From 20b8dad97a38172b9b76f983dda185c6cc830d03 Mon Sep 17 00:00:00 2001
From: Joerg Martin <joerg.martin@ptb.de>
Date: Tue, 25 Jan 2022 13:30:52 +0000
Subject: [PATCH] cubic and sine dataset added
---
EIVPackage/EIVData/cubic.py | 62 +++++++++++++++++++
EIVPackage/EIVData/sine.py | 63 ++++++++++++++++++++
Experiments/configurations/eiv_cubic.json | 21 +++++++
Experiments/configurations/eiv_sine.json | 21 +++++++
Experiments/configurations/noneiv_cubic.json | 20 +++++++
Experiments/configurations/noneiv_sine.json | 20 +++++++
6 files changed, 207 insertions(+)
create mode 100644 EIVPackage/EIVData/cubic.py
create mode 100644 EIVPackage/EIVData/sine.py
create mode 100644 Experiments/configurations/eiv_cubic.json
create mode 100644 Experiments/configurations/eiv_sine.json
create mode 100644 Experiments/configurations/noneiv_cubic.json
create mode 100644 Experiments/configurations/noneiv_sine.json
diff --git a/EIVPackage/EIVData/cubic.py b/EIVPackage/EIVData/cubic.py
new file mode 100644
index 0000000..5d5c63e
--- /dev/null
+++ b/EIVPackage/EIVData/cubic.py
@@ -0,0 +1,62 @@
+import torch
+import sys
+from torch.utils.data import TensorDataset
+
+from EIVGeneral.manipulate_tensors import add_noise
+
+total_number_of_datapoints = 2000
+input_range = [-4,4]
+slope = 1.0
+intercept = 0.0
+x_noise_strength = 0.05 * (input_range[1] - input_range[0])/2
+y_noise_strength = 3
+
+def load_data(seed=0, splitting_part=0.8, normalize=True,
+ return_ground_truth=False):
+ """
+ Loads one-dimensional, cubic data as in Hernandez-Lobato, Adams 2015.
+ :param seed: Seed for drawing and splitting the data.
+ :param splitting_part: Which fraction of the data to use as training
+ data. Defaults to 0.8.
+ :param normalize: Whether to normalize the data, defaults to True.
+ :param return_ground_truth: Boolean. If True, the unnoisy ground truth will
+ also be returned. Defaults to False.
+ :returns: cubic_trainset, cubic_testset if return_ground_truth is False,
+ else cubic_trainset, cubic_testset, true_cubic_trainset,
+ true_cubic_testset. The later two return **four tensors**: The true x,y and
+ their noisy counterparts.
+ """
+ random_generator = torch.Generator().manual_seed(seed)
+ # draw different seeds for noise and splitting
+ seeds = [int(t) for t in torch.randint(0,sys.maxsize,(3,),\
+ generator=random_generator)]
+ # create new generators from tensor seeds
+ true_x = input_range[0] + (input_range[1]-input_range[0])\
+ * torch.rand((total_number_of_datapoints,1),
+ generator=torch.Generator().manual_seed(seeds[0]))
+ true_y = slope * true_x**3 + intercept
+ # add noise and normalize x and y
+ (noisy_x, noisy_y), (true_x, true_y) = add_noise(
+ tensor_list=(true_x, true_y),
+ noise_strength_list=(x_noise_strength, y_noise_strength),
+ seed_list=seeds[1:3],
+ normalize=normalize)
+ # create datasets
+ dataset_len = noisy_x.shape[0]
+ train_len = int(dataset_len*splitting_part)
+ test_len = dataset_len - train_len
+ true_train_x, true_test_x = torch.split(true_x, [train_len, test_len])
+ true_train_y, true_test_y = torch.split(true_y, [train_len, test_len])
+ noisy_train_x, noisy_test_x = torch.split(noisy_x, [train_len, test_len])
+ noisy_train_y, noisy_test_y = torch.split(noisy_y, [train_len, test_len])
+ cubic_trainset = TensorDataset(noisy_train_x, noisy_train_y)
+ cubic_testset = TensorDataset(noisy_test_x, noisy_test_y)
+ true_cubic_trainset = TensorDataset(true_train_x, true_train_y,
+ noisy_train_x, noisy_train_y)
+ true_cubic_testset = TensorDataset(true_test_x, true_test_y,
+ noisy_test_x, noisy_test_y)
+ if not return_ground_truth:
+ return cubic_trainset, cubic_testset
+ else:
+ return cubic_trainset, cubic_testset, true_cubic_trainset,\
+ true_cubic_testset
diff --git a/EIVPackage/EIVData/sine.py b/EIVPackage/EIVData/sine.py
new file mode 100644
index 0000000..d9ca0b8
--- /dev/null
+++ b/EIVPackage/EIVData/sine.py
@@ -0,0 +1,63 @@
+import torch
+import sys
+from torch.utils.data import TensorDataset
+
+from EIVGeneral.manipulate_tensors import add_noise
+
+total_number_of_datapoints = 2000
+input_range = [-0.2,0.8]
+intercept = 0.0
+x_noise_strength = 0.02
+y_noise_strength = 0.05
+
+def load_data(seed=0, splitting_part=0.8, normalize=True,
+ return_ground_truth=False):
+ """
+ Loads one-dimensional, sine shaped data as in Blundell et al. 2014.
+ :param seed: Seed for drawing and splitting the data.
+ :param splitting_part: Which fraction of the data to use as training
+ data. Defaults to 0.8.
+ :param normalize: Whether to normalize the data, defaults to True.
+ :param return_ground_truth: Boolean. If True, the unnoisy ground truth will
+ also be returned. Defaults to False.
+ :returns: sine_trainset, sine_testset if return_ground_truth is False,
+ else sine_trainset, sine_testset, true_sine_trainset,
+ true_sine_testset. The later two return **four tensors**: The true x,y and
+ their noisy counterparts.
+ """
+ random_generator = torch.Generator().manual_seed(seed)
+ # draw different seeds for noise and splitting
+ seeds = [int(t) for t in torch.randint(0,sys.maxsize,(3,),\
+ generator=random_generator)]
+ # create new generators from tensor seeds
+ true_x = input_range[0] + (input_range[1]-input_range[0])\
+ * torch.rand((total_number_of_datapoints,1),
+ generator=torch.Generator().manual_seed(seeds[0]))
+ true_y = true_x +\
+ torch.sin(2 * torch.pi * true_x) +\
+ torch.sin(4 * torch.pi * true_x)
+ # add noise and normalize x and y
+ (noisy_x, noisy_y), (true_x, true_y) = add_noise(
+ tensor_list=(true_x, true_y),
+ noise_strength_list=(x_noise_strength, y_noise_strength),
+ seed_list=seeds[1:3],
+ normalize=normalize)
+ # create datasets
+ dataset_len = noisy_x.shape[0]
+ train_len = int(dataset_len*splitting_part)
+ test_len = dataset_len - train_len
+ true_train_x, true_test_x = torch.split(true_x, [train_len, test_len])
+ true_train_y, true_test_y = torch.split(true_y, [train_len, test_len])
+ noisy_train_x, noisy_test_x = torch.split(noisy_x, [train_len, test_len])
+ noisy_train_y, noisy_test_y = torch.split(noisy_y, [train_len, test_len])
+ sine_trainset = TensorDataset(noisy_train_x, noisy_train_y)
+ sine_testset = TensorDataset(noisy_test_x, noisy_test_y)
+ true_sine_trainset = TensorDataset(true_train_x, true_train_y,
+ noisy_train_x, noisy_train_y)
+ true_sine_testset = TensorDataset(true_test_x, true_test_y,
+ noisy_test_x, noisy_test_y)
+ if not return_ground_truth:
+ return sine_trainset, sine_testset
+ else:
+ return sine_trainset, sine_testset, true_sine_trainset,\
+ true_sine_testset
diff --git a/Experiments/configurations/eiv_cubic.json b/Experiments/configurations/eiv_cubic.json
new file mode 100644
index 0000000..6b81fef
--- /dev/null
+++ b/Experiments/configurations/eiv_cubic.json
@@ -0,0 +1,21 @@
+{
+ "long_dataname": "cubic",
+ "short_dataname": "cubic",
+ "lr": 1e-3,
+ "batch_size": 64,
+ "test_batch_size": 800,
+ "number_of_epochs": 100,
+ "unscaled_reg": 10,
+ "report_point": 5,
+ "p": 0.1,
+ "lr_update": 20,
+ "std_y_update_points": [1,40],
+ "eiv_prediction_number_of_draws": [100,5],
+ "eiv_prediction_number_of_batches": 10,
+ "init_std_y_list": [0.5],
+ "gamma": 0.5,
+ "hidden_layers": [128, 128, 128, 128],
+ "fixed_std_x": 0.05,
+ "seed_range": [0,10],
+ "gpu_number": 1
+}
diff --git a/Experiments/configurations/eiv_sine.json b/Experiments/configurations/eiv_sine.json
new file mode 100644
index 0000000..b632e1d
--- /dev/null
+++ b/Experiments/configurations/eiv_sine.json
@@ -0,0 +1,21 @@
+{
+ "long_dataname": "sine",
+ "short_dataname": "sine",
+ "lr": 1e-3,
+ "batch_size": 64,
+ "test_batch_size": 800,
+ "number_of_epochs": 100,
+ "unscaled_reg": 10,
+ "report_point": 5,
+ "p": 0.1,
+ "lr_update": 20,
+ "std_y_update_points": [1,40],
+ "eiv_prediction_number_of_draws": [100,5],
+ "eiv_prediction_number_of_batches": 10,
+ "init_std_y_list": [0.1],
+ "gamma": 0.5,
+ "hidden_layers": [128, 128, 128, 128],
+ "fixed_std_x": 0.02,
+ "seed_range": [0,10],
+ "gpu_number": 1
+}
diff --git a/Experiments/configurations/noneiv_cubic.json b/Experiments/configurations/noneiv_cubic.json
new file mode 100644
index 0000000..00fdce9
--- /dev/null
+++ b/Experiments/configurations/noneiv_cubic.json
@@ -0,0 +1,20 @@
+{
+ "long_dataname": "cubic",
+ "short_dataname": "cubic",
+ "lr": 1e-3,
+ "batch_size": 64,
+ "test_batch_size": 800,
+ "number_of_epochs": 100,
+ "unscaled_reg": 10,
+ "report_point": 5,
+ "p": 0.1,
+ "lr_update": 20,
+ "std_y_update_points": [1,40] ,
+ "noneiv_prediction_number_of_draws": 100,
+ "noneiv_prediction_number_of_batches": 10,
+ "init_std_y_list": [0.5],
+ "gamma": 0.5,
+ "hidden_layers": [128, 128, 128, 128],
+ "seed_range": [0,10],
+ "gpu_number": 1
+}
diff --git a/Experiments/configurations/noneiv_sine.json b/Experiments/configurations/noneiv_sine.json
new file mode 100644
index 0000000..c94c589
--- /dev/null
+++ b/Experiments/configurations/noneiv_sine.json
@@ -0,0 +1,20 @@
+{
+ "long_dataname": "sine",
+ "short_dataname": "sine",
+ "lr": 1e-3,
+ "batch_size": 64,
+ "test_batch_size": 800,
+ "number_of_epochs": 100,
+ "unscaled_reg": 10,
+ "report_point": 5,
+ "p": 0.1,
+ "lr_update": 20,
+ "std_y_update_points": [1,40] ,
+ "noneiv_prediction_number_of_draws": 100,
+ "noneiv_prediction_number_of_batches": 10,
+ "init_std_y_list": [0.1],
+ "gamma": 0.5,
+ "hidden_layers": [128, 128, 128, 128],
+ "seed_range": [0,10],
+ "gpu_number": 1
+}
--
GitLab