Skip to content

Commit

Permalink
complete the multivariate OMP example
Browse files Browse the repository at this point in the history
  • Loading branch information
Sylvain Chevallier committed Sep 17, 2015
1 parent 20da152 commit 8c209f6
Showing 1 changed file with 41 additions and 90 deletions.
131 changes: 41 additions & 90 deletions example_sparse_decomposition.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -98,101 +98,52 @@ def decomposition_random_dictionary(Gaussian=True, rng=None, n_features=65,
rmse[k, idx] = norm(r[0], 'fro')/norm(s, 'fro')*100
return rmse

rmse_gaussian = decomposition_random_dictionary(Gaussian=True, rng=rng_global,
rmse_gaussian1 = decomposition_random_dictionary(Gaussian=True, rng=rng_global,
n_features=65, n_dims=1)

rmse_uniform = decomposition_random_dictionary(Gaussian=False, rng=rng_global,
rmse_uniform1 = decomposition_random_dictionary(Gaussian=False, rng=rng_global,
n_features=65, n_dims=1)

fig = plt.figure(figsize=(10,8))
plt.title(r'Initialization of the random univariate dictionary')
km = fig.add_subplot(1,1,1)
km.errorbar(range(1, n_nonzero_coefs+1), rmse_uniform.mean(1),
yerr=rmse_uniform.std(1), label='Uniform')
km.errorbar(range(1, n_nonzero_coefs+1), rmse_gaussian.mean(1),
yerr=rmse_gaussian.std(1), color='r', label='Gaussian')
km.plot(range(n_nonzero_coefs+2), np.zeros(n_nonzero_coefs+2), 'k')
km.axis([0, n_nonzero_coefs+1, 0, 90])
km.set_xticks(range(0, n_nonzero_coefs+2, 5))
km.set_xlabel('k')
km.set_ylabel('rRMSE (%)')
km.legend(loc='upper right')
plt.savefig('sparse_decomposition_univariate.png')

rmse_gaussian = decomposition_random_dictionary(Gaussian=True, rng=rng_global,
rmse_gaussian2 = decomposition_random_dictionary(Gaussian=True, rng=rng_global,
n_features=65, n_dims=3)
rmse_uniform2 = decomposition_random_dictionary(Gaussian=False, rng=rng_global,
n_features=65, n_dims=3)
rmse_gaussian3 = decomposition_random_dictionary(Gaussian=True, rng=rng_global,
n_features=65, n_dims=5)

rmse_uniform = decomposition_random_dictionary(Gaussian=False, rng=rng_global,
rmse_uniform3 = decomposition_random_dictionary(Gaussian=False, rng=rng_global,
n_features=65, n_dims=5)

fig = plt.figure(figsize=(10,8))
plt.title(r'Initialization of the random multivariate dictionary')
km = fig.add_subplot(1,1,1)
km.errorbar(range(1, n_nonzero_coefs+1), rmse_uniform.mean(1),
yerr=rmse_uniform.std(1), label='Uniform')
km.errorbar(range(1, n_nonzero_coefs+1), rmse_gaussian.mean(1),
yerr=rmse_gaussian.std(1), color='r', label='Gaussian')
km.plot(range(n_nonzero_coefs+2), np.zeros(n_nonzero_coefs+2), 'k')
km.axis([0, n_nonzero_coefs+1, 0, 90])
km.set_xticks(range(0, n_nonzero_coefs+2, 5))
km.set_xlabel('k')
km.set_ylabel('rRMSE (%)')
km.legend(loc='upper right')
fig = plt.figure(figsize=(15,5))
uni = fig.add_subplot(1,3,1)
uni.set_title(r'Random univariate (n=1) dictionary')
uni.errorbar(range(1, n_nonzero_coefs+1), rmse_uniform1.mean(1),
yerr=rmse_uniform1.std(1), label='Uniform')
uni.errorbar(range(1, n_nonzero_coefs+1), rmse_gaussian1.mean(1),
yerr=rmse_gaussian1.std(1), color='r', label='Gaussian')
uni.plot(range(n_nonzero_coefs+2), np.zeros(n_nonzero_coefs+2), 'k')
uni.axis([0, n_nonzero_coefs+1, 0, 90])
uni.set_xticks(range(0, n_nonzero_coefs+2, 5))
uni.set_ylabel('rRMSE (%)')
uni.legend(loc='upper right')
mul1 = fig.add_subplot(1,3,2)
mul1.set_title(r'Random multivariate (n=3) dictionary')
mul1.errorbar(range(1, n_nonzero_coefs+1), rmse_uniform2.mean(1),
yerr=rmse_uniform2.std(1), label='Uniform')
mul1.errorbar(range(1, n_nonzero_coefs+1), rmse_gaussian2.mean(1),
yerr=rmse_gaussian2.std(1), color='r', label='Gaussian')
mul1.plot(range(n_nonzero_coefs+2), np.zeros(n_nonzero_coefs+2), 'k')
mul1.axis([0, n_nonzero_coefs+1, 0, 90])
mul1.set_xticks(range(0, n_nonzero_coefs+2, 5))
mul1.set_xlabel('k')
mul1.legend(loc='upper right')
mul2 = fig.add_subplot(1,3,3)
mul2.set_title(r'Random multivariate (n=5) dictionary')
mul2.errorbar(range(1, n_nonzero_coefs+1), rmse_uniform3.mean(1),
yerr=rmse_uniform3.std(1), label='Uniform')
mul2.errorbar(range(1, n_nonzero_coefs+1), rmse_gaussian3.mean(1),
yerr=rmse_gaussian3.std(1), color='r', label='Gaussian')
mul2.plot(range(n_nonzero_coefs+2), np.zeros(n_nonzero_coefs+2), 'k')
mul2.axis([0, n_nonzero_coefs+1, 0, 90])
mul2.set_xticks(range(0, n_nonzero_coefs+2, 5))
mul2.legend(loc='upper right')
plt.savefig('sparse_decomposition_multivariate.png')


# test uniform vs gaussian, for univariate, multivariate and kernelized version
# function avec Gaussian, n_dims return 2 rmse_uniform, rmse_gaussian


# Gaussian case
# gaussian_dict, X, code = _generate_testbed(kernel_init_len=kernel_init_len,
# n_nonzero_coefs=n_nonzero_coefs,
# n_kernels=n_kernels,
# n_samples=n_samples,
# n_features=n_features,
# n_dims=n_dims,
# rng=rng_global,
# Gaussian=True)
# rmse_gaussian = zeros(shape=(n_nonzero_coefs, n_samples))
# for k in range(n_nonzero_coefs):
# for idx, s in enumerate(X):
# r, _ = multivariate_sparse_encode(array(s, ndmin=3), gaussian_dict,
# n_nonzero_coefs=k+1,
# n_jobs=n_jobs,
# verbose=1)
# rmse_gaussian[k, idx] = norm(r[0], 'fro')/norm(s, 'fro')*100

# # Uniform case
# uniform_dict, X, code = _generate_testbed(kernel_init_len=kernel_init_len,
# n_nonzero_coefs=n_nonzero_coefs,
# n_kernels=n_kernels,
# n_samples=n_samples,
# n_features=n_features,
# n_dims=n_dims,
# rng=rng_global,
# Gaussian=False)
# rmse_uniform = zeros(shape=(n_nonzero_coefs, n_samples))
# for k in range(n_nonzero_coefs):
# for idx, s in enumerate(X):
# r, _ = multivariate_sparse_encode(array(s, ndmin=3), uniform_dict,
# n_nonzero_coefs=k+1,
# n_jobs=n_jobs,
# verbose=1)
# rmse_uniform[k, idx] = norm(r[0], 'fro')/norm(s, 'fro')*100

# fig = plt.figure(figsize=(10,8))
# plt.title(r'Initialization of the generating dictionary')
# km = fig.add_subplot(1,1,1)
# km.errorbar(range(1, n_nonzero_coefs+1), rmse_uniform.mean(1),
# yerr=rmse_uniform.std(1), label='Uniform')
# km.errorbar(range(1, n_nonzero_coefs+1), rmse_gaussian.mean(1),
# yerr=rmse_gaussian.std(1), color='r', label='Gaussian')
# km.plot(range(n_nonzero_coefs+2), np.zeros(n_nonzero_coefs+2), 'k')
# km.axis([0, n_nonzero_coefs+1, 0, 90])
# km.set_xticks(range(0, n_nonzero_coefs+2, 5))
# km.set_xlabel('k')
# km.set_ylabel('rRMSE (%)')
# km.legend(loc='upper right')
# plt.savefig('omp-rmse-UniformGaussian.png')

0 comments on commit 8c209f6

Please sign in to comment.