Home
Data Process
Genome
Transcriptome
Protein
Integration
Model
Train > supervised
Train > unsupervised
Train > semi-supervised
Use
Data Analyse
Help
``` ```
Train your supervised model
Home
Model
Train
1. Select strategy and upload files
Upload train/validation/testing datasets
Upload a dataset to conduct train/validation/testing split
Use stratified K-Fold cross validation method
Upload training set (Only datasets labeled with categories 0/1/2 to 6 can be uploaded)
Example
Upload validation set
Upload testing set
Upload a dataset (Only datasets labeled with categories 0/1/2 to 6 can be uploaded)
Example
Input a ratio of training:validation:testing (NOTE: The separator is a colon ':')
Upload training set
Example
Input K value (NOTE: If you want to use 3-Fold cross validation method, then K value is 3.)
Upload testing set
2. Choose model
CNN > CNN Excels at extracting discriminative features from 1D signals with local patterns (e.g., mass spectrometry)
LSTM > Designed for modeling data with long-term dependencies
RNN > Used for processing simple sequences with short-term dependencies
MLP > A general-purpose benchmark model for non-linear classification of feature vectors without explicit structure
AutoEncoder > Generates better feature representations for classification tasks through dimensionality reduction and denoising
Transformer > Models the global context via the self-attention mechanism
SOM > Suitable for discovering inherent cluster structures within data
RBFNN > Suitable for fast learning and local approximation of small-scale tabular data
All > All the above models are trained in parallel using the same set of data
3. Upload training hyperparameters
(1) Epoch
(2) Learning Rate
(3) EarlyStopping Patience
(4) Batch Size
(5) Label Number (NOTE: If you want to use multi-label classification, then the number of labels is greater than 2. Upper limit of 7.)
(6) Loss Function
CrossEntropyLoss
FocalLoss
NLLLoss
(7) Optimizer
Adam
SGD
RMSprop
(8) Random Seed
Your Email
Submit