Skip to content

P
PaddleRec

BaiXuePrincess / PaddleRec
与 Fork 源项目一致

Fork自 PaddlePaddle / PaddleRec

通知 1
Star 0
Fork 0
P
PaddleRec

体验新版 GitCode，发现更多精彩内容 >>

提交 13e62b7f 编写于 作者: X xujiaqi01
浏览文件
操作

fix

上级 726c61ca
显示空白变更内容
内联 并排
Showing with 224 addition and 217 deletion
models/contentunderstanding/classification/config.yaml
浏览文件 @ 13e62b7f
......@@ -18,7 +18,7 @@ train:
strategy: "async"
epochs: 10
workspace: "paddlerec.models.contentunderstandin.classification"
workspace: "paddlerec.models.contentunderstanding.classification"
reader:
batch_size: 5
......
models/contentunderstanding/classification/model.py
浏览文件 @ 13e62b7f
......@@ -26,9 +26,12 @@ class Model(ModelBase):
data = fluid.data(name="input", shape=[None, self.max_len], dtype='int64')
label = fluid.data(name="label", shape=[None, 1], dtype='int64')
seq_len = fluid.data(name="seq_len", shape=[None], dtype='int64')
self._data_var = [data, label, seq_len]
# embedding layer
emb = fluid.embedding(input=data, size=[self.dict_dim, self.emb_dim])
emb = fluid.layers.sequence_unpad(emb, length=self.seq_len)
emb = fluid.layers.sequence_unpad(emb, length=seq_len)
# convolution layer
conv = fluid.nets.sequence_conv_pool(
input=emb,
......@@ -38,7 +41,7 @@ class Model(ModelBase):
pool_type="max")
# full connect layer
fc_1 = fluid.layers.fc(input=[conv], size=hid_dim)
fc_1 = fluid.layers.fc(input=[conv], size=self.hid_dim)
# softmax layer
prediction = fluid.layers.fc(input=[fc_1], size=self.class_dim, act="softmax")
cost = fluid.layers.cross_entropy(input=prediction, label=label)
......@@ -46,18 +49,18 @@ class Model(ModelBase):
acc = fluid.layers.accuracy(input=prediction, label=label)
self.cost = avg_cost
self.metrics["acc"] = cos_pos
self._metrics["acc"] = acc
def get_cost_op(self):
return self.cost
def get_metrics(self):
return self.metrics
return self._metrics
def optimizer(self):
learning_rate = 0.01
sgd_optimizer = fluid.optimizer.Adagrad(learning_rate=learning_rate)
return sgd_optimizer
def infer_net(self, parameter_list):
def infer_net(self):
self.train_net()
models/contentunderstanding/classification/reader.py
浏览文件 @ 13e62b7f
......@@ -30,5 +30,9 @@ class TrainReader(Reader):
if data is None:
yield None
return
data = [int(i) for i in data]
label = [int(i) for i in label]
seq_len = [int(i) for i in seq_len]
print >>sys.stderr, str([('data', data), ('label', label), ('seq_len', seq_len)])
yield [('data', data), ('label', label), ('seq_len', seq_len)]
return data_iter
models/contentunderstanding/classification/train_data/part-0
浏览文件 @ 13e62b7f
12 27 13 0 25 52 89 20 39 4 9 1
78 10 61 58 29 79 85 16 46 41 9 1
81 77 44 4 5 57 43 97 42 89 6 0
7 77 86 3 98 89 56 24 7 59 9 1
7 77 86 3 98 89 56 24 7 59 9 1
65 89 99 27 65 98 16 89 42 0 3 0
66 14 48 38 66 5 56 89 98 19 4 1
78 7 10 20 77 16 37 43 59 23 6 1
......@@ -24,32 +24,32 @@
97 69 76 47 80 62 23 30 87 22 7 1
42 56 25 47 42 18 80 53 15 57 7 0
34 73 75 88 61 79 40 74 87 87 6 1
7 91 9 24 42 60 76 31 10 13 4 0
7 91 9 24 42 60 76 31 10 13 4 0
21 1 46 59 61 54 99 54 89 55 5 1
67 21 1 29 88 5 3 85 39 22 5 1
90 99 7 8 17 77 73 3 32 10 5 0
30 44 26 32 37 74 90 71 42 29 9 1
79 68 3 24 21 37 35 3 76 23 6 1
3 66 7 4 2 88 94 64 47 81 6 1
3 66 7 4 2 88 94 64 47 81 6 1
10 48 16 49 96 93 61 97 84 39 3 1
73 28 67 59 89 92 17 24 52 71 3 1
98 4 35 62 91 2 78 51 72 93 1 1
37 42 96 10 48 49 84 45 59 47 5 1
13 24 7 49 63 78 29 75 45 92 7 1
1 6 95 23 38 34 85 94 33 47 6 1
1 6 95 23 38 34 85 94 33 47 6 1
99 63 65 39 72 73 91 20 16 45 9 0
35 8 81 24 62 0 95 0 52 46 4 1
58 66 88 42 86 94 91 8 18 92 7 0
12 62 56 43 99 31 63 80 11 7 4 1
22 36 1 39 69 20 56 75 17 15 7 0
25 97 62 50 99 98 32 2 98 75 7 1
7 59 98 68 62 19 28 28 60 27 7 0
7 59 98 68 62 19 28 28 60 27 7 0
39 63 43 45 43 11 40 81 4 25 6 0
81 95 27 84 71 45 87 65 40 50 1 0
82 21 69 55 71 92 52 65 90 16 3 0
24 6 5 22 36 34 66 71 3 52 2 0
5 14 66 71 49 10 52 81 32 14 1 0
8 94 52 23 60 27 43 19 89 91 9 0
5 14 66 71 49 10 52 81 32 14 1 0
8 94 52 23 60 27 43 19 89 91 9 0
26 14 36 37 28 94 46 96 11 80 8 1
89 19 77 66 48 75 62 58 90 81 8 1
25 43 95 21 25 81 39 79 9 74 9 0
......@@ -75,9 +75,9 @@
70 29 79 29 44 56 33 27 25 4 3 1
44 20 87 67 65 41 93 37 99 78 1 1
93 57 87 11 33 40 21 3 47 87 9 1
8 3 24 49 99 48 40 22 99 41 2 0
8 3 24 49 99 48 40 22 99 41 2 0
19 90 9 83 93 22 36 96 44 73 7 1
4 73 2 88 79 90 32 48 45 12 5 0
4 73 2 88 79 90 32 48 45 12 5 0
24 58 34 67 85 62 84 48 14 79 5 1
54 69 19 18 59 78 84 48 61 46 4 0
72 69 95 26 30 74 49 30 95 61 8 0
......@@ -117,7 +117,7 @@
23 27 98 73 25 7 89 48 7 44 4 1
86 98 68 1 74 46 15 92 59 25 9 1
95 86 72 13 33 60 62 83 96 84 1 0
9 58 37 50 57 16 78 0 21 80 2 0
9 58 37 50 57 16 78 0 21 80 2 0
82 94 74 42 3 60 61 93 34 22 3 1
16 97 97 14 47 50 90 35 9 58 5 0
70 94 82 42 85 88 59 58 6 68 9 0
......@@ -125,7 +125,7 @@
22 23 7 82 39 28 96 92 23 40 5 1
40 31 72 94 20 81 89 4 42 1 5 0
57 63 71 41 28 2 39 67 90 54 6 0
9 74 4 41 11 31 15 21 44 32 6 1
9 74 4 41 11 31 15 21 44 32 6 1
31 28 66 66 61 78 72 80 82 88 3 1
79 18 1 59 35 62 0 72 78 97 7 0
14 19 30 63 38 37 12 15 54 15 6 1
......@@ -135,7 +135,7 @@
51 55 9 9 88 59 21 66 87 12 1 1
90 22 38 66 12 9 30 48 55 85 1 1
39 23 82 29 57 76 79 56 3 19 2 0
7 72 76 15 90 23 40 40 33 39 4 1
7 72 76 15 90 23 40 40 33 39 4 1
60 64 34 11 18 18 38 39 53 37 1 1
85 72 51 47 83 90 32 96 78 23 9 1
85 51 96 31 83 70 57 65 15 0 6 0
......@@ -146,10 +146,10 @@
34 19 68 93 11 9 14 87 22 70 9 0
63 77 27 20 20 37 65 51 29 29 9 1
22 79 98 57 56 97 43 49 4 80 4 1
6 4 35 54 4 36 1 79 85 35 6 0
6 4 35 54 4 36 1 79 85 35 6 0
12 55 68 61 91 43 49 5 93 27 8 0
64 22 69 16 63 20 28 60 13 35 7 1
9 19 60 89 62 29 47 33 6 13 4 0
9 19 60 89 62 29 47 33 6 13 4 0
14 15 39 86 47 75 7 70 57 60 6 1
90 63 12 43 28 46 39 97 83 42 6 0
49 3 3 64 59 46 30 13 61 10 2 0
......@@ -160,7 +160,7 @@
88 57 22 64 93 66 20 90 78 2 7 1
90 86 41 28 14 25 86 73 7 21 4 0
63 91 0 29 2 78 86 76 9 20 4 1
3 57 91 37 21 85 80 99 18 79 1 1
3 57 91 37 21 85 80 99 18 79 1 1
69 95 36 6 85 47 83 83 61 52 4 0
72 4 34 16 59 78 56 70 27 44 9 1
58 42 6 53 21 7 83 38 86 66 5 0
......@@ -179,17 +179,17 @@
50 66 99 67 76 25 43 12 24 67 9 0
74 56 61 97 23 63 22 63 6 83 2 1
10 96 13 49 43 20 58 19 99 58 7 1
2 95 31 4 99 91 27 90 85 32 3 0
2 95 31 4 99 91 27 90 85 32 3 0
41 23 20 71 41 75 75 35 16 12 3 1
21 33 87 57 19 27 94 36 80 10 6 0
8 0 25 74 14 61 86 8 42 82 9 0
8 0 25 74 14 61 86 8 42 82 9 0
23 33 91 19 84 99 95 92 29 31 8 0
94 94 5 6 98 23 37 65 14 25 6 1
42 16 39 32 2 20 86 81 90 91 8 0
72 39 20 63 88 52 65 81 77 96 4 0
48 73 65 75 89 36 75 36 11 35 8 0
79 74 3 29 63 20 76 46 8 82 5 0
7 46 38 77 79 92 71 98 30 35 6 0
7 46 38 77 79 92 71 98 30 35 6 0
44 69 93 31 22 68 91 70 32 86 5 0
45 38 77 87 64 44 69 19 28 82 9 0
93 63 92 84 22 44 51 94 4 99 9 0
......@@ -202,27 +202,27 @@
97 12 87 82 6 18 57 49 49 58 1 1
70 79 55 86 29 88 55 39 17 74 5 1
65 51 45 62 54 17 59 12 29 79 5 0
5 63 82 51 54 97 54 36 57 46 3 0
5 63 82 51 54 97 54 36 57 46 3 0
74 77 52 10 12 9 34 95 2 0 5 0
50 20 22 89 50 70 55 98 80 50 1 0
61 80 7 3 78 36 44 37 90 18 9 0
81 13 55 57 88 81 66 55 18 34 2 1
52 30 54 70 28 56 48 82 67 20 8 1
0 41 15 63 27 90 12 16 56 79 3 0
0 41 15 63 27 90 12 16 56 79 3 0
69 89 54 1 93 10 15 2 25 59 8 0
74 99 17 93 96 82 38 77 98 85 4 0
8 59 17 92 60 21 59 76 55 73 2 1
8 59 17 92 60 21 59 76 55 73 2 1
53 56 79 19 29 94 86 96 62 39 3 1
23 44 25 63 41 94 65 10 8 40 9 1
7 18 80 43 20 70 14 59 72 17 9 0
7 18 80 43 20 70 14 59 72 17 9 0
84 97 79 14 37 64 23 68 8 24 2 0
63 94 98 77 8 62 10 77 63 56 4 0
8 63 74 34 49 22 52 54 44 93 3 0
8 63 74 34 49 22 52 54 44 93 3 0
94 48 92 58 82 48 53 34 96 25 2 0
33 15 3 95 48 93 9 69 44 77 7 1
69 72 80 77 64 24 52 21 36 49 2 0
59 34 54 66 60 19 76 79 16 70 5 1
8 83 9 91 67 79 31 20 31 88 2 0
8 83 9 91 67 79 31 20 31 88 2 0
64 95 46 95 78 63 4 60 66 63 7 1
10 39 78 45 36 4 89 94 68 75 7 0
81 52 70 11 48 15 40 63 29 14 8 1
......@@ -231,8 +231,8 @@
71 50 79 54 17 58 30 16 17 99 1 1
74 79 74 61 61 36 28 39 89 36 6 0
53 45 45 23 51 32 93 26 10 8 3 0
1 97 6 67 88 20 41 63 49 6 8 0
3 64 41 19 41 80 75 71 69 90 8 0
1 97 6 67 88 20 41 63 49 6 8 0
3 64 41 19 41 80 75 71 69 90 8 0
31 90 38 93 52 0 38 86 41 68 9 1
50 94 53 9 73 59 94 7 24 57 3 0
87 11 4 62 96 7 0 59 46 11 6 1
......@@ -272,14 +272,14 @@
70 40 38 47 5 38 83 70 37 90 2 0
42 21 62 27 43 47 82 80 88 49 4 0
68 68 67 12 38 13 32 30 93 27 3 1
5 44 98 28 5 81 20 56 10 34 9 1
5 44 98 28 5 81 20 56 10 34 9 1
40 46 11 33 73 62 68 70 66 85 4 0
9 46 11 84 6 31 18 89 66 32 1 1
6 78 44 98 77 29 69 39 62 78 1 0
9 46 11 84 6 31 18 89 66 32 1 1
6 78 44 98 77 29 69 39 62 78 1 0
47 90 18 0 3 8 12 20 51 75 4 1
21 29 74 19 12 29 41 22 63 47 8 1
22 59 64 62 18 89 19 92 87 8 8 0
6 21 24 58 14 53 18 93 62 15 8 0
6 21 24 58 14 53 18 93 62 15 8 0
20 33 88 25 37 52 1 72 74 11 2 0
90 49 28 53 28 80 22 81 0 46 9 0
87 31 51 27 15 31 68 93 5 4 7 1
......@@ -314,7 +314,7 @@
65 74 40 2 8 40 18 57 30 38 1 1
76 44 64 6 10 32 84 70 74 24 1 1
14 29 59 34 27 8 0 37 27 68 3 0
6 47 5 77 15 41 93 49 59 83 4 1
6 47 5 77 15 41 93 49 59 83 4 1
39 88 43 89 32 98 82 0 5 12 9 0
78 79 30 26 58 6 9 58 37 65 8 1
25 28 66 41 70 87 76 62 29 39 7 1
models/contentunderstanding/readme.md
浏览文件 @ 13e62b7f
......@@ -71,13 +71,13 @@ python text2paddle.py raw_big_train_data/ raw_big_test_data/ train_big_data test
### 训练
```
python -m paddlerec.run -m paddlerec.models.rank.dnn -d cpu -e single
python -m paddlerec.run -m paddlerec.models.contentunderstanding.classification -d cpu -e single
```
### 预测
```
python -m paddlerec.run -m paddlerec.models.rank.dnn -d cpu -e single
python -m paddlerec.run -m paddlerec.models.contentunderstanding.classification -d cpu -e single
```
## 效果对比
......
models/contentunderstanding/tagspace/config.yaml
浏览文件 @ 13e62b7f
......@@ -18,7 +18,7 @@ train:
strategy: "async"
epochs: 10
workspace: "paddlerec.models.rank.tagspace"
workspace: "paddlerec.models.contentunderstanding.tagspace"
reader:
batch_size: 5
......
Markdown is supported
0% 或 .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑！
想要评论请 注册 或