https://github.com/monk1337/multilab
State of art Framework for Multi-label classification
https://github.com/monk1337/multilab
classification deep-learning keras machine-learning multi-label-classification tensorflow tensorflow-models tensorflow-tutorials text-classification text-processing
Last synced: 24 days ago
JSON representation
State of art Framework for Multi-label classification
- Host: GitHub
- URL: https://github.com/monk1337/multilab
- Owner: monk1337
- Created: 2019-11-26T12:19:06.000Z (almost 6 years ago)
- Default Branch: master
- Last Pushed: 2019-12-02T06:56:51.000Z (almost 6 years ago)
- Last Synced: 2025-01-31T03:01:35.154Z (10 months ago)
- Topics: classification, deep-learning, keras, machine-learning, multi-label-classification, tensorflow, tensorflow-models, tensorflow-tutorials, text-classification, text-processing
- Language: Jupyter Notebook
- Homepage:
- Size: 402 KB
- Stars: 5
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
MultiLab
A framework for Multi-label classification, Classical machine learning models to state of art deep learning models for multi label classification, preprocessing Multi-label datasets and load benchmark multi-label datasets, metrices for accuracy calculation
You can preprocess multilabel dataset simply as follows:
```python
from multilab.preprocess import Text_preprocessing
tp = Text_preprocessing()
preprocessded_dataset = tp.initial_preprocess(dataframe, chunk_value = 5)
```
### Loading traditional Models
```python
from multilab.models import BinaryRe
Bm = BinaryRe(x_train, y_train, x_test,y_test)
print(Bm.train())
```
output :
```python
{'accuracy': 0.4074074074074074, 'f1_score': 0.4395604395604396}
```
### Loading Bilstm model
```python
from multilab.models import Bilstm
config = {
'vocab_size' : 7000,
'no_of_labels' : 9,
'rnn_units' : 256,
'word_embedding_dim' : 300,
'learning_rate' : 0.001,
'pretrained_embedding_matrix': None,
'dropout' : 0.2,
'epoch' : 5,
'batch_size' : 128,
'result_path' : '/Users/aaditya/Desktop',
'last_output' : False,
'train_embedding' : True
}
bl = Bilstm(x_train, y_train, x_test,y_test, config)
bl.train()
```
output
```python
validation_acc {'subset_accuracy': 0.45166666666666666, 'hamming_score': 0.4601111111111112, 'hamming_loss': 0.1285185185185185, 'micro_ac': 0.4490395710185522, 'weight_ac': 0.2830056188426279, 'epoch': 0}
```
### Loading Elmo model
```python
from multilab.models import Elmo
config = {
'no_of_labels' : 9,
'learning_rate' : 0.001,
'epoch' : 5,
'batch_size' : 128,
'model_type' : 'base',
'result_path' : '.'
}
elmo_model = Elmo(x_train, y_train, x_test,y_test, config)
elmo_model.train()
```
output
```python
validation_acc {'subset_accuracy': 0.5966666666666666, 'hamming_score': 0.6, 'hamming_loss': 0.05907407407407408, 'micro_ac': 0.6943641132818982, 'weight_ac': 0.5731481624223015, 'epoch': 0}
```
##### adding more models, work in progress..