Searched refs:weights (Results 1 – 5 of 5) sorted by relevance
| /honee/tests/createPyTorchModel/ |
| H A D | update_weights.py | 10 weights = [] variable
12 weights.append(np.loadtxt(new_parameters_Path / 'w1.dat', skiprows=1).reshape(6, 20).T)
13 weights.append(np.loadtxt(new_parameters_Path / 'w2.dat', skiprows=1).reshape(20, 6).T)
50 model.net[layer].weight[...] = torch.from_numpy(weights[i])[...]
|
| H A D | README.md | 1 This directory exists to create a PyTorch model with certain weights and biases. It is mostly setup…
|
| /honee/src/ |
| H A D | dm-utils.c | 298 PetscReal *coords, *weights, *coords_1d; in PetscDTUniformTensorQuadrature() local
302 PetscCall(PetscMalloc1(num_total_points * num_comp, &weights)); in PetscDTUniformTensorQuadrature()
308 PetscCall(PetscFree(weights)); in PetscDTUniformTensorQuadrature()
310 PetscCall(PetscMalloc1(num_comp, &weights)); in PetscDTUniformTensorQuadrature()
313 for (PetscInt c = 0; c < num_comp; c++) weights[c] = 1.0; in PetscDTUniformTensorQuadrature()
321 for (PetscInt c = 0; c < num_comp; c++) weights[i * num_comp + c] = 1.0; in PetscDTUniformTensorQuadrature()
334 … for (PetscInt c = 0; c < num_comp; c++) weights[(i * num_points + j) * num_comp + c] = 1.0; in PetscDTUniformTensorQuadrature()
351 …for (PetscInt c = 0; c < num_comp; c++) weights[((i * num_points + j) * num_points + k) * num_comp… in PetscDTUniformTensorQuadrature()
365 PetscCall(PetscQuadratureSetData(*q, dim, num_comp, num_total_points, coords, weights)); in PetscDTUniformTensorQuadrature()
|
| /honee/doc/ |
| H A D | theory.md | 439 These files store the network weights (`w1.dat` and `w2.dat`), biases (`b1.dat` and `b2.dat`), and …
482 - Path to directory with data-driven model parameters (weights, biases, etc.)
|
| H A D | auxiliary.md | 59 This also allows the use of the full domain quadrature weights for the triple integral.
|