Upload a table of training samples

+

Output sample importances (optional)

Predict the values of:
 


Regenerate random data:

  


Architecture options
+
Max layers:
Max neurons:
Max weights:
Include bias
Sparse weights

Activation functions:
Step

ReLU
ReLU1
Firing rate ≤  %
Sigmoid
Tanh

Quantize weights & activations
Bits:
Zero int:
Range:



Training inputs (x, y) → output (color)

+

x: input #
min
max

y: input #
min
max

→ output #
min
max



Instructions: This tool builds neural networks that predict outputs from inputs.

  • Upload a table* of training samples containing both inputs and outputs. These examples will train the network. Specify which rows or columns are the outputs, and click Submit.
  • The trained network opens in a new page, where you can a) predict the outputs of new samples, and b) export it as computer code for use in your own programs.
* Sample table must be a fully numeric, comma-separated .csv file with no empty fields. Max allowed: 100 kB file size, 1000 rows/columns, 1 header row/column. Training time limited to 1 sec.


© 2025 C Deeply, LLC | APIs | Terms | Contact