-
A.1 使用:gunicorn+flask+service-streamer（flask下启动了3个模型，一个端口(例如8804)，3个app）
batch_size=64
A.2 说明:
1. 其中配置gunicorn+gevent启动报错(报错如C.报错信息）
2. 其中配置gunicorn+tornado启动报错(报错如C.报错信息）
3. uwsgi启动不…
-
在一些任务上，直接跑capsnet，相比于textcnn效果会差一些，考虑到background-noise的影响，您提出了3种策略，包括Orphan Category，Leaky-Softmax，和Coefficients Amendment。代码中好像只有Coefficients Amendment部分代码。请问其他两种方法的代码您还会更新上来吗？
-
My aim is to make a five-category text classification
I am running transformers fine tuning bert with `cnnbase` model but my program stops at `loss.backward()` without any prompt in `cmd`.
I de…
-
First, thanks for the contribution of this great tensorboardX.
Second, I try to visualize the network structure graph of a text classification--TextCNN in NLP. The beginning of the network is the emb…
-
Hello. I have stumbled upon the issue with code cells rendering.
**Describe the bug**
When I render one of my github notebooks, code cells are empty. It works well when I render previous notebooks…
-
这个模型是Yoon Kim, 2014的经典模型，我观察到docstring和签名中有这样几个问题：
- [ ] 参数列表：['init_embed', 'num_classes', '**kernel_nums**=(3, 4, 5)', 'kernel_sizes=(3, 4, 5)', 'padding=0', 'dropout=0.5']，这里的`kernel_nums`在下面的docs…
-
```
ImportError Traceback (most recent call last)
in
6 from sklearn.preprocessing import MultiLabelBinarizer
7 from sklearn.metrics import roc_auc_score…
-
使用BertEmbedding
graph的create_model方法
最下面，向word_embedding的model加载权重
# 构建网络层
self.word_embedding = Embeddings(hyper_parameters=hyper_parameters)
if os.path.exists(self.path_fi…
-
---------------------------------------------------------------------------
ImportError Traceback (most recent call last)
in
6 from sklearn.preprocessing impor…
-
why ksize of maxpooling of textcnn is [1, self.sentence_size - filter_size + 1, 1, 1]，what I mind is "sentence_size - filter_size + 1", can it be sentence_size, filter_size or any others, is there any…