[英]creating NLC classifier on bluemix
when i try to create a new classifier of NLC, based on this documentation: http://www.ibm.com/smarterplanet/us/en/ibmwatson/developercloud/natural-language-classifier/api/v1/?node#create_classifier 当我尝试根据此文档创建NLC的新分类器时: http : //www.ibm.com/smarterplanet/us/en/ibmwatson/developercloud/natural-language-classifier/api/v1/? node#create_classifier
I can't create a new classifier. 我无法创建新的分类器。 This is the error:
这是错误:
{ code: 400, error: undefined }
I shouldn't receive that kind of error. 我不应该收到这种错误。
The CSV file is very small, just for test. CSV文件非常小,仅供测试。
Maybe i am doing something wrong. 也许我做错了。 But that error: "undefined" doesn't say nothing...
但是那个错误:“未定义”什么也没说...
This is the code (The same as the documentation) 这是代码(与文档相同)
var params = { language: 'es', name: "Train_NODE", training_data: fs.createReadStream('./public/uploads/'+req.files.file.name) }; natural_language_classifier.create(params, function(err, response) { if (err){ console.log(err); res.status(500).render('./APIs/NLC/', { title: err.description }); }else{ console.log(JSON.stringify(response, null, 2)); res.status(200).render('./APIs/NLC/', { title: 'Natural Language Classifer' }); } });
Somebody knows what to do? 有人知道该怎么办吗?
Issue here was with the file path being used. 这里的问题是正在使用的文件路径。 It returned a file that did not exist.
它返回了不存在的文件。 Once the file path to the training.csv was hardcoded in, everything worked fine.
硬编码training.csv的文件路径后,一切工作正常。
So we got the 400 error when we used './public/uploads/'+req.files.file.name, but it worked when we used './public/uploads/MyTrainingFile.csv'. 因此,当我们使用'./public/uploads/'+req.files.file.name时,出现了400错误,但是当我们使用'./public/uploads/MyTrainingFile.csv'时,它可以工作。
We would anticipate some more descriptive error message though. 我们将期待一些更具描述性的错误消息。
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.