author int64 658 755k | date stringdate 2012-06-12 08:34:29 2024-07-22 14:51:21 | timezone int64 -46,800 43.2k | hash stringlengths 40 40 | message stringlengths 5 490 | mods listlengths 1 16 | language stringclasses 20
values | license stringclasses 3
values | repo stringlengths 5 68 | original_message stringlengths 12 491 |
|---|---|---|---|---|---|---|---|---|---|
641,005 | 09.02.2021 02:55:44 | -32,400 | 53828f6a1c7f495ace812b94061c7e353e79cdc0 | implement sample_reward and calc_ground_truth_policy_value | [
{
"change_type": "MODIFY",
"old_path": "examples/online/README.md",
"new_path": "examples/online/README.md",
"diff": "@@ -27,6 +27,7 @@ python evaluate_off_policy_estimators.py\\\n--n_runs $n_runs\\\n--n_rounds $n_rounds\\\n--n_actions $n_actions\\\n+ --n_sim $n_sim\\\n--dim_context $dim_context\\\n... | Python | Apache License 2.0 | st-tech/zr-obp | implement sample_reward and calc_ground_truth_policy_value |
641,006 | 20.02.2021 22:46:47 | -32,400 | a0a58f72a1e69d7d2c4c43146b85bde4f5209dd4 | apply review and fix minor tests | [
{
"change_type": "MODIFY",
"old_path": "obp/ope/regression_model.py",
"new_path": "obp/ope/regression_model.py",
"diff": "@@ -144,16 +144,16 @@ class RegressionModel(BaseEstimator):\n)\nif position.max() >= self.len_list:\nraise ValueError(\n- f\"position elements must be smaller than len_list, but ... | Python | Apache License 2.0 | st-tech/zr-obp | apply review and fix minor tests |
641,006 | 20.02.2021 23:14:22 | -32,400 | 41e6e488fffa2827fa88ae741b31e5f816911088 | remove switch-ipw | [
{
"change_type": "MODIFY",
"old_path": "obp/ope/__init__.py",
"new_path": "obp/ope/__init__.py",
"diff": "@@ -9,7 +9,6 @@ __all_estimators__ = [\n\"DirectMethod\",\n\"DoublyRobust\",\n\"DoublyRobustWithShrinkage\",\n- \"SwitchInverseProbabilityWeighting\",\n\"SwitchDoublyRobust\",\n\"SelfNormalizedD... | Python | Apache License 2.0 | st-tech/zr-obp | remove switch-ipw |
641,006 | 20.02.2021 23:52:37 | -32,400 | 72a634db7d699d0ed84dadd7d27c4b35ba03a705 | add relative estimated policy value when summmarizing ope | [
{
"change_type": "MODIFY",
"old_path": "obp/ope/meta.py",
"new_path": "obp/ope/meta.py",
"diff": "@@ -257,8 +257,18 @@ class OffPolicyEvaluation:\nrandom_state=random_state,\n)\n)\n-\n- return policy_value_df.T, policy_value_interval_df.T\n+ policy_value_of_behavior_policy = self.bandit_feedback[\"r... | Python | Apache License 2.0 | st-tech/zr-obp | add relative estimated policy value when summmarizing ope |
641,003 | 21.02.2021 14:38:33 | -32,400 | ed49ea64ac02c7e2ff7a0e1618dc51f385cb0d8a | split output of obtain_batch_bandit_feedback when is_timeseries_split is True | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/real.py",
"new_path": "obp/dataset/real.py",
"diff": "@@ -6,6 +6,8 @@ from dataclasses import dataclass\nfrom logging import getLogger, basicConfig, INFO\nfrom pathlib import Path\nfrom typing import Optional\n+from typing import Union\n+from typin... | Python | Apache License 2.0 | st-tech/zr-obp | split output of obtain_batch_bandit_feedback when is_timeseries_split is True |
641,003 | 21.02.2021 14:50:37 | -32,400 | b0700f0857dddfb6b81be1edb5adfa2db58b7bd9 | fix bandit feedback test | [
{
"change_type": "MODIFY",
"old_path": "tests/dataset/test_real.py",
"new_path": "tests/dataset/test_real.py",
"diff": "@@ -2,6 +2,8 @@ import pytest\nimport numpy as np\nimport pandas as pd\n+from typing import Tuple\n+\nfrom obp.dataset import OpenBanditDataset\n@@ -63,22 +65,26 @@ def test_obtain... | Python | Apache License 2.0 | st-tech/zr-obp | fix bandit feedback test |
641,014 | 21.02.2021 16:53:59 | -32,400 | a3d171b579e347b2aa04fb616ceea8a865cb7ae3 | set None to position array of synthetic and classification data | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/multiclass.py",
"new_path": "obp/dataset/multiclass.py",
"diff": "@@ -107,7 +107,7 @@ class MultiClassToBanditReduction(BaseBanditDataset):\n[ 0., 0., 4., ..., 15., 3., 0.]]),\n'action': array([6, 8, 5, ..., 2, 5, 9]),\n'reward': array([1., 1., 1.,... | Python | Apache License 2.0 | st-tech/zr-obp | set None to position array of synthetic and classification data |
641,003 | 21.02.2021 17:45:52 | -32,400 | 49f389710b4b0da259bef666def0e5a1cd153410 | remove reward_test and add n_rounds | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/real.py",
"new_path": "obp/dataset/real.py",
"diff": "@@ -149,13 +149,16 @@ class OpenBanditDataset(BaseRealBanditDataset):\nThis parameter is used as a ground-truth policy value in the evaluation of OPE estimators.\n\"\"\"\n- return (\n- cls(behav... | Python | Apache License 2.0 | st-tech/zr-obp | remove reward_test and add n_rounds |
641,003 | 21.02.2021 18:04:05 | -32,400 | 027d944c2c1d145101909e66ced9109276cecd50 | fix sample_bootstrap_bandit_feedback | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/real.py",
"new_path": "obp/dataset/real.py",
"diff": "@@ -299,6 +299,11 @@ class OpenBanditDataset(BaseRealBanditDataset):\n- action_context: item-related context vectors\n\"\"\"\n+ if is_timeseries_split:\n+ bandit_feedback = self.obtain_batch_ban... | Python | Apache License 2.0 | st-tech/zr-obp | fix sample_bootstrap_bandit_feedback |
641,014 | 22.02.2021 15:13:13 | -32,400 | b638601f2da125ab792ec03e99b9540e656a3e5a | fix reg model to allow position=None | [
{
"change_type": "MODIFY",
"old_path": "obp/ope/regression_model.py",
"new_path": "obp/ope/regression_model.py",
"diff": "@@ -135,13 +135,9 @@ class RegressionModel(BaseEstimator):\n)\nn_rounds = context.shape[0]\n- if self.len_list == 1:\n+ if position is None or self.len_list == 1:\nposition = np.... | Python | Apache License 2.0 | st-tech/zr-obp | fix reg model to allow position=None |
641,014 | 22.02.2021 19:57:50 | -32,400 | 5c122f756140121cb9302d45eba24f9fa8a1a38e | adjust benchmark/ to recent changes | [
{
"change_type": "MODIFY",
"old_path": "benchmark/README.md",
"new_path": "benchmark/README.md",
"diff": "---\nThis directory includes some benchmark experiments and demonstrations about off-policy evaluation using [the full size Open Bandit Dataset](https://research.zozo.com/data.html). The detaile... | Python | Apache License 2.0 | st-tech/zr-obp | adjust benchmark/ to recent changes |
641,014 | 22.02.2021 19:58:12 | -32,400 | 98d2432123ab00002a4ce4fa39fc00214627d68b | fix n_rounds of bandit_feedback_test | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/real.py",
"new_path": "obp/dataset/real.py",
"diff": "@@ -241,7 +241,7 @@ class OpenBanditDataset(BaseRealBanditDataset):\naction_context=self.action_context,\n)\nbandit_feedback_test = dict(\n- n_rounds=n_rounds_train,\n+ n_rounds=(self.n_rounds -... | Python | Apache License 2.0 | st-tech/zr-obp | fix n_rounds of bandit_feedback_test |
641,014 | 22.02.2021 19:58:32 | -32,400 | baaf874686ac5aef4faf081510f32be3e4a2148b | rm unnecessary value error about random_state | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/synthetic.py",
"new_path": "obp/dataset/synthetic.py",
"diff": "@@ -148,11 +148,8 @@ class SyntheticBanditDataset(BaseBanditDataset):\nraise ValueError(\nf\"reward_type must be either 'binary' or 'continuous, but {self.reward_type} is given.'\"\n)\... | Python | Apache License 2.0 | st-tech/zr-obp | rm unnecessary value error about random_state |
641,014 | 22.02.2021 19:59:05 | -32,400 | c53da163170e790d6abede6d0f22ab2839011f8b | change backend of Parallel | [
{
"change_type": "MODIFY",
"old_path": "examples/multiclass/evaluate_off_policy_estimators.py",
"new_path": "examples/multiclass/evaluate_off_policy_estimators.py",
"diff": "@@ -48,10 +48,10 @@ ope_estimators = [\nSelfNormalizedInverseProbabilityWeighting(),\nDoublyRobust(),\nSelfNormalizedDoublyRob... | Python | Apache License 2.0 | st-tech/zr-obp | change backend of Parallel |
641,007 | 23.02.2021 01:24:20 | -32,400 | bb9e1d348657d21cde84c3753736fdf4c250a195 | fix type hint at fit_predict & calc_ground_truth_policy_value method | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/multiclass.py",
"new_path": "obp/dataset/multiclass.py",
"diff": "@@ -301,7 +301,7 @@ class MultiClassToBanditReduction(BaseBanditDataset):\n)\nreturn np.expand_dims(pi_e, 2)\n- def calc_ground_truth_policy_value(self, action_dist: np.ndarray) -> n... | Python | Apache License 2.0 | st-tech/zr-obp | fix type hint at fit_predict & calc_ground_truth_policy_value method |
641,003 | 07.03.2021 17:17:32 | -32,400 | 4058abbed87fbf170b16950ac1077f9eb767f088 | add lints.yml | [
{
"change_type": "ADD",
"old_path": null,
"new_path": ".github/workflows/lints.yml",
"diff": "+name: Lints\n+\n+on:\n+ push:\n+ branches:\n+ - master\n+ pull_request: {}\n+\n+jobs:\n+ lints:\n+ runs-on: ubuntu-latest\n+\n+ steps:\n+ - name: Checkout\n+ uses: actions/checkout@v2\n+\n+ - name: Setup P... | Python | Apache License 2.0 | st-tech/zr-obp | add lints.yml |
641,003 | 07.03.2021 17:18:43 | -32,400 | cb92ca6a547dceab6d8fe5b02ea748a1f0ec07d8 | add tests.yml | [
{
"change_type": "ADD",
"old_path": null,
"new_path": ".github/workflows/tests.yml",
"diff": "+name: Tests\n+\n+on:\n+ push:\n+ branches:\n+ - master\n+ pull_request: {}\n+\n+jobs:\n+ tests:\n+ runs-on: ubuntu-latest\n+\n+ strategy:\n+ matrix:\n+ python-version: [3.7, 3.8, 3.9]\n+\n+ # Not intended ... | Python | Apache License 2.0 | st-tech/zr-obp | add tests.yml |
641,003 | 07.03.2021 17:22:41 | -32,400 | 302353d66b9a80272806ea98263f677bbcf1006f | adapt link to upstream repo | [
{
"change_type": "MODIFY",
"old_path": ".github/workflows/tests.yml",
"new_path": ".github/workflows/tests.yml",
"diff": "@@ -15,7 +15,7 @@ jobs:\npython-version: [3.7, 3.8, 3.9]\n# Not intended for forks.\n- if: github.repository == 'nmasahiro/zr-obp'\n+ if: github.repository == 'st-tech/zr-obp'\ns... | Python | Apache License 2.0 | st-tech/zr-obp | adapt link to upstream repo |
641,003 | 07.03.2021 17:34:11 | -32,400 | b24c6d513a34aae1bfeefee820325ebe4b212d09 | add description of CI to CONTRIBUTING.md | [
{
"change_type": "MODIFY",
"old_path": "CONTRIBUTING.md",
"new_path": "CONTRIBUTING.md",
"diff": "@@ -29,3 +29,7 @@ After installing flake8, you can check the coding style by the following command\n# perform checking of the coding style\n$ flake8 .\n```\n+\n+## Continuous Integration\n+\n+Open Bandi... | Python | Apache License 2.0 | st-tech/zr-obp | add description of CI to CONTRIBUTING.md |
641,006 | 07.03.2021 19:49:02 | -32,400 | 2e1cd0b2bdff685c2bbd0f285fd28a23f9ccd72c | fix hyperparameter validations of dr-variants | [
{
"change_type": "MODIFY",
"old_path": "obp/ope/estimators.py",
"new_path": "obp/ope/estimators.py",
"diff": "@@ -1098,10 +1098,12 @@ class SwitchDoublyRobust(DoublyRobust):\ndef __post_init__(self) -> None:\n\"\"\"Initialize Class.\"\"\"\n- if not isinstance(self.tau, float):\n+ if not isinstance(s... | Python | Apache License 2.0 | st-tech/zr-obp | fix hyperparameter validations of dr-variants |
641,005 | 08.03.2021 06:38:42 | -32,400 | d655183d41149a9d65da7889fafcfab728ec217d | add mypy-extensions | [
{
"change_type": "MODIFY",
"old_path": "requirements.txt",
"new_path": "requirements.txt",
"diff": "matplotlib>=3.2.2\n+mypy-extensions>=0.4.3\nnumpy>=1.18.1\npandas>=0.25.1\npyyaml>=5.1\n"
},
{
"change_type": "MODIFY",
"old_path": "setup.py",
"new_path": "setup.py",
"diff": "@@ ... | Python | Apache License 2.0 | st-tech/zr-obp | add mypy-extensions |
641,005 | 08.03.2021 21:00:07 | -32,400 | c4ee9ab91bb58a01ed9de99ba7efb822c983b0a5 | use predict_proa | [
{
"change_type": "MODIFY",
"old_path": "examples/opl/README.md",
"new_path": "examples/opl/README.md",
"diff": "@@ -81,8 +81,10 @@ python evaluate_off_policy_learners.py\\\n# =============================================\n# random_state=12345\n# ---------------------------------------------\n-# rand... | Python | Apache License 2.0 | st-tech/zr-obp | use predict_proa |
641,003 | 14.03.2021 14:34:16 | -32,400 | 9256813f7a184cf2a2db6969941378e9c5a73a2c | allow position is None | [
{
"change_type": "MODIFY",
"old_path": "obp/policy/offline.py",
"new_path": "obp/policy/offline.py",
"diff": "@@ -145,7 +145,7 @@ class IPWLearner(BaseOfflinePolicyLearner):\nif pscore is None:\nn_actions = np.int(action.max() + 1)\npscore = np.ones_like(action) / n_actions\n- if self.len_list == 1:... | Python | Apache License 2.0 | st-tech/zr-obp | allow position is None |
641,003 | 14.03.2021 15:00:15 | -32,400 | f209a674623b7d32b0e98e85eabe39bce5098bb8 | test offline estimation performance | [
{
"change_type": "ADD",
"old_path": null,
"new_path": "tests/policy/test_offline_estimation_performance.py",
"diff": "+import numpy as np\n+from pandas import DataFrame\n+from joblib import Parallel, delayed\n+from sklearn.experimental import enable_hist_gradient_boosting # NOQA\n+from sklearn.ensem... | Python | Apache License 2.0 | st-tech/zr-obp | test offline estimation performance |
641,003 | 14.03.2021 15:45:06 | -32,400 | 5250825ccfe2c056c660a1aaccf0b493efeb919a | add test offline learner performance | [
{
"change_type": "RENAME",
"old_path": "tests/policy/test_offline_estimation_performance.py",
"new_path": "tests/ope/test_offline_estimation_performance.py",
"diff": "@@ -141,7 +141,7 @@ ope_estimators = [\n\"n_rounds, n_actions, dim_context, base_model_for_evaluation_policy, base_model_for_reg_mode... | Python | Apache License 2.0 | st-tech/zr-obp | add test offline learner performance |
641,006 | 15.03.2021 00:28:46 | -32,400 | 142bb208eeec48988ce9a9ed9983a2bae1c9e385 | add slate bandit dataset | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/__init__.py",
"new_path": "obp/dataset/__init__.py",
"diff": "from .base import *\nfrom .real import *\nfrom .synthetic import *\n+from .synthetic_slate import *\nfrom .multiclass import *\n"
},
{
"change_type": "ADD",
"old_path": null,... | Python | Apache License 2.0 | st-tech/zr-obp | add slate bandit dataset |
641,006 | 15.03.2021 02:28:39 | -32,400 | 3d6ca4e79066e197e66bb1ee4152ad5a5a5a3c7d | add basic reward functions | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/synthetic_slate.py",
"new_path": "obp/dataset/synthetic_slate.py",
"diff": "@@ -127,6 +127,7 @@ class SyntheticSlateBanditDataset(BaseBanditDataset):\nlen_list: int\ndim_context: int = 1\nreward_type: str = \"binary\"\n+ reward_structure: str = \"R... | Python | Apache License 2.0 | st-tech/zr-obp | add basic reward functions |
641,006 | 16.03.2021 22:28:46 | -32,400 | 46ea0ce278cab090dc589ed69743e61eeefd63b1 | add reward generator | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/synthetic_slate.py",
"new_path": "obp/dataset/synthetic_slate.py",
"diff": "@@ -128,6 +128,7 @@ class SyntheticSlateBanditDataset(BaseBanditDataset):\ndim_context: int = 1\nreward_type: str = \"binary\"\nreward_structure: str = \"RIPS\"\n+ exam_wei... | Python | Apache License 2.0 | st-tech/zr-obp | add reward generator |
641,003 | 20.03.2021 11:02:31 | -32,400 | ef12cf06cb89b37dd64fc5b6bf385dbbc15fdb8e | change reward_type to enum | [
{
"change_type": "ADD",
"old_path": null,
"new_path": "obp/dataset/reward_type.py",
"diff": "+import enum\n+\n+\n+class RewardType(enum.Enum):\n+ \"\"\"Reward type.\n+\n+ Attributes\n+ ----------\n+ BINARY:\n+ The reward type is binary.\n+ CONTINUOUS:\n+ The reward type is continuous.\n+ \"\"\"\n+\n... | Python | Apache License 2.0 | st-tech/zr-obp | change reward_type to enum |
641,003 | 20.03.2021 11:24:42 | -32,400 | 83013355a91ee884f48b83d817c8988c4af9b180 | change policy_type to enum | [
{
"change_type": "MODIFY",
"old_path": "obp/policy/base.py",
"new_path": "obp/policy/base.py",
"diff": "@@ -9,6 +9,8 @@ from typing import Optional\nimport numpy as np\nfrom sklearn.utils import check_random_state\n+from .policy_type import PolicyType\n+\n@dataclass\nclass BaseContextFreePolicy(meta... | Python | Apache License 2.0 | st-tech/zr-obp | change policy_type to enum |
641,003 | 20.03.2021 12:19:55 | -32,400 | 9fdddedcf9911e43b4f28526876dc21eeb06814e | fix flake8 error F541 | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/multiclass.py",
"new_path": "obp/dataset/multiclass.py",
"diff": "@@ -150,7 +150,7 @@ class MultiClassToBanditReduction(BaseBanditDataset):\ndef __post_init__(self) -> None:\n\"\"\"Initialize Class.\"\"\"\nif not is_classifier(self.base_classifier_... | Python | Apache License 2.0 | st-tech/zr-obp | fix flake8 error F541 |
641,003 | 20.03.2021 12:44:35 | -32,400 | 8e42430714ffc3b450d3dd03acab507e666b6be6 | fix flake8 error for dataset/__init__.py | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/__init__.py",
"new_path": "obp/dataset/__init__.py",
"diff": "-from .base import *\n-from .real import *\n-from .synthetic import *\n-from .multiclass import *\n+from obp.dataset.base import BaseBanditDataset\n+from obp.dataset.base import BaseReal... | Python | Apache License 2.0 | st-tech/zr-obp | fix flake8 error for dataset/__init__.py |
641,003 | 20.03.2021 12:53:19 | -32,400 | 81b79b318c11c24b67f387a074ba3d2f2601ddd8 | fix flake8 error for simulation/__init__.py | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/__init__.py",
"new_path": "obp/dataset/__init__.py",
"diff": "@@ -2,6 +2,9 @@ from obp.dataset.base import BaseBanditDataset\nfrom obp.dataset.base import BaseRealBanditDataset\nfrom obp.dataset.real import OpenBanditDataset\nfrom obp.dataset.synth... | Python | Apache License 2.0 | st-tech/zr-obp | fix flake8 error for simulation/__init__.py |
641,003 | 20.03.2021 13:38:17 | -32,400 | 436dc900b9a1af49f390f8499159b730459212c3 | replace expand_dims with np.newaxis | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/multiclass.py",
"new_path": "obp/dataset/multiclass.py",
"diff": "@@ -299,7 +299,7 @@ class MultiClassToBanditReduction(BaseBanditDataset):\npi_e[np.arange(self.n_rounds_ev), preds] = (\nalpha_e + (1.0 - alpha_e) / self.n_actions\n)\n- return np.ex... | Python | Apache License 2.0 | st-tech/zr-obp | replace expand_dims with np.newaxis |
641,003 | 20.03.2021 13:51:07 | -32,400 | 7e998f7e1e276ef3fcdcf31f544a9911074f8678 | add flake8 to github actions | [
{
"change_type": "MODIFY",
"old_path": ".github/workflows/lints.yml",
"new_path": ".github/workflows/lints.yml",
"diff": "@@ -23,3 +23,9 @@ jobs:\nuses: psf/black@stable\nwith:\nargs: \". --check --diff\"\n+\n+ - name: flake8\n+ run: |\n+ python -m pip install --upgrade pip\n+ pip install flake8\n+ ... | Python | Apache License 2.0 | st-tech/zr-obp | add flake8 to github actions |
641,003 | 20.03.2021 14:51:00 | -32,400 | 8a633d71e410d9986d9118258c2ea8a4e765de80 | adopt auto in Enum | [
{
"change_type": "MODIFY",
"old_path": "obp/policy/policy_type.py",
"new_path": "obp/policy/policy_type.py",
"diff": "@@ -14,9 +14,9 @@ class PolicyType(enum.Enum):\nThe policy type is offline.\n\"\"\"\n- CONTEXT_FREE = 0\n- CONTEXTUAL = 1\n- OFFLINE = 2\n+ CONTEXT_FREE = enum.auto()\n+ CONTEXTUAL =... | Python | Apache License 2.0 | st-tech/zr-obp | adopt auto in Enum |
641,006 | 21.03.2021 17:44:01 | -32,400 | 4d7326c9684aaf0380814a7ae0a0b8036124a78a | fix sample_reward_given_expected_reward; add reward structure where rewards depend on those of other slots | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/synthetic_slate.py",
"new_path": "obp/dataset/synthetic_slate.py",
"diff": "@@ -128,6 +128,7 @@ class SyntheticSlateBanditDataset(BaseBanditDataset):\ndim_context: int = 1\nreward_type: str = \"binary\"\nreward_structure: str = \"RIPS\"\n+ reward_t... | Python | Apache License 2.0 | st-tech/zr-obp | fix sample_reward_given_expected_reward; add reward structure where rewards depend on those of other slots |
641,006 | 21.03.2021 18:50:27 | -32,400 | 95a54403d03fe914526177bb81ac14e9c7ad92dc | clean reward functions; add several comments | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/synthetic_slate.py",
"new_path": "obp/dataset/synthetic_slate.py",
"diff": "@@ -13,7 +13,8 @@ from tqdm import tqdm\nfrom .base import BaseBanditDataset\nfrom ..types import BanditFeedback\n-from ..utils import sigmoid, softmax\n+from ..utils impor... | Python | Apache License 2.0 | st-tech/zr-obp | clean reward functions; add several comments |
641,006 | 03.04.2021 01:06:33 | -32,400 | b88b0b85042f4fbfe440068b0ec46aa1d9371fd2 | fix reward structures; add test of slate dataset | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/synthetic_slate.py",
"new_path": "obp/dataset/synthetic_slate.py",
"diff": "@@ -13,8 +13,7 @@ from tqdm import tqdm\nfrom .base import BaseBanditDataset\nfrom ..types import BanditFeedback\n-from ..utils import softmax\n-from .synthetic import logi... | Python | Apache License 2.0 | st-tech/zr-obp | fix reward structures; add test of slate dataset |
641,006 | 03.04.2021 01:35:50 | -32,400 | e48e5def9d261b7935127d562eee4f025f73d074 | add option of calculating exact pscore marginal of random policy | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/synthetic_slate.py",
"new_path": "obp/dataset/synthetic_slate.py",
"diff": "@@ -276,6 +276,7 @@ class SyntheticSlateBanditDataset(BaseBanditDataset):\nbehavior_policy_logit_: np.ndarray,\nn_rounds: int,\nreturn_pscore_marginal: bool = True,\n+ retu... | Python | Apache License 2.0 | st-tech/zr-obp | add option of calculating exact pscore marginal of random policy |
641,006 | 03.04.2021 21:21:12 | -32,400 | a52d8602f98e46f58d56e795108c27f020961ae8 | add comment; add click models | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/synthetic_slate.py",
"new_path": "obp/dataset/synthetic_slate.py",
"diff": "@@ -51,11 +51,11 @@ class SyntheticSlateBanditDataset(BaseBanditDataset):\nWhen 'cascade_additive' or 'standard_additive' is given, action_effect_matrix (:math:`w`) is gene... | Python | Apache License 2.0 | st-tech/zr-obp | add comment; add click models |
641,006 | 04.04.2021 21:16:41 | -32,400 | 9a6edd92641f3f9b72fbacd1e2244b1316c5e672 | add comment and fix argument order (generate_symmetic_matrix) | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/synthetic_slate.py",
"new_path": "obp/dataset/synthetic_slate.py",
"diff": "@@ -221,8 +221,8 @@ class SyntheticSlateBanditDataset(BaseBanditDataset):\nself.exam_weight = np.ones(self.len_list)\nif self.reward_structure in [\"cascade_additive\", \"s... | Python | Apache License 2.0 | st-tech/zr-obp | add comment and fix argument order (generate_symmetic_matrix) |
641,006 | 04.04.2021 21:42:15 | -32,400 | 46bcc7a20b4d71ae202d2b91d0572b4efa767c02 | unify slot_weight and action_effect -> action_interaction | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/__init__.py",
"new_path": "obp/dataset/__init__.py",
"diff": "@@ -7,9 +7,9 @@ from obp.dataset.synthetic import linear_reward_function\nfrom obp.dataset.synthetic import linear_behavior_policy\nfrom obp.dataset.multiclass import MultiClassToBanditR... | Python | Apache License 2.0 | st-tech/zr-obp | unify slot_weight and action_effect -> action_interaction |
641,006 | 10.04.2021 14:41:34 | -32,400 | 4d96d7589935f865a79276e6c39f895c5fb25b77 | n_actions -> n_unique_action | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/synthetic_slate.py",
"new_path": "obp/dataset/synthetic_slate.py",
"diff": "@@ -30,7 +30,7 @@ class SyntheticSlateBanditDataset(BaseBanditDataset):\nParameters\n-----------\n- n_actions: int (>= len_list)\n+ n_unique_action: int (>= len_list)\nNumb... | Python | Apache License 2.0 | st-tech/zr-obp | n_actions -> n_unique_action |
641,006 | 10.04.2021 14:42:29 | -32,400 | 247bedd57ad07a0ee8afe067a9e3399889f56966 | result -> expected_reward_factual | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/synthetic_slate.py",
"new_path": "obp/dataset/synthetic_slate.py",
"diff": "@@ -807,12 +807,12 @@ def action_interaction_exponential_reward_function(\nexpected_reward_factual = sigmoid(expected_reward_factual)\n# q_l = \\sum_{a} a3d[i, a, l] q_a + ... | Python | Apache License 2.0 | st-tech/zr-obp | result -> expected_reward_factual |
641,006 | 10.04.2021 15:06:50 | -32,400 | c12b0e05aa004f4b3d211ee54baf0c7b468bf783 | remove exam weight from initialization | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/synthetic_slate.py",
"new_path": "obp/dataset/synthetic_slate.py",
"diff": "@@ -66,10 +66,6 @@ class SyntheticSlateBanditDataset(BaseBanditDataset):\nWhen 'cascade' is given, reward of each slot is sampled using the cascade model.\nWhen using some ... | Python | Apache License 2.0 | st-tech/zr-obp | remove exam weight from initialization |
641,006 | 10.04.2021 15:30:09 | -32,400 | 42be0a52a82769ddb59695f6f6b5aba3ed519138 | add input validation of slate dataset | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/synthetic_slate.py",
"new_path": "obp/dataset/synthetic_slate.py",
"diff": "@@ -205,6 +205,10 @@ class SyntheticSlateBanditDataset(BaseBanditDataset):\nraise ValueError(\nf\"reward_structure must be either 'cascade_additive', 'cascade_exponential',... | Python | Apache License 2.0 | st-tech/zr-obp | add input validation of slate dataset |
641,006 | 10.04.2021 17:14:05 | -32,400 | 9f6992593c209c470cf01d0c4e9cdcf9579ee5c9 | add slate ope estimators | [
{
"change_type": "MODIFY",
"old_path": "obp/ope/__init__.py",
"new_path": "obp/ope/__init__.py",
"diff": "@@ -7,6 +7,9 @@ from obp.ope.estimators import DoublyRobust\nfrom obp.ope.estimators import SelfNormalizedDoublyRobust\nfrom obp.ope.estimators import SwitchDoublyRobust\nfrom obp.ope.estimators... | Python | Apache License 2.0 | st-tech/zr-obp | add slate ope estimators |
641,006 | 11.04.2021 16:41:18 | -32,400 | 3a7aa95b9d6dbc4d462f83d3d8eb9d286fc9c300 | fix confidence interval of slate ope | [
{
"change_type": "MODIFY",
"old_path": "obp/ope/estimators_slate.py",
"new_path": "obp/ope/estimators_slate.py",
"diff": "@@ -75,6 +75,15 @@ class BaseSlateInverseProbabilityWeighting(BaseSlateOffPolicyEstimator):\niw = evaluation_policy_pscore / behavior_policy_pscore\nreturn reward * iw * reward_w... | Python | Apache License 2.0 | st-tech/zr-obp | fix confidence interval of slate ope |
641,006 | 11.04.2021 17:18:28 | -32,400 | c8df15c7026f93675c4dbf079cf798607c30c7cc | add comment of SIPS | [
{
"change_type": "MODIFY",
"old_path": "obp/ope/estimators_slate.py",
"new_path": "obp/ope/estimators_slate.py",
"diff": "@@ -14,7 +14,7 @@ from ..utils import check_confidence_interval_arguments\n@dataclass\nclass BaseSlateOffPolicyEstimator(metaclass=ABCMeta):\n- \"\"\"Base class for OPE estimator... | Python | Apache License 2.0 | st-tech/zr-obp | add comment of SIPS |
641,005 | 12.04.2021 17:53:42 | -32,400 | 9220b587b0f91d221151ea463bf54baf3a7534af | fix type check issue in Windows | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/synthetic.py",
"new_path": "obp/dataset/synthetic.py",
"diff": "@@ -239,7 +239,7 @@ class SyntheticBanditDataset(BaseBanditDataset):\nraise ValueError(\n\"the size of axis 0 of context must be the same as that of action\"\n)\n- if not np.issubdtype... | Python | Apache License 2.0 | st-tech/zr-obp | fix type check issue in Windows |
641,006 | 17.04.2021 14:07:40 | -32,400 | 8f806155bf55df19401a98ba38b5b1d450c34e09 | apply review of synthetic_slate | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/synthetic_slate.py",
"new_path": "obp/dataset/synthetic_slate.py",
"diff": "@@ -31,7 +31,7 @@ class SyntheticSlateBanditDataset(BaseBanditDataset):\nParameters\n-----------\nn_unique_action: int (>= len_list)\n- Number of actions.\n+ Number of uniq... | Python | Apache License 2.0 | st-tech/zr-obp | apply review of synthetic_slate |
641,006 | 17.04.2021 14:48:35 | -32,400 | 8066412f7ac18e8d913fdb29c0424cc8defa5f69 | fix bugs; apply review of test_synthetic_slate | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/synthetic_slate.py",
"new_path": "obp/dataset/synthetic_slate.py",
"diff": "@@ -34,7 +34,7 @@ class SyntheticSlateBanditDataset(BaseBanditDataset):\nNumber of unique actions.\nlen_list: int (> 1)\n- Length of a list of actions recommended in each i... | Python | Apache License 2.0 | st-tech/zr-obp | fix bugs; apply review of test_synthetic_slate |
641,006 | 19.04.2021 02:23:51 | -32,400 | 88ff2a9dfd63e1340fe78cb6b570f4382e5739a0 | fix slate ope estimators and add validations | [
{
"change_type": "MODIFY",
"old_path": "obp/ope/estimators_slate.py",
"new_path": "obp/ope/estimators_slate.py",
"diff": "@@ -9,7 +9,12 @@ from typing import Dict, Optional\nimport numpy as np\nfrom sklearn.utils import check_random_state\n-from ..utils import check_confidence_interval_arguments\n+f... | Python | Apache License 2.0 | st-tech/zr-obp | fix slate ope estimators and add validations |
641,006 | 19.04.2021 02:24:05 | -32,400 | 59b7e95144225e2a4a2e7f729a0be74d5b6e23d4 | add slate ope tests | [
{
"change_type": "ADD",
"old_path": null,
"new_path": "tests/ope/test_ipw_estimators_slate.py",
"diff": "+import pytest\n+import numpy as np\n+\n+from obp.ope import SlateStandardIPS, SlateIndependentIPS, SlateRecursiveIPS\n+\n+# setting\n+len_list = 3\n+sips = SlateStandardIPS(len_list=len_list)\n+... | Python | Apache License 2.0 | st-tech/zr-obp | add slate ope tests |
641,006 | 19.04.2021 04:05:41 | -32,400 | 4258c7c177de9186ed453d0a53c692072b9d5e26 | fix synthetic slate bug | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/synthetic_slate.py",
"new_path": "obp/dataset/synthetic_slate.py",
"diff": "@@ -591,7 +591,7 @@ def generate_symmetric_matrix(n_unique_action: int, random_state: int) -> np.nda\nsymmetric_matrix: array-like, shape (n_unique_action, n_unique_action)... | Python | Apache License 2.0 | st-tech/zr-obp | fix synthetic slate bug |
641,006 | 19.04.2021 04:06:18 | -32,400 | 6c245cd57617741f681763e884aa4d4492a510fa | add slate ope performance test | [
{
"change_type": "MODIFY",
"old_path": "tests/ope/test_ipw_estimators_slate.py",
"new_path": "tests/ope/test_ipw_estimators_slate.py",
"diff": "@@ -2,6 +2,11 @@ import pytest\nimport numpy as np\nfrom obp.ope import SlateStandardIPS, SlateIndependentIPS, SlateRecursiveIPS\n+from obp.dataset import (... | Python | Apache License 2.0 | st-tech/zr-obp | add slate ope performance test |
641,006 | 25.04.2021 18:29:01 | -32,400 | 6d13599b83c7d643b2586f31736edf1ee486bfc0 | fix column names related to evaluation_policy_pscore; add efficient bootstrap method | [
{
"change_type": "MODIFY",
"old_path": "obp/ope/__init__.py",
"new_path": "obp/ope/__init__.py",
"diff": "@@ -11,6 +11,7 @@ from obp.ope.estimators_slate import SlateStandardIPS\nfrom obp.ope.estimators_slate import SlateIndependentIPS\nfrom obp.ope.estimators_slate import SlateRecursiveIPS\nfrom ob... | Python | Apache License 2.0 | st-tech/zr-obp | fix column names related to evaluation_policy_pscore; add efficient bootstrap method |
641,005 | 01.05.2021 11:21:35 | -32,400 | 9e142a1c8bd47c4aa138c0e81e361cb1764ed2fd | Refactoring: add subclasses for linear and logistic policies | [
{
"change_type": "MODIFY",
"old_path": "obp/policy/base.py",
"new_path": "obp/policy/base.py",
"diff": "@@ -111,12 +111,6 @@ class BaseContextualPolicy(metaclass=ABCMeta):\nbatch_size: int, default=1\nNumber of samples used in a batch parameter update.\n- alpha_: float, default=1.\n- Prior parameter... | Python | Apache License 2.0 | st-tech/zr-obp | Refactoring: add subclasses for linear and logistic policies |
641,005 | 02.05.2021 13:46:27 | -32,400 | 73bf6f1072f59bcdc229127eff9a176ee3a6c8b4 | Enable to use a dict of estimated_rewards_by_regmodel | [
{
"change_type": "MODIFY",
"old_path": "benchmark/ope/benchmark_off_policy_estimators.py",
"new_path": "benchmark/ope/benchmark_off_policy_estimators.py",
"diff": "@@ -26,6 +26,7 @@ ope_estimators = [\nInverseProbabilityWeighting(),\nSelfNormalizedInverseProbabilityWeighting(),\nDoublyRobust(),\n+ D... | Python | Apache License 2.0 | st-tech/zr-obp | Enable to use a dict of estimated_rewards_by_regmodel |
641,011 | 05.05.2021 20:39:35 | -32,400 | 0b0df8cafa0f2fe80376f4e4ab2c86f3dd90fdf2 | fix pscore_item_position calculation | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/synthetic_slate.py",
"new_path": "obp/dataset/synthetic_slate.py",
"diff": "@@ -376,7 +376,7 @@ class SyntheticSlateBanditDataset(BaseBanditDataset):\nfor action_list in permutations(\nnp.arange(self.n_unique_action), self.len_list\n):\n- if sample... | Python | Apache License 2.0 | st-tech/zr-obp | fix pscore_item_position calculation |
641,011 | 06.05.2021 11:52:27 | -32,400 | 57cae57736f00a8f4887c4d3ac41a6544b880b34 | rm return_exact_uniform_pscore_item_position | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/synthetic_slate.py",
"new_path": "obp/dataset/synthetic_slate.py",
"diff": "@@ -75,7 +75,7 @@ class SyntheticSlateBanditDataset(BaseBanditDataset):\nbehavior_policy_function: Callable[[np.ndarray, np.ndarray], np.ndarray], default=None\nFunction ge... | Python | Apache License 2.0 | st-tech/zr-obp | rm return_exact_uniform_pscore_item_position |
641,006 | 16.05.2021 17:02:06 | -32,400 | b0ff440712d3bd240fc549dec3b218a01b785711 | add generate_evaluation_policy_pscore and calc_on_policy_value | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/synthetic_slate.py",
"new_path": "obp/dataset/synthetic_slate.py",
"diff": "from dataclasses import dataclass\nfrom typing import Optional, Callable, Tuple, Union, List\nfrom itertools import permutations\n+from math import factorial\nimport numpy ... | Python | Apache License 2.0 | st-tech/zr-obp | add generate_evaluation_policy_pscore and calc_on_policy_value |
641,006 | 16.05.2021 22:49:48 | -32,400 | a781e97296a2a39cde971a1985f96059db5eac77 | fix action_2d to action; fix function name of calc_on_policy_value | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/synthetic_slate.py",
"new_path": "obp/dataset/synthetic_slate.py",
"diff": "@@ -563,7 +563,9 @@ class SyntheticSlateBanditDataset(BaseBanditDataset):\npscore_item_position=pscore_item_position,\n)\n- def calc_on_policy_value(self, reward: np.ndarra... | Python | Apache License 2.0 | st-tech/zr-obp | fix action_2d to action; fix function name of calc_on_policy_value |
641,011 | 17.05.2021 10:18:49 | -32,400 | 60ac0f6f558e752f70dff80c8fd40d9c64ee58dd | reflect comments on PR88 | [
{
"change_type": "MODIFY",
"old_path": "examples/quickstart/README.md",
"new_path": "examples/quickstart/README.md",
"diff": "@@ -7,3 +7,4 @@ This page contains a list of quickstart notebooks written with the Open Bandit P\n- [`multiclass.ipynb`](./multiclass.ipynb): a quickstart guide to handle mul... | Python | Apache License 2.0 | st-tech/zr-obp | reflect comments on PR88 |
641,014 | 17.05.2021 16:12:07 | -32,400 | 5c180f177e343f362e861a98a50bc2fb5adfc398 | generalize click models and fix minor erros | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/synthetic_slate.py",
"new_path": "obp/dataset/synthetic_slate.py",
"diff": "from dataclasses import dataclass\nfrom typing import Optional, Callable, Tuple, Union, List\nfrom itertools import permutations\n-from math import factorial\nimport numpy ... | Python | Apache License 2.0 | st-tech/zr-obp | generalize click models and fix minor erros |
641,014 | 17.05.2021 16:12:30 | -32,400 | cd1ba090406f211053f05ccd59f65aaf632a0fe7 | add some tests about click models | [
{
"change_type": "MODIFY",
"old_path": "tests/dataset/test_synthetic_slate.py",
"new_path": "tests/dataset/test_synthetic_slate.py",
"diff": "@@ -13,7 +13,7 @@ from obp.dataset import (\nfrom obp.types import BanditFeedback\n-# n_unique_action, len_list, dim_context, reward_type, reward_structure, c... | Python | Apache License 2.0 | st-tech/zr-obp | add some tests about click models |
641,014 | 17.05.2021 16:27:17 | -32,400 | 71e644fb42b96e8be6ac81f6f353fee20a912a24 | fix eng in synthetic_slate.ipynb | [
{
"change_type": "MODIFY",
"old_path": "examples/quickstart/synthetic_slate.ipynb",
"new_path": "examples/quickstart/synthetic_slate.ipynb",
"diff": "\"---\\n\",\n\"This notebook provides an example of conducting OPE of several different evaluation policies with synthetic slate bandit feedback data.... | Python | Apache License 2.0 | st-tech/zr-obp | fix eng in synthetic_slate.ipynb |
641,011 | 17.05.2021 18:56:18 | -32,400 | cb7f3caf6d4e2f60d62321122f84aaf135f1bc62 | add decay_function | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/__init__.py",
"new_path": "obp/dataset/__init__.py",
"diff": "@@ -9,7 +9,7 @@ from obp.dataset.multiclass import MultiClassToBanditReduction\nfrom obp.dataset.synthetic_slate import SyntheticSlateBanditDataset\nfrom obp.dataset.synthetic_slate impo... | Python | Apache License 2.0 | st-tech/zr-obp | add decay_function |
641,011 | 18.05.2021 18:17:32 | -32,400 | 402140796514998e72291b443a83e42b9091c5bb | implement calc_ground_truth_policy_value function | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/synthetic_slate.py",
"new_path": "obp/dataset/synthetic_slate.py",
"diff": "@@ -316,10 +316,10 @@ class SyntheticSlateBanditDataset(BaseBanditDataset):\naction_interaction_weight_matrix[position_, position_] = 1\nreturn action_interaction_weight_ma... | Python | Apache License 2.0 | st-tech/zr-obp | implement calc_ground_truth_policy_value function |
641,011 | 20.05.2021 17:07:21 | -32,400 | d3638e83f79ec53669d47e9cb9db0606f905e559 | implement obtain_pscore_given_evaluation_policy_logit function | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/synthetic_slate.py",
"new_path": "obp/dataset/synthetic_slate.py",
"diff": "@@ -341,6 +341,65 @@ class SyntheticSlateBanditDataset(BaseBanditDataset):\n)\nreturn pscore_\n+ def obtain_pscore_given_evaluation_policy_logit(\n+ self,\n+ action: np.nda... | Python | Apache License 2.0 | st-tech/zr-obp | implement obtain_pscore_given_evaluation_policy_logit function |
641,011 | 23.05.2021 11:40:59 | -32,400 | 22a89152e61526db495e6504577721f19e4ab28a | add tests and minor fix | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/synthetic_slate.py",
"new_path": "obp/dataset/synthetic_slate.py",
"diff": "@@ -354,7 +354,7 @@ class SyntheticSlateBanditDataset(BaseBanditDataset):\naction: array-like, (n_rounds * len_list, )\nAction chosen by behavior policy.\n- evaluation_poli... | Python | Apache License 2.0 | st-tech/zr-obp | add tests and minor fix |
641,011 | 26.05.2021 16:04:19 | -32,400 | 14982edf3575c9f6d4835b6be5a635e97d659160 | add is_factorizable option | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/synthetic_slate.py",
"new_path": "obp/dataset/synthetic_slate.py",
"diff": "@@ -346,6 +346,7 @@ class SyntheticSlateBanditDataset(BaseBanditDataset):\naction: np.ndarray,\nevaluation_policy_logit_: np.ndarray,\nreturn_pscore_item_position: bool = T... | Python | Apache License 2.0 | st-tech/zr-obp | add is_factorizable option |
641,011 | 26.05.2021 19:36:48 | -32,400 | 059858ef0183024e2d03224e2e06e7c82569ddd8 | fix calc_ground_truth_policy_value | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/synthetic_slate.py",
"new_path": "obp/dataset/synthetic_slate.py",
"diff": "\"\"\"Class for Generating Synthetic Slate Logged Bandit Feedback.\"\"\"\nfrom dataclasses import dataclass\nfrom typing import Optional, Callable, Tuple, Union, List\n-fro... | Python | Apache License 2.0 | st-tech/zr-obp | fix calc_ground_truth_policy_value |
641,014 | 29.05.2021 08:52:36 | -32,400 | 60c31523957c35b12c9e5a5b5d3697f85661bc3c | make calc_pscore faster | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/synthetic_slate.py",
"new_path": "obp/dataset/synthetic_slate.py",
"diff": "@@ -321,30 +321,44 @@ class SyntheticSlateBanditDataset(BaseBanditDataset):\naction_interaction_weight_matrix[position_, position_] = 1\nreturn action_interaction_weight_ma... | Python | Apache License 2.0 | st-tech/zr-obp | make calc_pscore faster |
641,011 | 29.05.2021 16:25:42 | -32,400 | c95125a6f50560c701bd3c509964b57773631b58 | batch processing for calc_ground_truth_policy_value | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/synthetic_slate.py",
"new_path": "obp/dataset/synthetic_slate.py",
"diff": "@@ -831,7 +831,7 @@ class SyntheticSlateBanditDataset(BaseBanditDataset):\npscores.append(\nsoftmax(evaluation_policy_logit_)[:, action_list].prod(1)\n)\n- pscores = np.arr... | Python | Apache License 2.0 | st-tech/zr-obp | batch processing for calc_ground_truth_policy_value |
641,011 | 30.05.2021 07:12:56 | -32,400 | 586f1d467c22765db9371d3204f7c46fecd7f334 | faster calc_ground_truth_policy_value | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/synthetic_slate.py",
"new_path": "obp/dataset/synthetic_slate.py",
"diff": "@@ -300,7 +300,7 @@ class SyntheticSlateBanditDataset(BaseBanditDataset):\naction_interaction_weight_matrix[:, position_] = -self.decay_function(\nnp.abs(np.arange(len_list... | Python | Apache License 2.0 | st-tech/zr-obp | faster calc_ground_truth_policy_value |
641,005 | 30.05.2021 20:11:51 | -32,400 | 30dae82c3e493ebb625138b55eeff95798c2daa3 | change x lables | [
{
"change_type": "MODIFY",
"old_path": "obp/ope/meta.py",
"new_path": "obp/ope/meta.py",
"diff": "@@ -608,10 +608,6 @@ class OffPolicyEvaluation:\nestimated_round_rewards_df = DataFrame(\nestimated_round_rewards_dict[estimator_name]\n)\n- estimated_round_rewards_df.rename(\n- columns={key: key.upper... | Python | Apache License 2.0 | st-tech/zr-obp | change x lables |
641,011 | 31.05.2021 15:13:17 | -32,400 | 9463fb95680021319f757f3b2652d2efa05e345b | bug fix on batch processing | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/synthetic_slate.py",
"new_path": "obp/dataset/synthetic_slate.py",
"diff": "@@ -871,8 +871,9 @@ class SyntheticSlateBanditDataset(BaseBanditDataset):\nelse:\nn_batch = (\nn_rounds * n_enumerated_slate_actions * self.len_list - 1\n- ) // 10 ** 8 + 1... | Python | Apache License 2.0 | st-tech/zr-obp | bug fix on batch processing |
641,006 | 06.06.2021 19:53:13 | -32,400 | ddef53dcb93f7c86540e4907ce6ceea5f0566867 | add pscore mock test | [
{
"change_type": "MODIFY",
"old_path": "tests/dataset/test_synthetic_slate.py",
"new_path": "tests/dataset/test_synthetic_slate.py",
"diff": "@@ -2101,3 +2101,184 @@ def test_obtain_pscore_given_evaluation_policy_logit_value_check(\nelse bandit_feedback[\"pscore_item_position\"]\n== evaluation_polic... | Python | Apache License 2.0 | st-tech/zr-obp | add pscore mock test |
641,006 | 06.06.2021 20:01:42 | -32,400 | 9a98923c0365c6d9b984b0f4d58d424166b6c30f | fix bug of update params (reward_counts should not be the mean, but the count | [
{
"change_type": "MODIFY",
"old_path": "obp/policy/contextfree.py",
"new_path": "obp/policy/contextfree.py",
"diff": "@@ -89,8 +89,7 @@ class EpsilonGreedy(BaseContextFreePolicy):\n\"\"\"\nself.n_trial += 1\nself.action_counts_temp[action] += 1\n- n, old_reward = self.action_counts_temp[action], sel... | Python | Apache License 2.0 | st-tech/zr-obp | fix bug of update params (reward_counts should not be the mean, but the count |
641,006 | 06.06.2021 20:08:57 | -32,400 | 3722d00a8310b884f67212b81915fa58dd517d88 | delete action/pscore/reward/estimated_rewards_by_model from check_ope_inputs | [
{
"change_type": "MODIFY",
"old_path": "obp/utils.py",
"new_path": "obp/utils.py",
"diff": "@@ -301,8 +301,6 @@ def check_ope_inputs(\n# estimated_rewards_by_reg_model\nif estimated_rewards_by_reg_model is not None:\n- if not isinstance(estimated_rewards_by_reg_model, np.ndarray):\n- raise ValueErro... | Python | Apache License 2.0 | st-tech/zr-obp | delete action/pscore/reward/estimated_rewards_by_model from check_ope_inputs |
641,010 | 21.06.2021 23:36:15 | -28,800 | c773dfb8b11ef26e713822ef38cd743e1a02d8fd | docs: add buttons to open ipynb files in Colab | [
{
"change_type": "MODIFY",
"old_path": "README.md",
"new_path": "README.md",
"diff": "@@ -78,7 +78,7 @@ Open Bandit Pipeline consists of the following main modules.\n- [**dataset module**](./obp/dataset/): This module provides a data loader for Open Bandit Dataset and a flexible interface for handli... | Python | Apache License 2.0 | st-tech/zr-obp | docs: add buttons to open ipynb files in Colab |
641,006 | 27.06.2021 22:37:53 | -32,400 | 58332df77a41600f0e74fe616b44cfb2aa2b6462 | add faster method to calculate pscore item position | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/synthetic_slate.py",
"new_path": "obp/dataset/synthetic_slate.py",
"diff": "@@ -11,6 +11,7 @@ from scipy.stats import truncnorm\nfrom scipy.special import perm\nfrom sklearn.utils import check_random_state, check_scalar\nfrom tqdm import tqdm\n+fro... | Python | Apache License 2.0 | st-tech/zr-obp | add faster method to calculate pscore item position |
641,006 | 27.06.2021 22:38:29 | -32,400 | b0e8c698e99f65fa0a119cac1bcd2cce4e078a0f | add example of obtaining slate bandit feedback | [
{
"change_type": "ADD",
"old_path": null,
"new_path": "examples/synthetic/obtain_slate_bandit_feedback.py",
"diff": "+import argparse\n+\n+from obp.dataset import (\n+ logistic_reward_function,\n+ linear_behavior_policy_logit,\n+ SyntheticSlateBanditDataset,\n+)\n+\n+\n+if __name__ == \"__main__\":\... | Python | Apache License 2.0 | st-tech/zr-obp | add example of obtaining slate bandit feedback |
641,006 | 27.06.2021 23:26:53 | -32,400 | a4cfb2fdaf46594e49e7033bed8e8d5a35484e94 | add clip_logit_value to obtain_pscore_given_evaluation_policy_logit | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/synthetic_slate.py",
"new_path": "obp/dataset/synthetic_slate.py",
"diff": "@@ -405,6 +405,7 @@ class SyntheticSlateBanditDataset(BaseBanditDataset):\naction: np.ndarray,\nevaluation_policy_logit_: np.ndarray,\nreturn_pscore_item_position: bool = T... | Python | Apache License 2.0 | st-tech/zr-obp | add clip_logit_value to obtain_pscore_given_evaluation_policy_logit |
641,014 | 06.07.2021 13:29:57 | -32,400 | b1c817b0e1ff9c05104a1038cc5a362eaef4569f | implement SyntheticContinuousBanditDataset | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/__init__.py",
"new_path": "obp/dataset/__init__.py",
"diff": "@@ -6,6 +6,13 @@ from obp.dataset.synthetic import logistic_reward_function\nfrom obp.dataset.synthetic import linear_reward_function\nfrom obp.dataset.synthetic import linear_behavior_p... | Python | Apache License 2.0 | st-tech/zr-obp | implement SyntheticContinuousBanditDataset |
641,014 | 06.07.2021 13:30:24 | -32,400 | c77dd99fbc7e66d33473d2cbd5ecc2e93a4abc87 | add tests of synthetic_continuous.py | [
{
"change_type": "MODIFY",
"old_path": "tests/dataset/test_synthetic.py",
"new_path": "tests/dataset/test_synthetic.py",
"diff": "@@ -304,12 +304,12 @@ def test_synthetic_linear_behavior_policy():\naction_context = [1.0, 1.0]\nlinear_behavior_policy(context=np.ones([2, 2]), action_context=action_con... | Python | Apache License 2.0 | st-tech/zr-obp | add tests of synthetic_continuous.py |
641,014 | 06.07.2021 23:08:30 | -32,400 | c0874d4284a71852a5eca985a78957cb8f491bd9 | implement continuous ope estimators | [
{
"change_type": "MODIFY",
"old_path": "obp/ope/__init__.py",
"new_path": "obp/ope/__init__.py",
"diff": "@@ -10,8 +10,18 @@ from obp.ope.estimators import DoublyRobustWithShrinkage\nfrom obp.ope.estimators_slate import SlateStandardIPS\nfrom obp.ope.estimators_slate import SlateIndependentIPS\nfrom... | Python | Apache License 2.0 | st-tech/zr-obp | implement continuous ope estimators |
641,014 | 06.07.2021 23:09:03 | -32,400 | bf2eaf96bde790d4e46fa0240b526eb8ad8294ee | add tests of continuous ope estimators | [
{
"change_type": "ADD",
"old_path": null,
"new_path": "tests/ope/test_dr_estimators_continuous.py",
"diff": "+import pytest\n+import numpy as np\n+\n+from obp.ope import KernelizedDoublyRobust\n+from obp.dataset import (\n+ SyntheticContinuousBanditDataset,\n+ linear_reward_funcion_continuous,\n+ li... | Python | Apache License 2.0 | st-tech/zr-obp | add tests of continuous ope estimators |
641,014 | 06.07.2021 23:09:21 | -32,400 | ea3b2f915bacfd8dc91155a12ce702f784d924fb | add some check funcs for continuous ope | [
{
"change_type": "MODIFY",
"old_path": "obp/utils.py",
"new_path": "obp/utils.py",
"diff": "@@ -331,6 +331,150 @@ def check_ope_inputs(\nraise ValueError(\"pscore must be positive\")\n+def check_continuous_bandit_feedback_inputs(\n+ context: np.ndarray,\n+ action_by_behavior_policy: np.ndarray,\n+ r... | Python | Apache License 2.0 | st-tech/zr-obp | add some check funcs for continuous ope |
641,014 | 07.07.2021 10:18:30 | -32,400 | ce63f55866717e0155faf190336ae4f99176e3ad | add tests of meta_continuous | [
{
"change_type": "ADD",
"old_path": null,
"new_path": "tests/ope/test_meta_continuous.py",
"diff": "+from typing import Dict, Optional\n+from dataclasses import dataclass\n+import itertools\n+from copy import deepcopy\n+\n+import pytest\n+import numpy as np\n+import pandas as pd\n+from pandas.testin... | Python | Apache License 2.0 | st-tech/zr-obp | add tests of meta_continuous |
641,014 | 07.07.2021 10:19:34 | -32,400 | 6ee2fde20ad9ed66157aa9a441787a4129877772 | add synthetic_continuous_bandit_feedback | [
{
"change_type": "MODIFY",
"old_path": "tests/ope/conftest.py",
"new_path": "tests/ope/conftest.py",
"diff": "@@ -13,6 +13,7 @@ from obp.dataset import (\nlogistic_reward_function,\nlinear_behavior_policy,\nSyntheticSlateBanditDataset,\n+ SyntheticContinuousBanditDataset,\n)\nfrom obp.utils import s... | Python | Apache License 2.0 | st-tech/zr-obp | add synthetic_continuous_bandit_feedback |
641,014 | 07.07.2021 21:06:23 | -32,400 | 1e04f89055281813a3e3d82c338a879af666dd80 | add tests of ContinuousNNPolicyLearner | [
{
"change_type": "ADD",
"old_path": null,
"new_path": "tests/policy/test_offline_continuous.py",
"diff": "+import pytest\n+import numpy as np\n+import torch\n+\n+from obp.policy.offline_continuous import ContinuousNNPolicyLearner\n+\n+\n+# dim_context, pg_method, bandwidth, output_space, hidden_laye... | Python | Apache License 2.0 | st-tech/zr-obp | add tests of ContinuousNNPolicyLearner |
641,014 | 07.07.2021 21:40:50 | -32,400 | d92a4762e40ec08e04374770f7515241e5e92fff | add some check functions | [
{
"change_type": "MODIFY",
"old_path": "obp/utils.py",
"new_path": "obp/utils.py",
"diff": "@@ -331,6 +331,158 @@ def check_ope_inputs(\nraise ValueError(\"pscore must be positive\")\n+def check_continuous_bandit_feedback_inputs(\n+ context: np.ndarray,\n+ action_by_behavior_policy: np.ndarray,\n+ r... | Python | Apache License 2.0 | st-tech/zr-obp | add some check functions |
641,014 | 08.07.2021 10:39:27 | -32,400 | 8f5884ba6166f8e5b2e8615438b50a6e6e95d600 | move some arguments to init | [
{
"change_type": "MODIFY",
"old_path": "obp/dataset/synthetic_continuous.py",
"new_path": "obp/dataset/synthetic_continuous.py",
"diff": "@@ -28,6 +28,18 @@ class SyntheticContinuousBanditDataset(BaseBanditDataset):\ndim_context: int, default=1\nNumber of dimensions of context vectors.\n+ action_noi... | Python | Apache License 2.0 | st-tech/zr-obp | move some arguments to init |
641,014 | 08.07.2021 10:39:44 | -32,400 | 84e74c7e0de9c8c6d86b246ad75e9f3f305f705f | fix tests of SyntheticContinuousBanditDataset | [
{
"change_type": "MODIFY",
"old_path": "tests/dataset/test_synthetic_continuous.py",
"new_path": "tests/dataset/test_synthetic_continuous.py",
"diff": "@@ -12,35 +12,17 @@ from obp.dataset.synthetic_continuous import (\n)\n-def test_synthetic_continuous_init():\n- # dim_context\n- with pytest.raises... | Python | Apache License 2.0 | st-tech/zr-obp | fix tests of SyntheticContinuousBanditDataset |
641,014 | 08.07.2021 11:04:49 | -32,400 | 0582f84369024a87e189e95cf044d2cd45b8f4b0 | fix some tests to adjust the changes of SyntheticContinuousBanditDataset | [
{
"change_type": "MODIFY",
"old_path": "tests/ope/conftest.py",
"new_path": "tests/ope/conftest.py",
"diff": "@@ -72,14 +72,12 @@ def synthetic_continuous_bandit_feedback() -> BanditFeedback:\nmax_action_value = 10\ndataset = SyntheticContinuousBanditDataset(\ndim_context=dim_context,\n- random_stat... | Python | Apache License 2.0 | st-tech/zr-obp | fix some tests to adjust the changes of SyntheticContinuousBanditDataset |
641,014 | 08.07.2021 11:12:53 | -32,400 | 900c4ffa4b66e35c14f7580328e95be9af7ab208 | fix some tests to adjust to the changes of SyntheticContinuousBanditDataset | [
{
"change_type": "MODIFY",
"old_path": "tests/policy/test_offline_learner_continuous_performance.py",
"new_path": "tests/policy/test_offline_learner_continuous_performance.py",
"diff": "@@ -71,6 +71,10 @@ def test_offline_nn_policy_learner_performance(\n# synthetic data generator\ndataset = Syntheti... | Python | Apache License 2.0 | st-tech/zr-obp | fix some tests to adjust to the changes of SyntheticContinuousBanditDataset |
641,014 | 08.07.2021 12:47:50 | -32,400 | bfa4466de043aa6c053424882ac06de10fa578c6 | implement weight clipping and fix some typos | [
{
"change_type": "MODIFY",
"old_path": "obp/ope/estimators.py",
"new_path": "obp/ope/estimators.py",
"diff": "@@ -8,6 +8,7 @@ from typing import Dict, Optional, Union\nimport numpy as np\nimport torch\n+from sklearn.utils import check_scalar\nfrom ..utils import (\nestimate_confidence_interval_by_bo... | Python | Apache License 2.0 | st-tech/zr-obp | implement weight clipping and fix some typos |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.