Giters
modern-fortran
/
neural-fortran
A parallel framework for deep learning
Geek Repo:
Geek Repo
Github PK Tool:
Github PK Tool
Stargazers:
382
Watchers:
27
Issues:
88
Forks:
81
modern-fortran/neural-fortran Issues
Support of a method `network % evaluate`
Updated
a month ago
Comments count
5
Building client code with CMake
Closed
2 months ago
Comments count
6
Add topics `deep-learning`, `cnn`
Closed
2 months ago
Comments count
1
Introduce a separate `network % compile()` step
Updated
2 months ago
CNN training on MNIST does not converge
Updated
2 months ago
Comments count
3
Implement locally connected layer (1-d)
Updated
3 months ago
Implement dropout layer
Updated
3 months ago
Start-to-finish example
Updated
4 months ago
Comments count
1
Test failure with ifx
Updated
5 months ago
Comments count
6
Supported HDF5 version
Updated
5 months ago
Comments count
3
Some tests and examples fail with segmentation fault in serial production build, but not in debug build
Updated
6 months ago
Comments count
1
scaling
Updated
7 months ago
Comments count
2
Refactor `forward` and `backward` methods to allow passing a batch of data instead of one sample at a time
Updated
10 months ago
Implement `batchnorm` layer
Updated
10 months ago
Question about the decoupled weight decay in Adam
Closed
10 months ago
Comments count
2
could it be used for subrountine in FEM softwore ABAQUS
Updated
a year ago
Comments count
3
Support for neural network exchange format (NNEF)
Updated
a year ago
Comments count
3
Add batch normalization and Adam optimizer
Updated
a year ago
Implement Adam optimizer
Closed
a year ago
Comments count
6
Is CoArray that optional ?
Closed
a year ago
Comments count
2
Implement momentum and Nesterov modifications in SGD in example/quadratic.f90
Closed
a year ago
Comments count
1
v0.13.0 release
Closed
a year ago
Comments count
5
Error while building in latest main
Closed
a year ago
Comments count
1
Implement RMSprop in example/quadratic.f90
Closed
a year ago
Comments count
3
Question about input data to the quadratic function in example/quadratic.f90
Closed
a year ago
Comments count
2
Add example program to fit a quadratic function using different optimizers
Closed
a year ago
Comments count
1
Implement deconvolution / transpose convolution layer
Updated
a year ago
UNET implementation support
Updated
a year ago
Comments count
2
Implement upsampling layer type
Updated
a year ago
Create a contributing guide
Closed
a year ago
Possible to build without internet access aka pre-built dependencies?
Updated
a year ago
Comments count
9
Implement activation params type
Closed
a year ago
Comments count
5
MNIST example via PNG
Updated
a year ago
Comments count
1
Implement read-only forward pass
Closed
a year ago
Comments count
1
get_set_network_params.f90 does not print the output of a logical test
Closed
a year ago
Comments count
1
extract / set all network parameters via a single 1D real array
Closed
a year ago
Comments count
2
broken cmake compilation
Closed
2 years ago
Comments count
2
Add support for convolutional layers
Closed
2 years ago
Implement maxpool2d backward pass
Closed
2 years ago
Add a MNIST training example using conv2d layers
Closed
2 years ago
conv2d backward pass
Closed
2 years ago
Compiling on Windows 10
Updated
2 years ago
Comments count
7
Implement a reshape layer
Closed
2 years ago
Add GitHub Actions to run the tests on all pushes and pull requests on Linux and macOS
Closed
2 years ago
Memory leak in get_keras_h5_layers
Closed
2 years ago
Comments count
1
Initialize random number generation for consistent behavior between compilers
Updated
2 years ago
Implement a flatten layer
Closed
2 years ago
Use the directory structure to more obviously expose the user API
Closed
2 years ago
Comments count
2
Protect the main branch
Closed
2 years ago
Comments count
1
Implement a max-pooling layer
Closed
2 years ago
Previous
Next