my code wont get me good results

$10-30 USD

Closed

Posted

about 4 years ago

$10-30 USD

Paid on delivery

the tags are pre-processed (tokenization, stop word removal...) with nltk. After pre-processing, it must be mapped to a predefined list of expert labels (94 labels) by using a word embedding technique called FastText. I want to make use of the pre-trained functionality of FastText. I successfully installed the FastText model and successfully make it work but the results are unsatisfying. for example, the same prediction is assigned to all tag files.

Python

Project ID: 23933629

About the project

9 proposals

Remote project

Active 4 yrs ago

Looking to make some money?

Email address

Benefits of bidding on Freelancer

Set your budget and timeframe

Get paid for your work

Outline your proposal

It's free to sign up and bid on jobs

9 freelancers are bidding on average $28 USD for this job

@Techiedev

Hi I have good experiences developing code in Python Programming language for scientific application and Machine learning projects more than 3 years . In recent 2 years i worked with different deep learning Frameworks specially Keras , TensorFlow and Pytorch . As I know FastText created by Facebook and includes pre-trained model and embedding for different language and also make some NLP tasks easier. What language do you use? Arabic or English? And what is your NLP task? is label classification based on set of tags? could you share your codes? Thanks

$35 USD in 3 days

4.9

(27 reviews)

4.7

@muzammil360

Hi. It sounds like you need help with NLP. I am a machine learning engineer with more than 4 years of experience in Python, C++ and Matlab. Your project description is a bit unclear to me. You can always chat with me and explain your issue in detail. We will see what can we do about the challenges yourare facing.

$50 USD in 15 days

5.0

(5 reviews)

3.8

@trenzenokvs

Hi, I’m experienced in Python and finished numerous projects in NLP. So, I will easily help you. Looking forward for your reply.

$25 USD in 5 days

5.0

(4 reviews)

3.0

@kylezhaoxc

Hi, I have done a similar text classification task at Microsoft with tf+Keras, Conv1D, as well as FastText. Per your project description, your model input is like (x = tags, y = expert labels), and you are training a model predicting expertlabels, right? Though the model structure and optimization method is still not clear to me, I think I could give helpful suggestions. Hope we could solve it together.

$30 USD in 7 days