1445/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.5825
Epoch 1446/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.5808
Epoch 1447/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.5790
Epoch 1448/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.5772
Epoch 1449/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.5755
Epoch 1450/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.5737
Epoch 1451/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.5719
Epoch 1452/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.5702
Epoch 1453/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.5684
Epoch 1454/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.5666
Epoch 1455/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.5649
Epoch 1456/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.5631
Epoch 1457/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.5614
Epoch 1458/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.5597
Epoch 1459/1500
6/6 [==============================] - 0s 8ms/sample - loss: 1.5579
Epoch 1460/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.5562
Epoch 1461/1500
6/6 [==============================] - 0s 8ms/sample - loss: 1.5544
Epoch 1462/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.5527
Epoch 1463/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.5510
Epoch 1464/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.5493
Epoch 1465/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.5475
Epoch 1466/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.5458
Epoch 1467/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.5441
Epoch 1468/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.5424
Epoch 1469/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.5407
Epoch 1470/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.5389
Epoch 1471/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.5372
Epoch 1472/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.5355
Epoch 1473/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.5338
Epoch 1474/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.5321
Epoch 1475/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.5304
Epoch 1476/1500
6/6 [==============================] - 0s 8ms/sample - loss: 1.5288
Epoch 1477/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.5271
Epoch 1478/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.5254
Epoch 1479/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.5237
Epoch 1480/1500
6/6 [==============================] - 0s 8ms/sample - loss: 1.5220
Epoch 1481/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.5203
Epoch 1482/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.5187
Epoch 1483/1500
6/6 [==============================] - 0s 8ms/sample - loss: 1.5170
Epoch 1484/1500
6/6 [==============================] - 0s 8ms/sample - loss: 1.5153
Epoch 1485/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.5136
Epoch 1486/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.5120
Epoch 1487/1500
6/6 [==============================] - 0s 8ms/sample - loss: 1.5103
Epoch 1488/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.5087
Epoch 1489/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.5070
Epoch 1490/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.5053
Epoch 1491/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.5037
Epoch 1492/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.5020
Epoch 1493/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.5004
Epoch 1494/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.4987
Epoch 1495/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.4971
Epoch 1496/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.4955
Epoch 1497/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.4938
Epoch 1498/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.4922
Epoch 1499/1500
6/6 [==============================] - 0s 6ms/sample - loss: 1.4906
Epoch 1500/1500
6/6 [==============================] - 0s 7ms/sample - loss: 1.4889

predicting

testI=2
preY=myModel.predict(X)[testI]
indexTagDict=dict(list(zip(list(tagIndexDict.values()),list(tagIndexDict.keys()))))
indexWordDict=dict(list(zip(list(wordIndexDict.values()),list(wordIndexDict.keys()))))

sentenceList=[indexWordDict[wordItem] for wordItem in X[testI]]
sentenceList=sentenceList[:sentenceList.index("<end>")]

tagList=[indexTagDict[tagItem] for tagItem in preY]
tagList=tagList[:tagList.index("END")]

print(" ".join(sentenceList))
print(" ".join(tagList))
将 将 军 带 上 战 车
S B I B I B I

About

developed with tensorflow 2.1.0

Resources

Readme
Activity

Stars

0 stars

Watchers

0 watching

Forks

0 forks
Report repository

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 57.1%
  • Python 42.9%