我的代码有问题。我有一个名为test.txt的.txt文件，其中包含句子，而我有一个数据框，其中包含这些句子中的所有标记化单词。我的问题是我想专门查找并打印出特定令牌并保留其位置编号。我尝试了一些if statement（），但似乎覆盖了每个单词的索引计数。
public void createNotif(Poste poste, Comment comment, Boolean checkGroupSon) {
if (comment == null) {
emeteur = poste.getEmmet();
message = poste.getTx();
pathImg = poste.getImg();
} else {
emeteur = comment.getEmmet();
message = comment.getComment();
pathImg = comment.getImage();
}
//onDismiss Intent
Intent intent = new Intent(mContext, NotificationControllerReceiver.class);
PendingIntent mBroadcastIntentController = PendingIntent.getBroadcast(mContext, 0, intent, 0);
Intent notificationIntent = new Intent(mContext, CommentPhotoActivity.class);
notificationIntent.addFlags(Intent.FLAG_ACTIVITY_NEW_TASK); // Intent.FLAG_ACTIVITY_CLEAR_TASK|
notificationIntent.putExtra("pid", poste.getId());
PendingIntent contentIntent = PendingIntent.getActivity(mContext, 0, notificationIntent, PendingIntent.FLAG_UPDATE_CURRENT);
//TEST notification.setLatestEventInfo(getApplicationContext(), "YuYu", "Vous avez reçu un nouveau poste", contentIntent);
//TEST notification.flags = Notification.FLAG_AUTO_CANCEL;
//TEST
// NotificationCompat.Builder notificationBuilder = new NotificationCompat.Builder(getApplicationContext()).setContentTitle("Poste: "+emeteur).setSmallIcon(R.drawable.ic_launcher).setContentIntent(contentIntent).setContentText(message).setDeleteIntent(mBroadcastIntentController).setPriority(Notification.PRIORITY_MAX);
NotificationCompat.Builder notificationBuilder = null;
if (Build.VERSION.SDK_INT >= Build.VERSION_CODES.O) {
notificationBuilder = new NotificationCompat.Builder(mContext, "poste");
if (pathImg == null)
notificationBuilder.setContentTitle(emeteur).setSmallIcon(R.drawable.ic_launcher).setContentIntent(contentIntent).setContentText(emeteur + ": " + message).setDeleteIntent(mBroadcastIntentController).setPriority(NotificationCompat.PRIORITY_MAX);
else
notificationBuilder.setContentTitle(emeteur).setSmallIcon(R.drawable.ic_launcher).setContentIntent(contentIntent).setContentText(emeteur + ": Photo \uD83D\uDCF7 " + message).setDeleteIntent(mBroadcastIntentController).setPriority(NotificationCompat.PRIORITY_MAX);
}
// notificationBuilder.setContentTitle("Poste: "+emeteur).setSmallIcon(R.drawable.ic_launcher).setContentIntent(contentIntent).setContentText(message).setDeleteIntent(mBroadcastIntentController).setPriority(Notification.PRIORITY_MAX);
}
// long[] v = {500,1000};
// notificationBuilder.setVibrate(v);
notifications = null;
boolean b = notifHashMap.containsKey(poste.getId());
if (b) {
notifications = notifHashMap.get(poste.getId());
// Add your All messages here or use Loop to generate messages
if (pathImg == null)
notifications.add(emeteur + ":" + message);
else
notifications.add(emeteur + ": Photo \uD83D\uDCF7 " + message);
notifHashMap.put(poste.getId(), notifications);
} else {
notifications = new ArrayList<String>();
// Add your All messages here or use Loop to generate messages
if (pathImg == null)
notifications.add(emeteur + ":" + message);
else
notifications.add(emeteur + ": Photo \uD83D\uDCF7 " + message);
notifHashMap.put(poste.getId(), notifications);
}
// if (inboxStyle == null)
// inboxStyle = new NotificationCompat.InboxStyle();
//else
inboxStyle = new NotificationCompat.InboxStyle(notificationBuilder);
if (notifications.size() > 1) {
if (poste.getImg() == null)
inboxStyle.setBigContentTitle("Poste:" + poste.getTx());
else
inboxStyle.setBigContentTitle("Poste: \uD83D\uDCF7 " + poste.getTx());
inboxStyle.setSummaryText("Vous avez " + notifications.size() + " notifications.");
for (int i = 0; i < notifications.size(); i++) {
inboxStyle.addLine(notifications.get(i));
}
} else if (notifications.size() == 1) {
inboxStyle.addLine(notifications.get(0));
}
notificationBuilder.setStyle(inboxStyle);
notificationBuilder.setVibrate(new long[]{0L});
//notificationBuilder.setNumber(value++);
if (!checkGroupSon) {
Uri alarmSound = RingtoneManager.getDefaultUri(RingtoneManager.TYPE_NOTIFICATION);
notificationBuilder.setSound(alarmSound);
}
NotifManager = (NotificationManager) mContext.getSystemService(Context.NOTIFICATION_SERVICE);
if (Build.VERSION.SDK_INT >= Build.VERSION_CODES.O) {
/* Create or update. */
mChannel = new NotificationChannel("poste", "YuYu", NotificationManager.IMPORTANCE_LOW);
mChannel.enableLights(true);
mChannel.setLightColor(Color.RED);
mChannel.canShowBadge();
mChannel.setVibrationPattern(new long[]{ 0 });
mChannel.enableVibration(true);
/* if (!checkGroupSon) {
mChannel.enableVibration(true);
mChannel.setVibrationPattern(new long[]{100, 200, 300, 400, 500, 400, 300, 200, 400});
} */
assert NotifManager != null;
notificationBuilder.setChannelId("poste");
// NotifManager.deleteNotificationChannel("poste");
NotifManager.createNotificationChannel(mChannel);
}
if (Build.VERSION.SDK_INT >= Build.VERSION_CODES.N) {
summaryNotification = new NotificationCompat.Builder(mContext, "poste")
.setSmallIcon(R.drawable.ic_launcher)
.setStyle(new NotificationCompat.InboxStyle()
.addLine("1")
.addLine("2")
.setBigContentTitle("nouveau messages")
.setSummaryText("Nouveau message"))
.setPriority(NotificationCompat.PRIORITY_LOW)
.setGroup("example_group")
.setGroupAlertBehavior(NotificationCompat.GROUP_ALERT_CHILDREN)
.setGroupSummary(true).build();
}
Notification notification = notificationBuilder.setGroup("example_group").build();
notification.flags = Notification.FLAG_AUTO_CANCEL;
// assert NotifPostMsgManager != null;
//TEST manager.notify(ID_NOTIFICATION, notification);
NotifManager.notify(poste.getId(), notification);
if (Build.VERSION.SDK_INT >= Build.VERSION_CODES.N)
NotifManager.notify(1, summaryNotification);
}
我的test.txt文件中的示例句子是
def output ():
currCount = 0
for words in read():
add = len(words)
word_new = [' '.join(df.loc[t].values.tolist()) if t
in df.index is not None else t for t in word_tokenize(words)]
tag = ' '.join(word for word in word_new);
print('First:' + str(currCount) + '\n' + 'Last:' + str(currCount + add)
+ '\n' + 'Tag: ' + tag + '\n' + 'word: '+words + '\n')
currCount += add + 1
if words is ".":
currCount = 0
#Sample output #Output that i want
#First:0 #Assume that i only want
#Last:1 #PERSON tags
#Tag: PERSON
#word: I #First:0
#Last:1
#First:2 #Tag: PERSON
#Last:6 #word: I
#Tag: NOTHING
#word: like #First: 0
#Last: 3
#First:7 #Tag: Bob
#Last:12 #word: PERSON
#Tag: FOOD
#word: pizza
#First:13
#Last:14
#Tag: NOTHING
#word: .
#First:0
#Last:3
#Tag: Bob
#word: PERSON
#First:4
#Last:9
#Tag: NOTHING
#word: likes
#First:10
#Last:15
#Tag: FOOD
#word: pizza
#First:16
#Last:17
#Tag: NOTHING
#word: .
和我制作的标签样本
I like pizza .
Bob likes pizza .
I PERSON
Like NOTHING
Pizza FOOD
. NOTHING
Bob PERSON
likes NOTHING
pizza FOOD
. NOTHING
答案 0 :(得分：0)
这可能是一种更清洁，更轻松的方法：
words = []
for line in data:
start = 0
for word in nltk.word_tokenize(line):
word_tag = {}
word_tag['First'] = start
end = start + len(word)
word_tag['Last'] = end
word_tag['Word'] = word
# word_tag['Tag'] = <your statement for tagging>
words.append(word_tag)
start = end + 1
df = pd.DataFrame(words)
这是您的数据框的外观：
First Last Word Tag
0 1 I PERSON
2 6 like NOTHING
7 12 pizza FOOD
13 14 . NOTHING
0 3 Bob PERSON
4 9 likes NOTHING
10 15 pizza FOOD
16 17 . NOTHING
然后您可以使用以下方法过滤掉行：
df[df['Tag'] == 'PERSON']
输出：
First Last Word Tag
0 1 I PERSON
0 3 Bob PERSON