1.在cnn的嵌入层看到如下代码

embedding=tf.get_variable('embedding',[3,2])

这时创建了一个类似3x2矩阵的变量，而且这个变量是的。所以要进行初始化，或者赋值。

embedding=tf.get_variable('embedding',[3,2])
session = tf.Session()
session.run(tf.global_variables_initializer())

这两行代码就可以进行初始化，先捕捉变量再进行变量的初始化，结果如下：

2.在tensorflow中定义一个变量
如下：
`

x = tf.Variable(1)

此时定义了一个变量x，那个1是个0阶张量
但此时只是定义了一个结构类型是变量 x，初始值是1，单x现在在内存中还不是1.因为tensorflow实际是以graph图表结构的形式运行的，在执行sess.run（传入需要取值的节点）时才去计算该图表某个节点的值，在此之前的操作都是为了构建此graph的结构并没有真正的赋于实际的值。执行variable（1）时也就是只是定义结构（类型为变量，初始值为1）。只有执行变量初始化方法时才赋予其定义的值。
with tf.Session().as_default() as sess:
sess.run(tf.global_variables_initializer())
print(sess.run(x))`
3.

4.关于tf.nn.embedding_look()的使用说明，这个实在看textcnn工程发现的。现有分类样本如下，只举很小的一个例子。

体育   鲍勃库西奖归谁属？ NCAA最强控卫是坎巴还是弗神新浪体育讯如今，本赛季的NCAA进入到了末段，各项奖项的评选结果也即将出炉，其中评选最佳控卫的鲍勃-库西奖就将在下周最终四强战时公布，
鲍勃-库西奖是由奈史密斯篮球名人堂提供，旨在奖励年度最佳大学控卫。最终获奖的球员也即将在以下几名热门人选中产生。〈〈〈 NCAA疯狂三月专题主页上线，点击链接查看精彩内容吉梅尔-弗雷戴特，
杨百翰大学“弗神”吉梅尔-弗雷戴特一直都备受关注，他不仅仅是一名射手，他会用“终结对手脚踝”
娱乐  林俊杰为电影《夏日乐悠悠》献首唱HOLD住全场新浪娱乐讯 近日，林俊杰(微博)在成都举办歌友会，现场演唱了其为由马楚成(微博)执导的电影《夏日乐悠悠》创作的主题曲《LOVE YOU YOU》，
这是林俊杰首次公开演唱这首新创作的歌曲却在现场出乎意料的引发全场歌迷大合唱场面感人。目前电影《夏日乐悠悠》计划于9月30日在全国上映，该片由当红明星Angelababy(微博)、彭于晏(微博)、朱雨辰(微博)、周扬(微博)
游戏  《海湾战争-空中王者》歼-20保卫升空一款绚丽的战斗画面，横越海、陆、空三大战场，加上紧张的剧情，就构成了今天小编给各大玩家介绍的这款射击游戏：《海湾战争-空中王者》。剧情介绍：海湾地区储存了丰富的石油和天然气，加上地处亚非欧交界，历为兵家必争之地。经过新一次全球经济危机之后，海湾战争一触即发，中国也被卷入纷争之中。为了掌握实地军情，我军派出了有空中王者之称的歼-20，远赴海湾地区……更多精彩震撼感觉，立即下载该款游戏尽情体验吧。玩家交流才是王道，讯易游戏玩家交流中心

a:现在开始读取数据，需要将样本进行进行向量化。上面共有三个标签，分别读取每个标签的第一行，通过如下代码实现：

def read_file(filename):"""读取文件数据"""contents, labels = [], []with open_file(filename) as f:for line in f:try:label, content = line.strip().split('\t')if content:contents.append(list(native_content(content)))labels.append(native_content(label))except:passreturn contents, labels

经过这段代码后 labels里包含[‘体育’,‘娱乐’,‘游戏’]

`
contents [[‘鲍’, ‘勃’, ‘库’, ‘西’, ‘奖’, ‘归’, ‘谁’, ‘属’, ‘？’, ’ ', ‘N’, ‘C’, ‘A’, ‘A’, ‘最’, ‘强’, ‘控’, ‘卫’, ‘是’, ‘坎’, ‘巴’, ‘还’, ‘是’, ‘弗’, ‘神’, ‘新’, ‘浪’, ‘体’, ‘育’, ‘讯’, ‘如’, ‘今’, ‘，’, ‘本’, ‘赛’, ‘季’, ‘的’, ‘N’, ‘C’, ‘A’, ‘A’, ‘进’, ‘入’, ‘到’, ‘了’, ‘末’, ‘段’, ‘，’, ‘各’, ‘项’, ‘奖’, ‘项’, ‘的’, ‘评’, ‘选’, ‘结’, ‘果’, ‘也’, ‘即’, ‘将’, ‘出’, ‘炉’, ‘，’, ‘其’, ‘中’, ‘评’, ‘选’, ‘最’, ‘佳’, ‘控’, ‘卫’, ‘的’, ‘鲍’, ‘勃’, ‘-’, ‘库’, ‘西’, ‘奖’, ‘就’, ‘将’, ‘在’, ‘下’, ‘周’, ‘最’, ‘终’, ‘四’, ‘强’, ‘战’, ‘时’, ‘公’, ‘布’, ‘，’], [‘林’, ‘俊’, ‘杰’, ‘为’, ‘电’, ‘影’, ‘《’, ‘夏’, ‘日’, ‘乐’, ‘悠’, ‘悠’, ‘》’, ‘献’, ‘首’, ‘唱’, ‘H’, ‘O’, ‘L’, ‘D’, ‘住’, ‘全’, ‘场’, ‘新’, ‘浪’, ‘娱’, ‘乐’, ‘讯’, ’ ', ‘近’, ‘日’, ‘，’, ‘林’, ‘俊’, ‘杰’, ‘(’, ‘微’, ‘博’, ‘)’, ‘在’, ‘成’, ‘都’, ‘举’, ‘办’, ‘歌’, ‘友’, ‘会’, ‘，’, ‘现’, ‘场’, ‘演’, ‘唱’, ‘了’, ‘其’, ‘为’, ‘由’, ‘马’, ‘楚’, ‘成’, ‘(’, ‘微’, ‘博’, ‘)’, ‘执’, ‘导’, ‘的’, ‘电’, ‘影’, ‘《’, ‘夏’, ‘日’, ‘乐’, ‘悠’, ‘悠’, ‘》’, ‘创’, ‘作’, ‘的’, ‘主’, ‘题’, ‘曲’, ‘《’, ‘L’, ‘O’, ‘V’, ‘E’, ’ ‘, ‘Y’, ‘O’, ‘U’, ’ ‘, ‘Y’, ‘O’, ‘U’, ‘》’, ‘，’], [’《’, ‘海’, ‘湾’, ‘战’, ‘争’, ‘-’, ‘空’, ‘中’, ‘王’, ‘者’, ‘》’, ‘歼’, ‘-’, ‘2’, ‘0’, ‘保’, ‘卫’, ‘升’, ‘空’, ‘一’, ‘款’, ‘绚’, ‘丽’, ‘的’, ‘战’, ‘斗’, ‘画’, ‘面’, ‘，’, ‘横’, ‘越’, ‘海’, ‘、’, ‘陆’, ‘、’, ‘空’, ‘三’, ‘大’, ‘战’, ‘场’, ‘，’, ‘加’, ‘上’, ‘紧’, ‘张’, ‘的’, ‘剧’, ‘情’, ‘，’, ‘就’, ‘构’, ‘成’, ‘了’, ‘今’, ‘天’, ‘小’, ‘编’, ‘给’, ‘各’, ‘大’, ‘玩’, ‘家’, ‘介’, ‘绍’, ‘的’, ‘这’, ‘款’, ‘射’, ‘击’, ‘游’, ‘戏’, ‘：’, ‘《’, ‘海’, ‘湾’, ‘战’, ‘争’, ‘-’, ‘空’, ‘中’, ‘王’, ‘者’, ‘》’, ‘。’, ‘剧’, ‘情’, ‘介’, ‘绍’, ‘：’, ‘海’, ‘湾’, ‘地’, ‘区’, ‘储’, ‘存’, ‘了’, ‘丰’, ‘富’, ‘的’, ‘石’, ‘油’, ‘和’, ‘天’, ‘然’, ‘气’, ‘，’, ‘加’, ‘上’, ‘地’, ‘处’, ‘亚’, ‘非’, ‘欧’, ‘交’, ‘界’, ‘，’, ‘历’, ‘为’, ‘兵’, ‘家’, ‘必’, ‘争’, ‘之’, ‘地’, ‘。’, ‘经’, ‘过’, ‘新’, ‘一’, ‘次’, ‘全’, ‘球’, ‘经’, ‘济’, ‘危’, ‘机’, ‘之’, ‘后’, ‘，’, ‘海’, ‘湾’, ‘战’, ‘争’, ‘一’, ‘触’, ‘即’, ‘发’, ‘，’, ‘中’, ‘国’, ‘也’, ‘被’, ‘卷’, ‘入’, ‘纷’, ‘争’, ‘之’, ‘中’, ‘。’, ‘为’, ‘了’, ‘掌’, ‘握’, ‘实’, ‘地’, ‘军’, ‘情’, ‘，’, ‘我’, ‘军’, ‘派’, ‘出’, ‘了’, ‘有’, ‘空’, ‘中’, ‘王’, ‘者’, ‘之’, ‘称’, ‘的’, ‘歼’, ‘-’, ‘2’, ‘0’, ‘，’, ‘远’, ‘赴’, ‘海’, ‘湾’, ‘地’, ‘区’, ‘…’, ‘…’, ‘更’, ‘多’, ‘精’, ‘彩’, ‘震’, ‘撼’, ‘感’, ‘觉’, ‘，’, ‘立’, ‘即’, ‘下’, ‘载’, ‘该’, ‘款’, ‘游’, ‘戏’, ‘尽’, ‘情’, ‘体’, ‘验’, ‘吧’, ‘。’, ‘玩’, ‘家’, ‘交’, ‘流’, ‘才’, ‘是’, ‘王’, ‘道’, ‘，’, ‘讯’, ‘易’, ‘游’, ‘戏’, ‘玩’, ‘家’, ‘交’, ‘流’, ‘中’, ‘心’]]

上面是contents包含的内容。
接着构建词汇表，准确来说是字汇表，我看的文章是字符集的文本分类。构建词汇表可以根据以下代码实现

def build_vocab(train_dir, vocab_dir, vocab_size=5000):train_dir就是上面样本的路径data_train, _ = read_file(train_dir)#data_train就是上面的contentsall_data = []for content in data_train:   #content对应的就是标签后的那一行print('content',content)all_data.extend(content)print('all_data:',all_data)all_data的内容可以和data_train比较下all_data:all_data: ['鲍', '勃', '库', '西', '奖', '归', '谁', '属', '？', ' ', 'N', 'C', 'A', 'A', '最', '强', '控', '卫', '是', '坎', '巴', '还', '是', '弗', '神', '新', '浪', '体', '育', '讯', '如', '今', '，', '本', '赛', '季', '的', 'N', 'C', 'A', 'A', '进', '入', '到', '了', '末', '段', '，', '各', '项', '奖', '项', '的', '评', '选', '结', '果', '也', '即', '将', '出', '炉', '，', '其', '中', '评', '选', '最', '佳', '控', '卫', '的', '鲍', '勃', '-', '库', '西', '奖', '就', '将', '在', '下', '周', '最', '终', '四', '强', '战', '时', '公', '布', '，', '林', '俊', '杰', '为', '电', '影', '《', '夏', '日', '乐', '悠', '悠', '》', '献', '首', '唱', 'H', 'O', 'L', 'D', '住', '全', '场', '新', '浪', '娱', '乐', '讯', ' ', '近', '日', '，', '林', '俊', '杰', '(', '微', '博', ')', '在', '成', '都', '举', '办', '歌', '友', '会', '，', '现', '场', '演', '唱', '了', '其', '为', '由', '马', '楚', '成', '(', '微', '博', ')', '执', '导', '的', '电', '影', '《', '夏', '日', '乐', '悠', '悠', '》', '创', '作', '的', '主', '题', '曲', '《', 'L', 'O', 'V', 'E', ' ', 'Y', 'O', 'U', ' ', 'Y', 'O', 'U', '》', '，', '《', '海', '湾', '战', '争', '-', '空', '中', '王', '者', '》', '歼', '-', '2', '0', '保', '卫', '升', '空', '一', '款', '绚', '丽', '的', '战', '斗', '画', '面', '，', '横', '越', '海', '、', '陆', '、', '空', '三', '大', '战', '场', '，', '加', '上', '紧', '张', '的', '剧', '情', '，', '就', '构', '成', '了', '今', '天', '小', '编', '给', '各', '大', '玩', '家', '介', '绍', '的', '这', '款', '射', '击', '游', '戏', '：', '《', '海', '湾', '战', '争', '-', '空', '中', '王', '者', '》', '。', '剧', '情', '介', '绍', '：', '海', '湾', '地', '区', '储', '存', '了', '丰', '富', '的', '石', '油', '和', '天', '然', '气', '，', '加', '上', '地', '处', '亚', '非', '欧', '交', '界', '，', '历', '为', '兵', '家', '必', '争', '之', '地', '。', '经', '过', '新', '一', '次', '全', '球', '经', '济', '危', '机', '之', '后', '，', '海', '湾', '战', '争', '一', '触', '即', '发', '，', '中', '国', '也', '被', '卷', '入', '纷', '争', '之', '中', '。', '为', '了', '掌', '握', '实', '地', '军', '情', '，', '我', '军', '派', '出', '了', '有', '空', '中', '王', '者', '之', '称', '的', '歼', '-', '2', '0', '，', '远', '赴', '海', '湾', '地', '区', '…', '…', '更', '多', '精', '彩', '震', '撼', '感', '觉', '，', '立', '即', '下', '载', '该', '款', '游', '戏', '尽', '情', '体', '验', '吧', '。', '玩', '家', '交', '流', '才', '是', '王', '道', '，', '讯', '易', '游', '戏', '玩', '家', '交', '流', '中', '心']counter = Counter(all_data)counter的内容 Counter({'，': 18, '的': 10, '中': 7, '了': 6, '战': 6, '海': 6, '-': 5, '《': 5, '》': 5, '湾': 5, '争': 5, '空': 5, '地': 5, ' ': 4, 'A': 4, '为': 4, '悠': 4, 'O': 4, '王': 4, '情': 4, '家': 4, '。': 4, '之': 4, '奖': 3, '最': 3, '卫': 3, '是': 3, '新': 3, '讯': 3, '即': 3, '日': 3, '乐': 3, '场': 3, '成': 3, '者': 3, '一': 3, '款': 3, '玩': 3, '游': 3, '戏': 3, '交': 3, '鲍': 2, '勃': 2, '库': 2, '西': 2, 'N': 2, 'C': 2, '强': 2, '控': 2, '浪': 2, '体': 2, '今': 2, '入': 2, '各': 2, '项': 2, '评': 2, '选': 2, '也': 2, '将': 2, '出': 2, '其': 2, '就': 2, '在': 2, '下': 2, '林': 2, '俊': 2, '杰': 2, '电': 2, '影': 2, '夏': 2, '唱': 2, 'L': 2, '全': 2, '(': 2, '微': 2, '博': 2, ')': 2, 'Y': 2, 'U': 2, '歼': 2, '2': 2, '0': 2, '、': 2, '大': 2, '加': 2, '上': 2, '剧': 2, '天': 2, '介': 2, '绍': 2, '：': 2, '区': 2, '经': 2, '军': 2, '…': 2, '流': 2, '归': 1, '谁': 1, '属': 1, '？': 1, '坎': 1, '巴': 1, '还': 1, '弗': 1, '神': 1, '育': 1, '如': 1, '本': 1, '赛': 1, '季': 1, '进': 1, '到': 1, '末': 1, '段': 1, '结': 1, '果': 1, '炉': 1, '佳': 1, '周': 1, '终': 1, '四': 1, '时': 1, '公': 1, '布': 1, '献': 1, '首': 1, 'H': 1, 'D': 1, '住': 1, '娱': 1, '近': 1, '都': 1, '举': 1, '办': 1, '歌': 1, '友': 1, '会': 1, '现': 1, '演': 1, '由': 1, '马': 1, '楚': 1, '执': 1, '导': 1, '创': 1, '作': 1, '主': 1, '题': 1, '曲': 1, 'V': 1, 'E': 1, '保': 1, '升': 1, '绚': 1, '丽': 1, '斗': 1, '画': 1, '面': 1, '横': 1, '越': 1, '陆': 1, '三': 1, '紧': 1, '张': 1, '构': 1, '小': 1, '编': 1, '给': 1, '这': 1, '射': 1, '击': 1, '储': 1, '存': 1, '丰': 1, '富': 1, '石': 1, '油': 1, '和': 1, '然': 1, '气': 1, '处': 1, '亚': 1, '非': 1, '欧': 1, '界': 1, '历': 1, '兵': 1, '必': 1, '过': 1, '次': 1, '球': 1, '济': 1, '危': 1, '机': 1, '后': 1, '触': 1, '发': 1, '国': 1, '被': 1, '卷': 1, '纷': 1, '掌': 1, '握': 1, '实': 1, '我': 1, '派': 1, '有': 1, '称': 1, '远': 1, '赴': 1, '更': 1, '多': 1, '精': 1, '彩': 1, '震': 1, '撼': 1, '感': 1, '觉': 1, '立': 1, '载': 1, '该': 1, '尽': 1, '验': 1, '吧': 1, '才': 1, '道': 1, '易': 1, '心': 1})count_pairs = counter.most_common(vocab_size - 1)#vocab_size是不同字符的个数，自己定取多少个[('，', 18), ('的', 10), ('中', 7), ('了', 6), ('战', 6), ('海', 6), ('-', 5), ('《', 5), ('》', 5), ('湾', 5), ('争', 5), ('空', 5), ('地', 5), (' ', 4), ('A', 4), ('为', 4), ('悠', 4), ('O', 4), ('王', 4), ('情', 4), ('家', 4), ('。', 4), ('之', 4), ('奖', 3), ('最', 3), ('卫', 3), ('是', 3), ('新', 3), ('讯', 3), ('即', 3), ('日', 3), ('乐', 3), ('场', 3), ('成', 3), ('者', 3), ('一', 3), ('款', 3), ('玩', 3), ('游', 3), ('戏', 3), ('交', 3), ('鲍', 2), ('勃', 2), ('库', 2), ('西', 2), ('N', 2), ('C', 2), ('强', 2), ('控', 2), ('浪', 2), ('体', 2), ('今', 2), ('入', 2), ('各', 2), ('项', 2), ('评', 2), ('选', 2), ('也', 2), ('将', 2), ('出', 2), ('其', 2), ('就', 2), ('在', 2), ('下', 2), ('林', 2), ('俊', 2)]words, _ = list(zip(*count_pairs))#自己参考zip的用法，zip返回的是个迭代器，里面是tuple组成的，因此用列表显示，在这里list(zip(*count_pairs))包含两个tuple，一个是词汇字符，一个是字符相对应所出现的频次print(' words', words)#words是tuplewords  ('，', '的', '中', '了', '战', '海', '-', '《', '》', '湾', '争', '空', '地', ' ', 'A', '为', '悠', 'O', '王', '情', '家', '。', '之', '奖', '最', '卫', '是', '新', '讯', '即', '日', '乐', '场', '成', '者', '一', '款', '玩', '游', '戏', '交', '鲍', '勃', '库', '西', 'N', 'C', '强', '控', '浪', '体', '今', '入', '各', '项', '评', '选', '也', '将', '出', '其', '就', '在', '下', '林', '俊')# 添加一个 <PAD> 来将所有文本pad为同一长度words = ['<PAD>'] + list(words)open_file(vocab_dir, mode='w').write('\n'.join(words) + '\n')#vocab_dir是建立的字汇表的地址词汇表建立后如下图

b：建立词汇表后，将词汇表数字化，以及内容数字化.。词汇表数字化使用以下代码

def read_vocab(vocab_dir):"""读取词汇表"""# words = open_file(vocab_dir).read().strip().split('\n')with open_file(vocab_dir) as fp:# 如果是py2 则每个值都转化为unicodewords = [_.strip() for _ in fp.readlines()]print('words2', words)word_to_id = dict(zip(words, range(len(words))))print('word_to_id',word_to_id)return words, word_to_id

word_to_id {’’: 0, ‘，’: 1, ‘的’: 2, ‘中’: 3, ‘了’: 4, ‘战’: 5, ‘海’: 6, ‘-’: 7, ‘《’: 8, ‘》’: 9, ‘湾’: 10, ‘争’: 11, ‘空’: 12, ‘地’: 13, ‘’: 14, ‘A’: 15, ‘为’: 16, ‘悠’: 17, ‘O’: 18, ‘王’: 19, ‘情’: 20, ‘家’: 21, ‘。’: 22, ‘之’: 23, ‘奖’: 24, ‘最’: 25, ‘卫’: 26, ‘是’: 27, ‘新’: 28, ‘讯’: 29, ‘即’: 30, ‘日’: 31, ‘乐’: 32, ‘场’: 33, ‘成’: 34, ‘者’: 35, ‘一’: 36, ‘款’: 37, ‘玩’: 38, ‘游’: 39, ‘戏’: 40, ‘交’: 41, ‘鲍’: 42, ‘勃’: 43, ‘库’: 44, ‘西’: 45, ‘N’: 46, ‘C’: 47, ‘强’: 48, ‘控’: 49, ‘浪’: 50, ‘体’: 51, ‘今’: 52, ‘入’: 53, ‘各’: 54, ‘项’: 55, ‘评’: 56, ‘选’: 57, ‘也’: 58, ‘将’: 59, ‘出’: 60, ‘其’: 61, ‘就’: 62, ‘在’: 63, ‘下’: 64, ‘林’: 65, ‘俊’: 66}

c:将种类进行数字化

def read_category():"""读取分类目录，固定"""categories = ['体育', '财经', '房产', '家居', '教育', '科技', '时尚', '时政', '游戏', '娱乐']categories = [native_content(x) for x in categories]cat_to_id = dict(zip(categories, range(len(categories))))return categories, cat_to_id

cat_to_id {‘体育’: 0, ‘财经’: 1, ‘房产’: 2, ‘家居’: 3, ‘教育’: 4, ‘科技’: 5, ‘时尚’: 6, ‘时政’: 7, ‘游戏’: 8, ‘娱乐’: 9}

d：在词汇表和种类都进行数字化后，现在进行文本内容，也就是上面contents的数字化

def process_file(filename, word_to_id, cat_to_id, max_length=600):"""将文件转换为id表示"""contents, labels = read_file(filename)print('----------------------fengexian--------------------------------------------')print('contents',contents)print('labels',labels)data_id, label_id = [], []for i in range(len(contents)):data_id.append([word_to_id[x] for x in contents[i] if x in word_to_id])print('data_id',data_id)label_id.append(cat_to_id[labels[i]])print('label_id', label_id)print('length',len(data_id[0]))# 使用keras提供的pad_sequences来将文本pad为固定长度x_pad = kr.preprocessing.sequence.pad_sequences(data_id, 68)#x_pad可以看做二维矩阵，每一行都是每一行内容的数字化。长度可以调整，如果不够就用0补，这样每行的维数机一样，此处x_pad形式很重要print('x_pad',x_pad)y_pad = kr.utils.to_categorical(label_id, num_classes=len(cat_to_id))  # 将标签转换为one-hot表示print('y_pad',y_pad)return x_pad, y_pad

contents [[‘鲍’, ‘勃’, ‘库’, ‘西’, ‘奖’, ‘归’, ‘谁’, ‘属’, ‘？’, ’ ', ‘N’, ‘C’, ‘A’, ‘A’, ‘最’, ‘强’, ‘控’, ‘卫’, ‘是’, ‘坎’, ‘巴’, ‘还’, ‘是’, ‘弗’, ‘神’, ‘新’, ‘浪’, ‘体’, ‘育’, ‘讯’, ‘如’, ‘今’, ‘，’, ‘本’, ‘赛’, ‘季’, ‘的’, ‘N’, ‘C’, ‘A’, ‘A’, ‘进’, ‘入’, ‘到’, ‘了’, ‘末’, ‘段’, ‘，’, ‘各’, ‘项’, ‘奖’, ‘项’, ‘的’, ‘评’, ‘选’, ‘结’, ‘果’, ‘也’, ‘即’, ‘将’, ‘出’, ‘炉’, ‘，’, ‘其’, ‘中’, ‘评’, ‘选’, ‘最’, ‘佳’, ‘控’, ‘卫’, ‘的’, ‘鲍’, ‘勃’, ‘-’, ‘库’, ‘西’, ‘奖’, ‘就’, ‘将’, ‘在’, ‘下’, ‘周’, ‘最’, ‘终’, ‘四’, ‘强’, ‘战’, ‘时’, ‘公’, ‘布’, ‘，’], [‘林’, ‘俊’, ‘杰’, ‘为’, ‘电’, ‘影’, ‘《’, ‘夏’, ‘日’, ‘乐’, ‘悠’, ‘悠’, ‘》’, ‘献’, ‘首’, ‘唱’, ‘H’, ‘O’, ‘L’, ‘D’, ‘住’, ‘全’, ‘场’, ‘新’, ‘浪’, ‘娱’, ‘乐’, ‘讯’, ’ ', ‘近’, ‘日’, ‘，’, ‘林’, ‘俊’, ‘杰’, ‘(’, ‘微’, ‘博’, ‘)’, ‘在’, ‘成’, ‘都’, ‘举’, ‘办’, ‘歌’, ‘友’, ‘会’, ‘，’, ‘现’, ‘场’, ‘演’, ‘唱’, ‘了’, ‘其’, ‘为’, ‘由’, ‘马’, ‘楚’, ‘成’, ‘(’, ‘微’, ‘博’, ‘)’, ‘执’, ‘导’, ‘的’, ‘电’, ‘影’, ‘《’, ‘夏’, ‘日’, ‘乐’, ‘悠’, ‘悠’, ‘》’, ‘创’, ‘作’, ‘的’, ‘主’, ‘题’, ‘曲’, ‘《’, ‘L’, ‘O’, ‘V’, ‘E’, ’ ‘, ‘Y’, ‘O’, ‘U’, ’ ‘, ‘Y’, ‘O’, ‘U’, ‘》’, ‘，’], [’《’, ‘海’, ‘湾’, ‘战’, ‘争’, ‘-’, ‘空’, ‘中’, ‘王’, ‘者’, ‘》’, ‘歼’, ‘-’, ‘2’, ‘0’, ‘保’, ‘卫’, ‘升’, ‘空’, ‘一’, ‘款’, ‘绚’, ‘丽’, ‘的’, ‘战’, ‘斗’, ‘画’, ‘面’, ‘，’, ‘横’, ‘越’, ‘海’, ‘、’, ‘陆’, ‘、’, ‘空’, ‘三’, ‘大’, ‘战’, ‘场’, ‘，’, ‘加’, ‘上’, ‘紧’, ‘张’, ‘的’, ‘剧’, ‘情’, ‘，’, ‘就’, ‘构’, ‘成’, ‘了’, ‘今’, ‘天’, ‘小’, ‘编’, ‘给’, ‘各’, ‘大’, ‘玩’, ‘家’, ‘介’, ‘绍’, ‘的’, ‘这’, ‘款’, ‘射’, ‘击’, ‘游’, ‘戏’, ‘：’, ‘《’, ‘海’, ‘湾’, ‘战’, ‘争’, ‘-’, ‘空’, ‘中’, ‘王’, ‘者’, ‘》’, ‘。’, ‘剧’, ‘情’, ‘介’, ‘绍’, ‘：’, ‘海’, ‘湾’, ‘地’, ‘区’, ‘储’, ‘存’, ‘了’, ‘丰’, ‘富’, ‘的’, ‘石’, ‘油’, ‘和’, ‘天’, ‘然’, ‘气’, ‘，’, ‘加’, ‘上’, ‘地’, ‘处’, ‘亚’, ‘非’, ‘欧’, ‘交’, ‘界’, ‘，’, ‘历’, ‘为’, ‘兵’, ‘家’, ‘必’, ‘争’, ‘之’, ‘地’, ‘。’, ‘经’, ‘过’, ‘新’, ‘一’, ‘次’, ‘全’, ‘球’, ‘经’, ‘济’, ‘危’, ‘机’, ‘之’, ‘后’, ‘，’, ‘海’, ‘湾’, ‘战’, ‘争’, ‘一’, ‘触’, ‘即’, ‘发’, ‘，’, ‘中’, ‘国’, ‘也’, ‘被’, ‘卷’, ‘入’, ‘纷’, ‘争’, ‘之’, ‘中’, ‘。’, ‘为’, ‘了’, ‘掌’, ‘握’, ‘实’, ‘地’, ‘军’, ‘情’, ‘，’, ‘我’, ‘军’, ‘派’, ‘出’, ‘了’, ‘有’, ‘空’, ‘中’, ‘王’, ‘者’, ‘之’, ‘称’, ‘的’, ‘歼’, ‘-’, ‘2’, ‘0’, ‘，’, ‘远’, ‘赴’, ‘海’, ‘湾’, ‘地’, ‘区’, ‘…’, ‘…’, ‘更’, ‘多’, ‘精’, ‘彩’, ‘震’, ‘撼’, ‘感’, ‘觉’, ‘，’, ‘立’, ‘即’, ‘下’, ‘载’, ‘该’, ‘款’, ‘游’, ‘戏’, ‘尽’, ‘情’, ‘体’, ‘验’, ‘吧’, ‘。’, ‘玩’, ‘家’, ‘交’, ‘流’, ‘才’, ‘是’, ‘王’, ‘道’, ‘，’, ‘讯’, ‘易’, ‘游’, ‘戏’, ‘玩’, ‘家’, ‘交’, ‘流’, ‘中’, ‘心’]]
labels [‘体育’, ‘娱乐’, ‘游戏’]
data_id [[42, 43, 44, 45, 24, 46, 47, 15, 15, 25, 48, 49, 26, 27, 27, 28, 50, 51, 29, 52, 1, 2, 46, 47, 15, 15, 53, 4, 1, 54, 55, 24, 55, 2, 56, 57, 58, 30, 59, 60, 1, 61, 3, 56, 57, 25, 49, 26, 2, 42, 43, 7, 44, 45, 24, 62, 59, 63, 64, 25, 48, 5, 1], [65, 66, 16, 8, 31, 32, 17, 17, 9, 18, 33, 28, 50, 32, 29, 31, 1, 65, 66, 63, 34, 1, 33, 4, 61, 16, 34, 2, 8, 31, 32, 17, 17, 9, 2, 8, 18, 18, 18, 9, 1], [8, 6, 10, 5, 11, 7, 12, 3, 19, 35, 9, 7, 26, 12, 36, 37, 2, 5, 1, 6, 12, 5, 33, 1, 2, 20, 1, 62, 34, 4, 52, 54, 38, 21, 2, 37, 39, 40, 8, 6, 10, 5, 11, 7, 12, 3, 19, 35, 9, 22, 20, 6, 10, 13, 4, 2, 1, 13, 41, 1, 16, 21, 11, 23, 13, 22, 28, 36, 23, 1, 6, 10, 5, 11, 36, 30, 1, 3, 58, 53, 11, 23, 3, 22, 16, 4, 13, 20, 1, 60, 4, 12, 3, 19, 35, 23, 2, 7, 1, 6, 10, 13, 1, 30, 64, 37, 39, 40, 20, 51, 22, 38, 21, 41, 27, 19, 1, 29, 39, 40, 38, 21, 41, 3]]
label_id [0, 9, 8]
x_pad [[ 0 0 0 0 0 42 43 44 45 24 46 47 15 15 25 48 49 26 27 27 28 50 51 29
52 1 2 46 47 15 15 53 4 1 54 55 24 55 2 56 57 58 30 59 60 1 61 3
56 57 25 49 26 2 42 43 7 44 45 24 62 59 63 64 25 48 5 1]
[ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 65 66 16 8 31 32 17 17 9 18 33 28 50 32 29 31 1 65 66 63 34
1 33 4 61 16 34 2 8 31 32 17 17 9 2 8 18 18 18 9 1]
[ 1 13 41 1 16 21 11 23 13 22 28 36 23 1 6 10 5 11 36 30 1 3 58 53
11 23 3 22 16 4 13 20 1 60 4 12 3 19 35 23 2 7 1 6 10 13 1 30
64 37 39 40 20 51 22 38 21 41 27 19 1 29 39 40 38 21 41 3]]
y_pad [[1. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[0. 0. 0. 0. 0. 0. 0. 0. 0. 1.]
[0. 0. 0. 0. 0. 0. 0. 0. 1. 0.]]
这里的x_pad和y_pad的形式是非常重要的
e：生成批次数据（批量梯度下降，每次更新部分权重）

def batch_iter(x, y, batch_size=1):"""生成批次数据"""#batchs_size每批的训练大小print(1111111111111111111111)data_len = len(x)print(22222222222222222222222)print(len(x))num_batch = int((data_len - 1) / batch_size) + 1indices = np.random.permutation(np.arange(data_len))x_shuffle = x[indices]y_shuffle = y[indices]for i in range(num_batch):start_id = i * batch_sizeend_id = min((i + 1) * batch_size, data_len)yield x_shuffle[start_id:end_id], y_shuffle[start_id:end_id]x和y就是x_pad,  y_pad

x_shuffle[start_id:end_id]:
[[ 1 13 41 1 16 21 11 23 13 22 28 36 23 1 6 10 5 11 36 30 1 3 58 53
11 23 3 22 16 4 13 20 1 60 4 12 3 19 35 23 2 7 1 6 10 13 1 30
64 37 39 40 20 51 22 38 21 41 27 19 1 29 39 40 38 21 41 3]]

y_shuffle[start_id:end_id]:
[[0. 0. 0. 0. 0. 0. 0. 0. 1. 0.]]
到这里你可能就明白yield和return的关系和区别了，带yield的函数是一个生成器，而不是一个函数了，这个生成器有一个函数就是next函数，next就相当于“下一步”生成哪个数，这一次的next开始的地方是接着上一次的next停止的地方执行的，所以调用next的时候，生成器并不会从foo函数的开始执行，只是接着上一步停止的地方开始，然后遇到yield后，return出要生成的数，此步就结束。
---------------------
版权声明：本文为CSDN博主「冯爽朗」的原创文章，遵循CC 4.0 by-sa版权协议，转载请附上原文出处链接及本声明。
原文链接：https://blog.csdn.net/mieleizhi0522/article/details/82142856
f：生成批次数据后

…下面说cnn的模型
第一层嵌入层

词向量映射

      with tf.device('/cpu:0'):embedding = tf.get_variable('embedding', [self.config.vocab_size, self.config.embedding_dim])#创建了一个66x6 的矩阵 变量 ：self.config.vocab_size：66self.config.embedding_dim：6.一个字向量的维度       embedding_inputs = tf.nn.embedding_lookup(embedding, self.input_x)具体使用方法如下init = tf.global_variables_initializer()
embedding = tf.get_variable('embedding', [66, 6])
embedding_inputs = tf.nn.embedding_lookup(embedding, self.input_x)---------------------------------------------------------------------
input_x,input_y=batch_iter(xx, yy, batch_size=1)
print('input_x',input_x)session = tf.Session()
embedding = tf.get_variable('embedding', [66, 6])
init = tf.global_variables_initializer()
session.run(init)print(session.run(embedding))
embedding_inputs = tf.nn.embedding_lookup(embedding, input_x)
print('embedding_inputs',session.run(embedding_inputs))

input_x
[[ 1 13 41 1 16 21 11 23 13 22 28 36 23 1 6 10 5 11 36 30 1 3 58 53
11 23 3 22 16 4 13 20 1 60 4 12 3 19 35 23 2 7 1 6 10 13 1 30
64 37 39 40 20 51 22 38 21 41 27 19 1 29 39 40 38 21 41 3]]

随机初始化的embedding
[[ 0.23691297 -0.13519788 -0.17290153 -0.06241937 -0.05596925 -0.14235374]
[-0.24571183 -0.05539428 0.15108624 -0.28432798 -0.22648674 -0.10877597]
[-0.26434112 -0.22787322 0.15392077 0.13395855 0.09772652 -0.18564492]
[ 0.14206117 0.11317515 -0.05528121 -0.05800201 0.2836451 -0.11509615]
[-0.15899974 0.0658133 -0.28170696 -0.1541525 -0.07255393 0.13900462]
[ 0.08123055 0.2070345 -0.21851322 0.2779314 -0.08152774 0.24302483]
[-0.13986439 0.1926142 -0.2029559 0.26328248 -0.01662454 0.0631927 ]
[ 0.00827599 -0.19770452 -0.1522938 -0.02354962 0.04330006 0.25638068]
[-0.15717566 0.04652867 -0.02429664 0.23104036 0.03586441 0.08314246]
[-0.12403797 0.01842996 -0.24247035 0.16313967 0.01510769 -0.03565237]
[-0.1276895 -0.03677085 -0.19485027 0.11100221 0.06780049 -0.19717574]
[-0.13580988 0.18453643 -0.17972282 0.26238257 -0.0425182 0.04440424]
[-0.15336685 -0.05864498 0.28834653 0.24184489 0.21647209 -0.09705402]
[-0.2526745 -0.26769823 0.05841517 -0.27402672 0.2282387 0.00096136]
[ 0.13156486 -0.07441346 -0.09149754 0.12538543 -0.10951564 0.1413089 ]
[-0.15028904 -0.14594609 0.24763787 -0.10334942 0.15272108 -0.19012119]
[ 0.00184816 0.11910158 0.25898772 0.17714614 -0.21949929 -0.17058921]
[ 0.09135431 0.09987381 0.12331295 0.2502839 0.0403257 0.01253623]
[-0.09576726 -0.13063213 0.05545032 0.03134313 -0.06821208 0.07928216]
[-0.08161914 0.06568494 0.14552012 0.16656539 -0.1164633 -0.0806136 ]
[-0.10627581 0.06169692 0.24996006 0.07337654 0.14071411 0.04785666]
[ 0.25510186 -0.12553987 -0.24126157 0.21673346 0.14857164 0.25489032]
[-0.02202603 -0.1186287 0.21785933 0.0280619 0.0017722 0.14117366]
[-0.0535892 0.07938072 -0.2857826 -0.24757418 0.15214473 0.12090454]
[ 0.2737862 0.03695834 -0.06650086 -0.19313148 0.23660904 0.10879511]
[-0.28669882 0.07020518 0.06695697 -0.15798223 0.2875694 0.01970571]
[ 0.09032515 -0.15521571 0.0489772 0.18398878 0.22454399 0.0077832 ]
[ 0.08065709 0.19518784 -0.21418074 -0.0028725 -0.13971834 -0.01058358]
[-0.22876623 -0.12268238 -0.22381066 0.18612084 -0.1495206 0.12250948]
[ 0.1780909 0.10810417 -0.11891362 0.15357238 -0.21714506 -0.20010895]
[-0.08518472 -0.04857677 0.2497778 0.20756474 0.18140996 0.07720825]
[-0.21426567 -0.21975863 0.08462957 0.20638472 -0.05134472 -0.2534072 ]
[-0.14999406 -0.03410277 -0.28500742 0.05089357 -0.04791309 0.06126118]
[-0.26611164 0.1410932 -0.06462757 0.15332225 0.23302484 -0.12114364]
[-0.1494725 -0.08487259 -0.19930506 0.01653457 -0.2679908 0.0472686 ]
[ 0.06406471 0.05587634 -0.16891722 -0.09462978 0.04923108 0.13883916]
[-0.18711463 0.15864715 -0.15546672 0.20826581 0.2781158 -0.20220985]
[-0.06363125 0.12919554 0.14578372 -0.01874146 0.14159101 0.15668741]
[-0.11848497 -0.05116902 0.09591931 0.01752862 0.10756946 -0.04643519]
[-0.03719392 0.17145976 -0.18463153 -0.07408771 -0.16771841 0.17747995]
[ 0.22013861 0.16776392 -0.13306917 -0.04167722 -0.17991196 0.25754595]
[ 0.21247488 0.22724444 0.04058126 -0.01098627 0.1850268 0.04928353]
[-0.24639142 -0.04634091 0.00045529 -0.25679696 -0.05259308 -0.24200112]
[ 0.07113779 -0.25441384 -0.1824894 -0.13744785 -0.05465193 -0.28369966]
[ 0.1663951 0.05346683 -0.15225552 0.16807556 -0.15423337 -0.2456753 ]
[-0.03412658 0.08400601 0.1980876 -0.14623068 0.21056029 0.1800251 ]
[ 0.23165774 0.19597226 -0.2495565 -0.16574113 0.1525026 0.2566433 ]
[ 0.0630953 -0.2786023 -0.16081005 -0.02538711 -0.27419418 0.1025582 ]
[-0.25362533 0.02197579 -0.00617951 -0.14856662 0.11153242 0.23738497]
[-0.20182483 0.1876134 0.14892706 -0.15894248 0.10338148 0.28538036]
[-0.16415553 -0.2628631 -0.09699531 0.01421681 -0.01112854 0.09936133]
[-0.08882944 0.02432218 0.20564589 -0.11377952 0.1974028 0.16832677]
[ 0.25703806 -0.00071028 0.16990384 -0.10639983 -0.0831883 -0.09193347]
[-0.03947866 -0.17047392 -0.05723536 0.11617863 -0.1890268 0.18543178]
[ 0.11309478 -0.1983316 0.06863239 -0.14372447 -0.2759583 -0.05608061]
[-0.04162106 0.07707885 -0.27668318 -0.06008887 0.26514602 -0.11434726]
[-0.26868257 0.1296862 0.15483758 -0.15717937 0.21677113 0.19152117]
[-0.05864732 0.0278258 0.19932055 -0.0688611 -0.12320758 -0.12650473]
[-0.19307345 -0.08884493 -0.20212801 0.26287162 -0.25741225 0.25926608]
[ 0.24767327 0.25195622 -0.05632921 -0.23977941 -0.11114164 0.01533225]
[ 0.19079602 -0.16030592 -0.27824146 -0.2586743 0.13144979 -0.10577594]
[ 0.20931724 0.02251613 0.11316517 0.03618035 -0.04443905 -0.10311459]
[ 0.07290316 0.08613291 -0.15033028 -0.14176294 -0.03231186 0.05009356]
[ 0.06722745 0.0038574 -0.24184425 -0.26024234 -0.03536853 -0.01136014]
[ 0.13659015 0.17667219 0.26687944 -0.05812459 0.24382049 -0.02167881]
[-0.27565438 0.03991991 -0.15689795 -0.11174311 -0.1111488 0.18866181]]

embedding_inputs #这是cnn模型第二层的输入向量，可以看到这是一个三维的
[[[-0.24571183 -0.05539428 0.15108624 -0.28432798 -0.22648674
-0.10877597]
[-0.2526745 -0.26769823 0.05841517 -0.27402672 0.2282387
0.00096136]
[ 0.21247488 0.22724444 0.04058126 -0.01098627 0.1850268
0.04928353]
[-0.24571183 -0.05539428 0.15108624 -0.28432798 -0.22648674
-0.10877597]
[ 0.00184816 0.11910158 0.25898772 0.17714614 -0.21949929
-0.17058921]
[ 0.25510186 -0.12553987 -0.24126157 0.21673346 0.14857164
0.25489032]
[-0.13580988 0.18453643 -0.17972282 0.26238257 -0.0425182
0.04440424]
[-0.0535892 0.07938072 -0.2857826 -0.24757418 0.15214473
0.12090454]
[-0.2526745 -0.26769823 0.05841517 -0.27402672 0.2282387
0.00096136]
[-0.02202603 -0.1186287 0.21785933 0.0280619 0.0017722
0.14117366]
[-0.22876623 -0.12268238 -0.22381066 0.18612084 -0.1495206
0.12250948]
[-0.18711463 0.15864715 -0.15546672 0.20826581 0.2781158
-0.20220985]
[-0.0535892 0.07938072 -0.2857826 -0.24757418 0.15214473
0.12090454]
[-0.24571183 -0.05539428 0.15108624 -0.28432798 -0.22648674
-0.10877597]
[-0.13986439 0.1926142 -0.2029559 0.26328248 -0.01662454
0.0631927 ]
[-0.1276895 -0.03677085 -0.19485027 0.11100221 0.06780049
-0.19717574]
[ 0.08123055 0.2070345 -0.21851322 0.2779314 -0.08152774
0.24302483]
[-0.13580988 0.18453643 -0.17972282 0.26238257 -0.0425182
0.04440424]
[-0.18711463 0.15864715 -0.15546672 0.20826581 0.2781158
-0.20220985]
[-0.08518472 -0.04857677 0.2497778 0.20756474 0.18140996
0.07720825]
[-0.24571183 -0.05539428 0.15108624 -0.28432798 -0.22648674
-0.10877597]
[ 0.14206117 0.11317515 -0.05528121 -0.05800201 0.2836451
-0.11509615]
[-0.19307345 -0.08884493 -0.20212801 0.26287162 -0.25741225
0.25926608]
[-0.03947866 -0.17047392 -0.05723536 0.11617863 -0.1890268
0.18543178]
[-0.13580988 0.18453643 -0.17972282 0.26238257 -0.0425182
0.04440424]
[-0.0535892 0.07938072 -0.2857826 -0.24757418 0.15214473
0.12090454]
[ 0.14206117 0.11317515 -0.05528121 -0.05800201 0.2836451
-0.11509615]
[-0.02202603 -0.1186287 0.21785933 0.0280619 0.0017722
0.14117366]
[ 0.00184816 0.11910158 0.25898772 0.17714614 -0.21949929
-0.17058921]
[-0.15899974 0.0658133 -0.28170696 -0.1541525 -0.07255393
0.13900462]
[-0.2526745 -0.26769823 0.05841517 -0.27402672 0.2282387
0.00096136]
[-0.10627581 0.06169692 0.24996006 0.07337654 0.14071411
0.04785666]
[-0.24571183 -0.05539428 0.15108624 -0.28432798 -0.22648674
-0.10877597]
[ 0.19079602 -0.16030592 -0.27824146 -0.2586743 0.13144979
-0.10577594]
[-0.15899974 0.0658133 -0.28170696 -0.1541525 -0.07255393
0.13900462]
[-0.15336685 -0.05864498 0.28834653 0.24184489 0.21647209
-0.09705402]
[ 0.14206117 0.11317515 -0.05528121 -0.05800201 0.2836451
-0.11509615]
[-0.08161914 0.06568494 0.14552012 0.16656539 -0.1164633
-0.0806136 ]
[ 0.06406471 0.05587634 -0.16891722 -0.09462978 0.04923108
0.13883916]
[-0.0535892 0.07938072 -0.2857826 -0.24757418 0.15214473
0.12090454]
[-0.26434112 -0.22787322 0.15392077 0.13395855 0.09772652
-0.18564492]
[ 0.00827599 -0.19770452 -0.1522938 -0.02354962 0.04330006
0.25638068]
[-0.24571183 -0.05539428 0.15108624 -0.28432798 -0.22648674
-0.10877597]
[-0.13986439 0.1926142 -0.2029559 0.26328248 -0.01662454
0.0631927 ]
[-0.1276895 -0.03677085 -0.19485027 0.11100221 0.06780049
-0.19717574]
[-0.2526745 -0.26769823 0.05841517 -0.27402672 0.2282387
0.00096136]
[-0.24571183 -0.05539428 0.15108624 -0.28432798 -0.22648674
-0.10877597]
[-0.08518472 -0.04857677 0.2497778 0.20756474 0.18140996
0.07720825]
[ 0.13659015 0.17667219 0.26687944 -0.05812459 0.24382049
-0.02167881]
[-0.06363125 0.12919554 0.14578372 -0.01874146 0.14159101
0.15668741]
[-0.03719392 0.17145976 -0.18463153 -0.07408771 -0.16771841
0.17747995]
[ 0.22013861 0.16776392 -0.13306917 -0.04167722 -0.17991196
0.25754595]
[-0.10627581 0.06169692 0.24996006 0.07337654 0.14071411
0.04785666]
[-0.08882944 0.02432218 0.20564589 -0.11377952 0.1974028
0.16832677]
[-0.02202603 -0.1186287 0.21785933 0.0280619 0.0017722
0.14117366]
[-0.11848497 -0.05116902 0.09591931 0.01752862 0.10756946
-0.04643519]
[ 0.25510186 -0.12553987 -0.24126157 0.21673346 0.14857164
0.25489032]
[ 0.21247488 0.22724444 0.04058126 -0.01098627 0.1850268
0.04928353]
[ 0.08065709 0.19518784 -0.21418074 -0.0028725 -0.13971834
-0.01058358]
[-0.08161914 0.06568494 0.14552012 0.16656539 -0.1164633
-0.0806136 ]
[-0.24571183 -0.05539428 0.15108624 -0.28432798 -0.22648674
-0.10877597]
[ 0.1780909 0.10810417 -0.11891362 0.15357238 -0.21714506
-0.20010895]
[-0.03719392 0.17145976 -0.18463153 -0.07408771 -0.16771841
0.17747995]
[ 0.22013861 0.16776392 -0.13306917 -0.04167722 -0.17991196
0.25754595]
[-0.11848497 -0.05116902 0.09591931 0.01752862 0.10756946
-0.04643519]
[ 0.25510186 -0.12553987 -0.24126157 0.21673346 0.14857164
0.25489032]
[ 0.21247488 0.22724444 0.04058126 -0.01098627 0.1850268
0.04928353]
[ 0.14206117 0.11317515 -0.05528121 -0.05800201 0.2836451
-0.11509615]]]

tensorflow基础知识（CNN）相关推荐

3.1 Tensorflow基础知识
3.1 Tensorflow基础知识 3.1.1 深度学习框架介绍框架对比: 1.技术层面对于个人开发者,它的主要用途为科研,而企业用户的主要用途为进行AI应用,提升生产效率.在此,主要从两类用户 ...
教程▍一步步上手TensorFlow——基础知识
作者|blackblog 编辑| 布袋熊 AI基础技能学习一步步上手TensorFlow TensorFlow 是一个用于人工智能的开源神器.作为常用的机器学习框架,可被用于语音识别或图像识别等多 ...
【深度学习】基础知识--CNN:图像分类（上）
作者信息: 华校专,曾任阿里巴巴资深算法工程师.智易科技首席算法研究员,现任腾讯高级研究员,<Python 大战机器学习>的作者. 编者按: 算法工程师必备系列更新啦!继上次推出了算法工程 ...
机器学习Tensorflow基础知识、张量与变量
TensorFlow是一个采用数据流图(data flow graphs),用于数值计算的开源软件库.节点(Nodes)在图中表示数学操作,图中的线(edges)则表示在节点间相互联系的多维数据数组, ...
Tensorflow基础知识
1. tf.expand_dims() 作用:给函数增加纬度. 参数: tf.expand_dims( input, # 输入张量 axis=None, # 给定张量输入input,此操作为选择维度索 ...
tensorflow基础知识10，三维可视化embedding
import tensorflow as tf from tensorflow.examples.tutorials.mnist import input_data from tensorflow.c ...
低阶 TensorFlow 基础知识
(1) Hello world https://colab.research.google.com/notebooks/mlcc/hello_world.ipynb?hl=zh-cn (2) 演示基本 ...
python 基础知识之（numpy、pandas、matplotlib、tensorflow）
1.Numpy库: 参考网页 1.1创建Numpy变量创建一个 ndarray 只需调用 NumPy 的 array 函数即可:numpy.array(object, dtype = None, c ...
CV：计算机视觉技术之图像基础知识(二)—以python的skimage和numpy库来了解计算机视觉图像基础(图像存储原理-模糊核-锐化核-边缘检测核，进阶卷积神经网络(CNN)的必备基础)
CV:计算机视觉技术之图像基础知识(二)-以python的skimage和numpy库来了解计算机视觉图像基础(图像存储原理-模糊核-锐化核-边缘检测核,进阶卷积神经网络(CNN)的必备基础) 目录 ...

tensorflow基础知识（CNN）

词向量映射

tensorflow基础知识（CNN）相关推荐

最新文章

热门文章