小编Md.*_*man的帖子

有没有办法在scikit学习库上实现跳过克？我已经手动生成了一个包含n-skim gram的列表,并将其作为scikit-learn方法的词汇表传递给skipgrams .

不幸的是,它在预测方面的表现非常差:只有63%的准确率.但是,CountVectorizer()使用CountVectorizer()默认代码时,我的准确率为77-80%.

是否有更好的方法来实施scikit中的skip-gram学习？

这是我的代码部分:

corpus = GetCorpus() # This one get text from file as a list

vocabulary = list(GetVocabulary(corpus,k,n))  
# this one returns a k-skip n-gram   

vec = CountVectorizer(
          tokenizer=lambda x: x.split(),
          ngram_range=(2,2),
          stop_words=stopWords,
          vocabulary=vocabulary)

public class MainActivity extends AppCompatActivity {

private static final String TAG = "MainActivity";
private static final int REQUEST_CODE = 1234;
private int mScreenDensity;
private MediaProjectionManager mProjectionManager;
private static final int DISPLAY_WIDTH = 720;
private static final int DISPLAY_HEIGHT = 1280;
private MediaProjection mMediaProjection;
private VirtualDisplay mVirtualDisplay;
private ToggleButton mToggleButton;
private MediaRecorder mMediaRecorder;
private static final SparseIntArray ORIENTATIONS = new SparseIntArray();
private static final int REQUEST_PERMISSIONS = 10;

static {
    ORIENTATIONS.append(Surface.ROTATION_0, 90);
    ORIENTATIONS.append(Surface.ROTATION_90, 0);
    ORIENTATIONS.append(Surface.ROTATION_180, 270);
    ORIENTATIONS.append(Surface.ROTATION_270, 180);
}

@Override
public …