Merge branch 'develop' of https://github.com/PaddlePaddle/models into yxp1216

PaddleNLP/docs/embeddings.md

@@ -31,14 +31,14 @@ PaddleNLP提供多个开源的预训练Embedding模型，用户仅需在使用`p
 | Co-occurrence 类型          | 目标词向量 | 上下文词向量  |
 | --------------------------- | ------   | ---- |
 |    Word → Word              | w2v.baidu_encyclopedia.target.word-word.dim300     |   w2v.baidu_encyclopedia.context.word-word.dim300    |
-|    Word → Ngram (1-2)       |  w2v.baidu_encyclopedia.target.word-ngram.1-2.dim300    |   暂无    |
-|    Word → Ngram (1-3)       |  暂无    |   暂无    |
-|    Ngram (1-2) → Ngram (1-2)|  暂无   |   暂无    |
+|    Word → Ngram (1-2)       |  w2v.baidu_encyclopedia.target.word-ngram.1-2.dim300    |   w2v.baidu_encyclopedia.context.word-ngram.1-2.dim300    |
+|    Word → Ngram (1-3)       |  w2v.baidu_encyclopedia.target.word-ngram.1-3.dim300    |   w2v.baidu_encyclopedia.context.word-ngram.1-3.dim300    |
+|    Ngram (1-2) → Ngram (1-2)|  w2v.baidu_encyclopedia.target.word-ngram.2-2.dim300   |   w2v.baidu_encyclopedia.target.word-ngram.2-2.dim300    |
 |    Word → Character (1)     |  w2v.baidu_encyclopedia.target.word-character.char1-1.dim300    |  w2v.baidu_encyclopedia.context.word-character.char1-1.dim300     |
 |    Word → Character (1-2)   |  w2v.baidu_encyclopedia.target.word-character.char1-2.dim300    |  w2v.baidu_encyclopedia.context.word-character.char1-2.dim300     |
 |    Word → Character (1-4)   |  w2v.baidu_encyclopedia.target.word-character.char1-4.dim300    |  w2v.baidu_encyclopedia.context.word-character.char1-4.dim300     |
-|    Word → Word (left/right) |   暂无   |   暂无    |
-|    Word → Word (distance)   |   暂无   |   暂无    |
+|    Word → Word (left/right) |   w2v.baidu_encyclopedia.target.word-wordLR.dim300   |   w2v.baidu_encyclopedia.context.word-wordLR.dim300    |
+|    Word → Word (distance)   |   w2v.baidu_encyclopedia.target.word-wordPosition.dim300   |   w2v.baidu_encyclopedia.context.word-wordPosition.dim300    |
 
 ## 英文词向量
 

PaddleNLP/examples/text_generation/vae-seq2seq/README.md

@@ -117,6 +117,7 @@ python predict.py \
         --max_grad_norm 5.0 \
         --dataset yahoo \
         --use_gpu True \
+        --infer_output_file infer_output.txt \
         --init_from_ckpt yahoo_model/49 \
 
 ```

PaddleNLP/examples/word_embedding/README.md

@@ -58,7 +58,7 @@ nohup python train.py --vocab_path='./dict.txt' --use_gpu=True --lr=1e-4 --batch
 以上参数表示：
 
 * `vocab_path`: 词汇表文件路径。
-* `use_gpu`: 是否使用GPU进行训练， 默认为`False`。
+* `use_gpu`: 是否使用GPU进行训练， 默认为`True`。
 * `lr`: 学习率， 默认为5e-4。
 * `batch_size`: 运行一个batch大小，默认为64。
 * `epochs`: 训练轮次，默认为5。
@@ -96,3 +96,7 @@ Eval Acc：
 
 ## 致谢
 - 感谢 [Chinese-Word-Vectors](https://github.com/Embedding/Chinese-Word-Vectors)提供Word2Vec中文Embedding来源。
+
+## 参考论文
+- Li, Shen, et al. "Analogical reasoning on chinese morphological and semantic relations." arXiv preprint arXiv:1805.06504 (2018).
+- Qiu, Yuanyuan, et al. "Revisiting correlations between intrinsic and extrinsic evaluations of word embeddings." Chinese Computational Linguistics and Natural Language Processing Based on Naturally Annotated Big Data. Springer, Cham, 2018. 209-221.

PaddleNLP/paddlenlp/embeddings/README.md

@@ -25,20 +25,21 @@ token_embedding = TokenEmbedding(embedding_name="w2v.baidu_encyclopedia.target.w
 # 查看token_embedding详情
 print(token_embedding)
 
-Object   type: <paddlenlp.embeddings.token_embedding.TokenEmbedding object at 0x7f67fd192e30>
-Unknown index: 1
+Object   type: <paddlenlp.embeddings.token_embedding.TokenEmbedding object at 0x7fda7eb5f290>
+Unknown index: 635963
 Unknown token: [UNK]
-Padding index: 0
+Padding index: 635964
 Padding token: [PAD]
 Parameter containing:
-Tensor(shape=[636015, 300], dtype=float32, place=CPUPlace, stop_gradient=False,
-       [[ 0.        ,  0.        ,  0.        , ...,  0.        ,  0.        ,  0.        ],
-        [ 0.00372404,  0.01534354,  0.01341010, ..., -0.00605236, -0.02150303,  0.02372430],
-        [-0.24200200,  0.13931701,  0.07378800, ...,  0.14103900,  0.05592300, -0.08004800],
+Tensor(shape=[635965, 300], dtype=float32, place=CPUPlace, stop_gradient=False,
+       [[-0.24200200,  0.13931701,  0.07378800, ...,  0.14103900,  0.05592300, -0.08004800],
+        [-0.08671700,  0.07770800,  0.09515300, ...,  0.11196400,  0.03082200, -0.12893000],
+        [-0.11436500,  0.12201900,  0.02833000, ...,  0.11068700,  0.03607300, -0.13763499],
         ...,
-        [ 0.01615800, -0.00266300, -0.00628300, ...,  0.01484100,  0.00196600, -0.01032000],
-        [ 0.01705700,  0.00040400, -0.01222000, ...,  0.02837200,  0.02402500, -0.00814800],
-        [ 0.02628800, -0.00008300, -0.00393500, ...,  0.00654000,  0.00024600, -0.00662600]])
+        [ 0.02628800, -0.00008300, -0.00393500, ...,  0.00654000,  0.00024600, -0.00662600],
+        [-0.00924490,  0.00652097,  0.01049327, ..., -0.01796000,  0.03498908, -0.02209341],
+        [ 0.        ,  0.        ,  0.        , ...,  0.        ,  0.        ,  0.        ]])
+
 ```
 
 ## 查询embedding结果
@@ -93,5 +94,5 @@ words = tokenizer.cut("中国人民")
 print(words) # ['中国人', '民']
 
 tokens = tokenizer.encode("中国人民")
-print(tokens) # [12532, 1336]
+print(tokens) # [12530, 1334]
 ```

PaddleNLP/paddlenlp/embeddings/constant.py

@@ -22,17 +22,27 @@ PAD_TOKEN = '[PAD]'
 UNK_TOKEN = '[UNK]'
 
 EMBEDDING_NAME_LIST = [
+    # Word2Vec
     # baidu_encyclopedia
     "w2v.baidu_encyclopedia.target.word-word.dim300",
     "w2v.baidu_encyclopedia.target.word-character.char1-1.dim300",
     "w2v.baidu_encyclopedia.target.word-character.char1-2.dim300",
     "w2v.baidu_encyclopedia.target.word-character.char1-4.dim300",
     "w2v.baidu_encyclopedia.target.word-ngram.1-2.dim300",
+    "w2v.baidu_encyclopedia.target.word-ngram.1-3.dim300",
+    "w2v.baidu_encyclopedia.target.word-ngram.2-2.dim300",
+    "w2v.baidu_encyclopedia.target.word-wordLR.dim300",
+    "w2v.baidu_encyclopedia.target.word-wordPosition.dim300",
     "w2v.baidu_encyclopedia.target.bigram-char.dim300",
     "w2v.baidu_encyclopedia.context.word-word.dim300",
     "w2v.baidu_encyclopedia.context.word-character.char1-1.dim300",
     "w2v.baidu_encyclopedia.context.word-character.char1-2.dim300",
     "w2v.baidu_encyclopedia.context.word-character.char1-4.dim300",
+    "w2v.baidu_encyclopedia.context.word-ngram.1-2.dim300",
+    "w2v.baidu_encyclopedia.context.word-ngram.1-3.dim300",
+    "w2v.baidu_encyclopedia.context.word-ngram.2-2.dim300",
+    "w2v.baidu_encyclopedia.context.word-wordLR.dim300",
+    "w2v.baidu_encyclopedia.context.word-wordPosition.dim300",
     # wikipedia
     "w2v.wiki.target.bigram-char.dim300",
     "w2v.wiki.target.word-char.dim300",