diff --git a/preview_pr/_downloads/07fcc19ba03226cd3d83d4e40ec44385/auto_examples_python.zip b/preview_pr/_downloads/07fcc19ba03226cd3d83d4e40ec44385/auto_examples_python.zip index f24e70378..ed060a155 100644 Binary files a/preview_pr/_downloads/07fcc19ba03226cd3d83d4e40ec44385/auto_examples_python.zip and b/preview_pr/_downloads/07fcc19ba03226cd3d83d4e40ec44385/auto_examples_python.zip differ diff --git a/preview_pr/_downloads/6f1e7a639e0699d6164445b55e6c116d/auto_examples_jupyter.zip b/preview_pr/_downloads/6f1e7a639e0699d6164445b55e6c116d/auto_examples_jupyter.zip index bd5356975..053171d79 100644 Binary files a/preview_pr/_downloads/6f1e7a639e0699d6164445b55e6c116d/auto_examples_jupyter.zip and b/preview_pr/_downloads/6f1e7a639e0699d6164445b55e6c116d/auto_examples_jupyter.zip differ diff --git a/preview_pr/_images/sphx_glr_plot_TS_bandit_001.png b/preview_pr/_images/sphx_glr_plot_TS_bandit_001.png index 2662475ba..75d2f1949 100644 Binary files a/preview_pr/_images/sphx_glr_plot_TS_bandit_001.png and b/preview_pr/_images/sphx_glr_plot_TS_bandit_001.png differ diff --git a/preview_pr/_images/sphx_glr_plot_TS_bandit_002.png b/preview_pr/_images/sphx_glr_plot_TS_bandit_002.png index 0033810a4..cd5f044cf 100644 Binary files a/preview_pr/_images/sphx_glr_plot_TS_bandit_002.png and b/preview_pr/_images/sphx_glr_plot_TS_bandit_002.png differ diff --git a/preview_pr/_images/sphx_glr_plot_TS_bandit_thumb.png b/preview_pr/_images/sphx_glr_plot_TS_bandit_thumb.png index 7bd8a8c64..27318e3c0 100644 Binary files a/preview_pr/_images/sphx_glr_plot_TS_bandit_thumb.png and b/preview_pr/_images/sphx_glr_plot_TS_bandit_thumb.png differ diff --git a/preview_pr/_images/sphx_glr_plot_agent_manager_001.png b/preview_pr/_images/sphx_glr_plot_agent_manager_001.png index 029a0feda..e43f70166 100644 Binary files a/preview_pr/_images/sphx_glr_plot_agent_manager_001.png and b/preview_pr/_images/sphx_glr_plot_agent_manager_001.png differ diff --git a/preview_pr/_images/sphx_glr_plot_agent_manager_thumb.png b/preview_pr/_images/sphx_glr_plot_agent_manager_thumb.png index 255ac346e..51e9c14cb 100644 Binary files a/preview_pr/_images/sphx_glr_plot_agent_manager_thumb.png and b/preview_pr/_images/sphx_glr_plot_agent_manager_thumb.png differ diff --git a/preview_pr/_images/sphx_glr_plot_exp3_bandit_001.png b/preview_pr/_images/sphx_glr_plot_exp3_bandit_001.png index 41b83bf73..d22fd9a0c 100644 Binary files a/preview_pr/_images/sphx_glr_plot_exp3_bandit_001.png and b/preview_pr/_images/sphx_glr_plot_exp3_bandit_001.png differ diff --git a/preview_pr/_images/sphx_glr_plot_exp3_bandit_thumb.png b/preview_pr/_images/sphx_glr_plot_exp3_bandit_thumb.png index 532a2dea1..9686512f4 100644 Binary files a/preview_pr/_images/sphx_glr_plot_exp3_bandit_thumb.png and b/preview_pr/_images/sphx_glr_plot_exp3_bandit_thumb.png differ diff --git a/preview_pr/_images/sphx_glr_plot_mirror_bandit_001.png b/preview_pr/_images/sphx_glr_plot_mirror_bandit_001.png index e7be1aa0d..efb7c8f44 100644 Binary files a/preview_pr/_images/sphx_glr_plot_mirror_bandit_001.png and b/preview_pr/_images/sphx_glr_plot_mirror_bandit_001.png differ diff --git a/preview_pr/_images/sphx_glr_plot_mirror_bandit_thumb.png b/preview_pr/_images/sphx_glr_plot_mirror_bandit_thumb.png index e7fcce3ac..6bba3c904 100644 Binary files a/preview_pr/_images/sphx_glr_plot_mirror_bandit_thumb.png and b/preview_pr/_images/sphx_glr_plot_mirror_bandit_thumb.png differ diff --git a/preview_pr/_images/sphx_glr_plot_smooth_001.png b/preview_pr/_images/sphx_glr_plot_smooth_001.png index 8d37935bd..7691eb761 100644 Binary files a/preview_pr/_images/sphx_glr_plot_smooth_001.png and b/preview_pr/_images/sphx_glr_plot_smooth_001.png differ diff --git a/preview_pr/_images/sphx_glr_plot_smooth_002.png b/preview_pr/_images/sphx_glr_plot_smooth_002.png index ee1ef8f0a..5eba80116 100644 Binary files a/preview_pr/_images/sphx_glr_plot_smooth_002.png and b/preview_pr/_images/sphx_glr_plot_smooth_002.png differ diff --git a/preview_pr/_images/sphx_glr_plot_smooth_thumb.png b/preview_pr/_images/sphx_glr_plot_smooth_thumb.png index a9d75591c..d726af1da 100644 Binary files a/preview_pr/_images/sphx_glr_plot_smooth_thumb.png and b/preview_pr/_images/sphx_glr_plot_smooth_thumb.png differ diff --git a/preview_pr/_images/sphx_glr_plot_ucb_bandit_001.png b/preview_pr/_images/sphx_glr_plot_ucb_bandit_001.png index 5298c5773..73466d1c2 100644 Binary files a/preview_pr/_images/sphx_glr_plot_ucb_bandit_001.png and b/preview_pr/_images/sphx_glr_plot_ucb_bandit_001.png differ diff --git a/preview_pr/_images/sphx_glr_plot_ucb_bandit_thumb.png b/preview_pr/_images/sphx_glr_plot_ucb_bandit_thumb.png index 26c01fddf..878c1389c 100644 Binary files a/preview_pr/_images/sphx_glr_plot_ucb_bandit_thumb.png and b/preview_pr/_images/sphx_glr_plot_ucb_bandit_thumb.png differ diff --git a/preview_pr/_images/sphx_glr_plot_writer_wrapper_001.png b/preview_pr/_images/sphx_glr_plot_writer_wrapper_001.png index 9c3cdaf9b..aea8c4791 100644 Binary files a/preview_pr/_images/sphx_glr_plot_writer_wrapper_001.png and b/preview_pr/_images/sphx_glr_plot_writer_wrapper_001.png differ diff --git a/preview_pr/_images/sphx_glr_plot_writer_wrapper_002.png b/preview_pr/_images/sphx_glr_plot_writer_wrapper_002.png index 29e10a4eb..81cec3960 100644 Binary files a/preview_pr/_images/sphx_glr_plot_writer_wrapper_002.png and b/preview_pr/_images/sphx_glr_plot_writer_wrapper_002.png differ diff --git a/preview_pr/_images/sphx_glr_plot_writer_wrapper_thumb.png b/preview_pr/_images/sphx_glr_plot_writer_wrapper_thumb.png index ba707a941..870c3b106 100644 Binary files a/preview_pr/_images/sphx_glr_plot_writer_wrapper_thumb.png and b/preview_pr/_images/sphx_glr_plot_writer_wrapper_thumb.png differ diff --git a/preview_pr/_modules/rlberry/manager/utils.html b/preview_pr/_modules/rlberry/manager/utils.html index 765fa1bcc..ed011f8c7 100644 --- a/preview_pr/_modules/rlberry/manager/utils.html +++ b/preview_pr/_modules/rlberry/manager/utils.html @@ -137,9 +137,26 @@

Source code for rlberry.manager.utils

 
 
[docs]def tensorboard_folder_to_dataframe(path_to_tensorboard_data):
     """
-    path_to_tensorboard_data : path to the tensorboard data. It must be the parent folder of all the training, and the event have to be in this kind of path : <path_to_tensorboard_data/algo_name/n_simu/events.out.tfevents.xxxxx>
-
-    Return a dict of panda dataframe (key = tag, value = panda.dataframe)
+    Function to convert 'tensorboard log' to 'Panda DataFrames'
+
+    To convert the 'tensorboard log', the input must be must be the path to "the parent folder of all the training log" (path_to_tensorboard_data), and the 'events.out.tfevents' files have to be in this kind of path :
+      < path_to_tensorboard_data/algo_name/n_simu/events.out.tfevents.xxxxx >
+
+    The output format is a dictionary.
+    key = tag (type of data)
+    value = Panda DataFrame with the following structure (4 column) :
+        "name" = algo_name
+        "n_simu" = n_simu (seed)
+        "x" = step number
+        "y" = value of the data
+
+    Parameters
+    ----------
+    path_to_tensorboard_data : path to the parent folder of the tensorboard's data.
+
+    Returns
+    -------
+    Dict : dict of Panda DataFrame (key = tag, value = Panda.DataFrame)
     """
     from tensorboard.backend.event_processing import event_accumulator
 
diff --git a/preview_pr/_sources/auto_examples/demo_bandits/plot_TS_bandit.rst.txt b/preview_pr/_sources/auto_examples/demo_bandits/plot_TS_bandit.rst.txt
index 5a9fe0de3..c4e6b3f58 100644
--- a/preview_pr/_sources/auto_examples/demo_bandits/plot_TS_bandit.rst.txt
+++ b/preview_pr/_sources/auto_examples/demo_bandits/plot_TS_bandit.rst.txt
@@ -54,21 +54,21 @@ For the Gaussian case, we use a Gaussian prior and compare it to a sub-Gaussian
 
  .. code-block:: none
 
-    [INFO] 10:02: ... trained! 
-    [INFO] 10:02: Saved ExperimentManager(Bounded UCB Agent) using pickle. 
-    [INFO] 10:02: The ExperimentManager was saved in : 'rlberry_data/temp/manager_data/Bounded UCB Agent_2024-10-16_10-02-15_bd80da50/manager_obj.pickle' 
-    [INFO] 10:02: Running ExperimentManager fit() for Bernoulli TS Agent with n_fit = 10 and max_workers = None. 
-    [INFO] 10:02: ... trained! 
-    [INFO] 10:02: Saved ExperimentManager(Bernoulli TS Agent) using pickle. 
-    [INFO] 10:02: The ExperimentManager was saved in : 'rlberry_data/temp/manager_data/Bernoulli TS Agent_2024-10-16_10-02-15_847030c5/manager_obj.pickle' 
-    [INFO] 10:02: Running ExperimentManager fit() for Gaussian UCB Agent with n_fit = 10 and max_workers = None. 
-    [INFO] 10:02: ... trained! 
-    [INFO] 10:02: Saved ExperimentManager(Gaussian UCB Agent) using pickle. 
-    [INFO] 10:02: The ExperimentManager was saved in : 'rlberry_data/temp/manager_data/Gaussian UCB Agent_2024-10-16_10-02-26_8ade7c97/manager_obj.pickle' 
-    [INFO] 10:02: Running ExperimentManager fit() for Gaussian TS Agent with n_fit = 10 and max_workers = None. 
-    [INFO] 10:02: ... trained! 
-    [INFO] 10:02: Saved ExperimentManager(Gaussian TS Agent) using pickle. 
-    [INFO] 10:02: The ExperimentManager was saved in : 'rlberry_data/temp/manager_data/Gaussian TS Agent_2024-10-16_10-02-26_ae7d0a45/manager_obj.pickle' 
+    [INFO] 12:45: ... trained! 
+    [INFO] 12:45: Saved ExperimentManager(Bounded UCB Agent) using pickle. 
+    [INFO] 12:45: The ExperimentManager was saved in : 'rlberry_data/temp/manager_data/Bounded UCB Agent_2024-10-16_12-45-04_319cb7e6/manager_obj.pickle' 
+    [INFO] 12:45: Running ExperimentManager fit() for Bernoulli TS Agent with n_fit = 10 and max_workers = None. 
+    [INFO] 12:45: ... trained! 
+    [INFO] 12:45: Saved ExperimentManager(Bernoulli TS Agent) using pickle. 
+    [INFO] 12:45: The ExperimentManager was saved in : 'rlberry_data/temp/manager_data/Bernoulli TS Agent_2024-10-16_12-45-04_279a356f/manager_obj.pickle' 
+    [INFO] 12:45: Running ExperimentManager fit() for Gaussian UCB Agent with n_fit = 10 and max_workers = None. 
+    [INFO] 12:45: ... trained! 
+    [INFO] 12:45: Saved ExperimentManager(Gaussian UCB Agent) using pickle. 
+    [INFO] 12:45: The ExperimentManager was saved in : 'rlberry_data/temp/manager_data/Gaussian UCB Agent_2024-10-16_12-45-15_f7d90f4f/manager_obj.pickle' 
+    [INFO] 12:45: Running ExperimentManager fit() for Gaussian TS Agent with n_fit = 10 and max_workers = None. 
+    [INFO] 12:45: ... trained! 
+    [INFO] 12:45: Saved ExperimentManager(Gaussian TS Agent) using pickle. 
+    [INFO] 12:45: The ExperimentManager was saved in : 'rlberry_data/temp/manager_data/Gaussian TS Agent_2024-10-16_12-45-15_9c0d4b7a/manager_obj.pickle' 
 
 
 
@@ -224,7 +224,7 @@ For the Gaussian case, we use a Gaussian prior and compare it to a sub-Gaussian
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** (0 minutes 20.469 seconds)
+   **Total running time of the script:** (0 minutes 20.350 seconds)
 
 
 .. _sphx_glr_download_auto_examples_demo_bandits_plot_TS_bandit.py:
diff --git a/preview_pr/_sources/auto_examples/demo_bandits/plot_compare_index_bandits.rst.txt b/preview_pr/_sources/auto_examples/demo_bandits/plot_compare_index_bandits.rst.txt
index b26f44423..8b7aff3a1 100644
--- a/preview_pr/_sources/auto_examples/demo_bandits/plot_compare_index_bandits.rst.txt
+++ b/preview_pr/_sources/auto_examples/demo_bandits/plot_compare_index_bandits.rst.txt
@@ -276,7 +276,7 @@ how to use subplots in with `plot_writer_data`
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** (0 minutes 41.147 seconds)
+   **Total running time of the script:** (0 minutes 41.099 seconds)
 
 
 .. _sphx_glr_download_auto_examples_demo_bandits_plot_compare_index_bandits.py:
diff --git a/preview_pr/_sources/auto_examples/demo_bandits/plot_exp3_bandit.rst.txt b/preview_pr/_sources/auto_examples/demo_bandits/plot_exp3_bandit.rst.txt
index 7b8250720..a371f0460 100644
--- a/preview_pr/_sources/auto_examples/demo_bandits/plot_exp3_bandit.rst.txt
+++ b/preview_pr/_sources/auto_examples/demo_bandits/plot_exp3_bandit.rst.txt
@@ -155,7 +155,7 @@ randomized algorithm.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** (0 minutes 7.648 seconds)
+   **Total running time of the script:** (0 minutes 7.662 seconds)
 
 
 .. _sphx_glr_download_auto_examples_demo_bandits_plot_exp3_bandit.py:
diff --git a/preview_pr/_sources/auto_examples/demo_bandits/plot_mirror_bandit.rst.txt b/preview_pr/_sources/auto_examples/demo_bandits/plot_mirror_bandit.rst.txt
index 3477d781e..d5ff6c204 100644
--- a/preview_pr/_sources/auto_examples/demo_bandits/plot_mirror_bandit.rst.txt
+++ b/preview_pr/_sources/auto_examples/demo_bandits/plot_mirror_bandit.rst.txt
@@ -45,32 +45,32 @@ and finally definition of the experiment.
 
  .. code-block:: none
 
-    [INFO] 10:03:        agent_name  worker  reward  action  max_global_step
-                             SH        0     -0.821    2           4 
-    [INFO] 10:03:        agent_name  worker  reward  action  max_global_step
-                             SH        0     -0.465    7           9 
-    [INFO] 10:03:        agent_name  worker  reward  action  max_global_step
-                             SH        0     -0.459    7           17 
-    [INFO] 10:03:        agent_name  worker  reward  action  max_global_step
-                             SH        0     -0.464    7           25 
-    [INFO] 10:03:        agent_name  worker  reward  action  max_global_step
-                             SH        0     -0.314    7           33 
-    [INFO] 10:03:        agent_name  worker  reward  action  max_global_step
-                             SH        0     -0.301    3           43 
-    [INFO] 10:03:        agent_name  worker  reward  action  max_global_step
-                             SH        0     -0.313    6           53 
-    [INFO] 10:03:        agent_name  worker  reward  action  max_global_step
-                             SH        0     -0.306    3           63 
-    [INFO] 10:03:        agent_name  worker  reward  action  max_global_step
-                             SH        0     -0.308    6           73 
-    [INFO] 10:03:        agent_name  worker  reward  action  max_global_step
-                             SH        0     -0.304    6           83 
-    [INFO] 10:03:        agent_name  worker  reward  action  max_global_step
-                             SH        0     -0.307    6           93 
-    [INFO] 10:03: ... trained! 
-    [INFO] 10:03: Saved ExperimentManager(SH) using pickle. 
-    [INFO] 10:03: The ExperimentManager was saved in : 'rlberry_data/temp/manager_data/SH_2024-10-16_10-03-17_feb7766d/manager_obj.pickle' 
-    The optimal action (fastest server) is server number  7
+    [INFO] 12:46:        agent_name  worker  reward  action  max_global_step
+                             SH        0     -0.625    3           5 
+    [INFO] 12:46:        agent_name  worker  reward  action  max_global_step
+                             SH        0     -0.458    1           11 
+    [INFO] 12:46:        agent_name  worker  reward  action  max_global_step
+                             SH        0     -0.306    2           20 
+    [INFO] 12:46:        agent_name  worker  reward  action  max_global_step
+                             SH        0     -0.61     1           27 
+    [INFO] 12:46:        agent_name  worker  reward  action  max_global_step
+                             SH        0     -0.301    6           36 
+    [INFO] 12:46:        agent_name  worker  reward  action  max_global_step
+                             SH        0     -0.309    4           46 
+    [INFO] 12:46:        agent_name  worker  reward  action  max_global_step
+                             SH        0     -0.303    6           56 
+    [INFO] 12:46:        agent_name  worker  reward  action  max_global_step
+                             SH        0     -0.306    7           66 
+    [INFO] 12:46:        agent_name  worker  reward  action  max_global_step
+                             SH        0     -0.303    6           73 
+    [INFO] 12:46:        agent_name  worker  reward  action  max_global_step
+                             SH        0     -0.302    6           83 
+    [INFO] 12:46:        agent_name  worker  reward  action  max_global_step
+                             SH        0     -0.301    6           93 
+    [INFO] 12:46: ... trained! 
+    [INFO] 12:46: Saved ExperimentManager(SH) using pickle. 
+    [INFO] 12:46: The ExperimentManager was saved in : 'rlberry_data/temp/manager_data/SH_2024-10-16_12-46-06_13e762a8/manager_obj.pickle' 
+    The optimal action (fastest server) is server number  8
 
 
 
@@ -251,7 +251,7 @@ and finally definition of the experiment.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** (0 minutes 37.576 seconds)
+   **Total running time of the script:** (0 minutes 36.308 seconds)
 
 
 .. _sphx_glr_download_auto_examples_demo_bandits_plot_mirror_bandit.py:
diff --git a/preview_pr/_sources/auto_examples/demo_bandits/plot_ucb_bandit.rst.txt b/preview_pr/_sources/auto_examples/demo_bandits/plot_ucb_bandit.rst.txt
index e6a64390b..f5865fdab 100644
--- a/preview_pr/_sources/auto_examples/demo_bandits/plot_ucb_bandit.rst.txt
+++ b/preview_pr/_sources/auto_examples/demo_bandits/plot_ucb_bandit.rst.txt
@@ -105,7 +105,7 @@ This script shows how to define a bandit environment and an UCB Index-based algo
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** (0 minutes 4.924 seconds)
+   **Total running time of the script:** (0 minutes 4.902 seconds)
 
 
 .. _sphx_glr_download_auto_examples_demo_bandits_plot_ucb_bandit.py:
diff --git a/preview_pr/_sources/auto_examples/demo_bandits/sg_execution_times.rst.txt b/preview_pr/_sources/auto_examples/demo_bandits/sg_execution_times.rst.txt
index d6cbd8e17..0752b60e9 100644
--- a/preview_pr/_sources/auto_examples/demo_bandits/sg_execution_times.rst.txt
+++ b/preview_pr/_sources/auto_examples/demo_bandits/sg_execution_times.rst.txt
@@ -6,16 +6,16 @@
 
 Computation times
 =================
-**01:51.763** total execution time for **auto_examples_demo_bandits** files:
+**01:50.320** total execution time for **auto_examples_demo_bandits** files:
 
 +--------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_auto_examples_demo_bandits_plot_compare_index_bandits.py` (``plot_compare_index_bandits.py``) | 00:41.147 | 0.0 MB |
+| :ref:`sphx_glr_auto_examples_demo_bandits_plot_compare_index_bandits.py` (``plot_compare_index_bandits.py``) | 00:41.099 | 0.0 MB |
 +--------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_auto_examples_demo_bandits_plot_mirror_bandit.py` (``plot_mirror_bandit.py``)                 | 00:37.576 | 0.0 MB |
+| :ref:`sphx_glr_auto_examples_demo_bandits_plot_mirror_bandit.py` (``plot_mirror_bandit.py``)                 | 00:36.308 | 0.0 MB |
 +--------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_auto_examples_demo_bandits_plot_TS_bandit.py` (``plot_TS_bandit.py``)                         | 00:20.469 | 0.0 MB |
+| :ref:`sphx_glr_auto_examples_demo_bandits_plot_TS_bandit.py` (``plot_TS_bandit.py``)                         | 00:20.350 | 0.0 MB |
 +--------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_auto_examples_demo_bandits_plot_exp3_bandit.py` (``plot_exp3_bandit.py``)                     | 00:07.648 | 0.0 MB |
+| :ref:`sphx_glr_auto_examples_demo_bandits_plot_exp3_bandit.py` (``plot_exp3_bandit.py``)                     | 00:07.662 | 0.0 MB |
 +--------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_auto_examples_demo_bandits_plot_ucb_bandit.py` (``plot_ucb_bandit.py``)                       | 00:04.924 | 0.0 MB |
+| :ref:`sphx_glr_auto_examples_demo_bandits_plot_ucb_bandit.py` (``plot_ucb_bandit.py``)                       | 00:04.902 | 0.0 MB |
 +--------------------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/preview_pr/_sources/auto_examples/plot_agent_manager.rst.txt b/preview_pr/_sources/auto_examples/plot_agent_manager.rst.txt
index ed3725e70..63fa51f16 100644
--- a/preview_pr/_sources/auto_examples/plot_agent_manager.rst.txt
+++ b/preview_pr/_sources/auto_examples/plot_agent_manager.rst.txt
@@ -49,18 +49,18 @@ Finally, we compare with a baseline provided by a random policy using the Experi
 
  .. code-block:: none
 
-    [INFO] 10:01: ... trained! 
-    [INFO] 10:01: Saved ExperimentManager(ValueIterationAgent) using pickle. 
-    [INFO] 10:01: The ExperimentManager was saved in : 'rlberry_data/temp/manager_data/ValueIterationAgent_2024-10-16_10-01-58_8d70b7e4/manager_obj.pickle' 
-    [INFO] 10:01: Running ExperimentManager fit() for RandomAgent with n_fit = 1 and max_workers = None. 
-    [INFO] 10:01: ... trained! 
-    [INFO] 10:01: Saved ExperimentManager(RandomAgent) using pickle. 
-    [INFO] 10:01: The ExperimentManager was saved in : 'rlberry_data/temp/manager_data/RandomAgent_2024-10-16_10-01-59_00aad013/manager_obj.pickle' 
-    [INFO] 10:01: Evaluating ValueIterationAgent... 
-    [INFO] 10:01: Computing 10 evaluations. 
+    [INFO] 12:44: ... trained! 
+    [INFO] 12:44: Saved ExperimentManager(ValueIterationAgent) using pickle. 
+    [INFO] 12:44: The ExperimentManager was saved in : 'rlberry_data/temp/manager_data/ValueIterationAgent_2024-10-16_12-44-47_80b8b7fe/manager_obj.pickle' 
+    [INFO] 12:44: Running ExperimentManager fit() for RandomAgent with n_fit = 1 and max_workers = None. 
+    [INFO] 12:44: ... trained! 
+    [INFO] 12:44: Saved ExperimentManager(RandomAgent) using pickle. 
+    [INFO] 12:44: The ExperimentManager was saved in : 'rlberry_data/temp/manager_data/RandomAgent_2024-10-16_12-44-48_7ba13ec9/manager_obj.pickle' 
+    [INFO] 12:44: Evaluating ValueIterationAgent... 
+    [INFO] 12:44: Computing 10 evaluations. 
     [INFO] Evaluation:..........  Evaluation finished 
-    [INFO] 10:01: Evaluating RandomAgent... 
-    [INFO] 10:01: Computing 10 evaluations. 
+    [INFO] 12:44: Evaluating RandomAgent... 
+    [INFO] 12:44: Computing 10 evaluations. 
     [INFO] Evaluation:..........  Evaluation finished 
 
 
@@ -186,7 +186,7 @@ Finally, we compare with a baseline provided by a random policy using the Experi
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** (0 minutes 1.233 seconds)
+   **Total running time of the script:** (0 minutes 1.153 seconds)
 
 
 .. _sphx_glr_download_auto_examples_plot_agent_manager.py:
diff --git a/preview_pr/_sources/auto_examples/plot_checkpointing.rst.txt b/preview_pr/_sources/auto_examples/plot_checkpointing.rst.txt
index 6024042d0..b890ce836 100644
--- a/preview_pr/_sources/auto_examples/plot_checkpointing.rst.txt
+++ b/preview_pr/_sources/auto_examples/plot_checkpointing.rst.txt
@@ -41,37 +41,37 @@ your agents, and how to restore from a previous checkpoint.
 
  .. code-block:: none
 
-    [INFO] 10:02: ... trained! 
-    [INFO] 10:02: Saved ExperimentManager(my-agent) using pickle. 
-    [INFO] 10:02: The ExperimentManager was saved in : 'rlberry_data/temp/manager_data/my-agent_2024-10-16_10-02-00_9aa9c456/manager_obj.pickle' 
-    [INFO] 10:02: Saved ExperimentManager(my-agent) using pickle. 
-    [INFO] 10:02: The ExperimentManager was saved in : 'rlberry_data/temp/manager_data/my-agent_2024-10-16_10-02-00_9aa9c456/manager_obj.pickle' 
-
-     Saved manager at rlberry_data/temp/manager_data/my-agent_2024-10-16_10-02-00_9aa9c456/manager_obj.pickle.
-
-    [INFO] 10:02: Running ExperimentManager fit() for my-agent with n_fit = 2 and max_workers = None. 
-    checkpoint at rlberry_data/temp/manager_data/my-agent_2024-10-16_10-02-00_9aa9c456/output_1/checkpoint.pickle (timestep = 500)
-    checkpoint at rlberry_data/temp/manager_data/my-agent_2024-10-16_10-02-00_9aa9c456/output_0/checkpoint.pickle (timestep = 500)
-    checkpoint at rlberry_data/temp/manager_data/my-agent_2024-10-16_10-02-00_9aa9c456/output_1/checkpoint.pickle (timestep = 1000)
-    checkpoint at rlberry_data/temp/manager_data/my-agent_2024-10-16_10-02-00_9aa9c456/output_0/checkpoint.pickle (timestep = 1000)
-    [INFO] 10:02: ... trained! 
-    [INFO] 10:02: Saved ExperimentManager(my-agent) using pickle. 
-    [INFO] 10:02: The ExperimentManager was saved in : 'rlberry_data/temp/manager_data/my-agent_2024-10-16_10-02-00_9aa9c456/manager_obj.pickle' 
-
-     Loading manager from rlberry_data/temp/manager_data/my-agent_2024-10-16_10-02-00_9aa9c456/manager_obj.pickle.
-
-    [INFO] 10:02: Running ExperimentManager fit() for my-agent with n_fit = 2 and max_workers = None. 
+    [INFO] 12:44: ... trained! 
+    [INFO] 12:44: Saved ExperimentManager(my-agent) using pickle. 
+    [INFO] 12:44: The ExperimentManager was saved in : 'rlberry_data/temp/manager_data/my-agent_2024-10-16_12-44-49_e91ad515/manager_obj.pickle' 
+    [INFO] 12:44: Saved ExperimentManager(my-agent) using pickle. 
+    [INFO] 12:44: The ExperimentManager was saved in : 'rlberry_data/temp/manager_data/my-agent_2024-10-16_12-44-49_e91ad515/manager_obj.pickle' 
+
+     Saved manager at rlberry_data/temp/manager_data/my-agent_2024-10-16_12-44-49_e91ad515/manager_obj.pickle.
+
+    [INFO] 12:44: Running ExperimentManager fit() for my-agent with n_fit = 2 and max_workers = None. 
+    checkpoint at rlberry_data/temp/manager_data/my-agent_2024-10-16_12-44-49_e91ad515/output_0/checkpoint.pickle (timestep = 500)
+    checkpoint at rlberry_data/temp/manager_data/my-agent_2024-10-16_12-44-49_e91ad515/output_1/checkpoint.pickle (timestep = 500)
+    checkpoint at rlberry_data/temp/manager_data/my-agent_2024-10-16_12-44-49_e91ad515/output_1/checkpoint.pickle (timestep = 1000)
+    checkpoint at rlberry_data/temp/manager_data/my-agent_2024-10-16_12-44-49_e91ad515/output_0/checkpoint.pickle (timestep = 1000)
+    [INFO] 12:44: ... trained! 
+    [INFO] 12:44: Saved ExperimentManager(my-agent) using pickle. 
+    [INFO] 12:44: The ExperimentManager was saved in : 'rlberry_data/temp/manager_data/my-agent_2024-10-16_12-44-49_e91ad515/manager_obj.pickle' 
+
+     Loading manager from rlberry_data/temp/manager_data/my-agent_2024-10-16_12-44-49_e91ad515/manager_obj.pickle.
+
+    [INFO] 12:44: Running ExperimentManager fit() for my-agent with n_fit = 2 and max_workers = None. 
  
-     --> MyAgent loaded from checkpoint: rlberry_data/temp/manager_data/my-agent_2024-10-16_10-02-00_9aa9c456/output_1/checkpoint.pickle 
+     --> MyAgent loaded from checkpoint: rlberry_data/temp/manager_data/my-agent_2024-10-16_12-44-49_e91ad515/output_0/checkpoint.pickle 
 
  
-     --> MyAgent loaded from checkpoint: rlberry_data/temp/manager_data/my-agent_2024-10-16_10-02-00_9aa9c456/output_0/checkpoint.pickle 
+     --> MyAgent loaded from checkpoint: rlberry_data/temp/manager_data/my-agent_2024-10-16_12-44-49_e91ad515/output_1/checkpoint.pickle 
 
-    checkpoint at rlberry_data/temp/manager_data/my-agent_2024-10-16_10-02-00_9aa9c456/output_1/checkpoint.pickle (timestep = 1500)
-    checkpoint at rlberry_data/temp/manager_data/my-agent_2024-10-16_10-02-00_9aa9c456/output_0/checkpoint.pickle (timestep = 1500)
-    [INFO] 10:02: ... trained! 
-    [INFO] 10:02: Saved ExperimentManager(my-agent) using pickle. 
-    [INFO] 10:02: The ExperimentManager was saved in : 'rlberry_data/temp/manager_data/my-agent_2024-10-16_10-02-00_9aa9c456/manager_obj.pickle' 
+    checkpoint at rlberry_data/temp/manager_data/my-agent_2024-10-16_12-44-49_e91ad515/output_0/checkpoint.pickle (timestep = 1500)
+    checkpoint at rlberry_data/temp/manager_data/my-agent_2024-10-16_12-44-49_e91ad515/output_1/checkpoint.pickle (timestep = 1500)
+    [INFO] 12:44: ... trained! 
+    [INFO] 12:44: Saved ExperimentManager(my-agent) using pickle. 
+    [INFO] 12:44: The ExperimentManager was saved in : 'rlberry_data/temp/manager_data/my-agent_2024-10-16_12-44-49_e91ad515/manager_obj.pickle' 
 
 
 
@@ -166,7 +166,7 @@ your agents, and how to restore from a previous checkpoint.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** (0 minutes 1.589 seconds)
+   **Total running time of the script:** (0 minutes 1.500 seconds)
 
 
 .. _sphx_glr_download_auto_examples_plot_checkpointing.py:
diff --git a/preview_pr/_sources/auto_examples/plot_kernels.rst.txt b/preview_pr/_sources/auto_examples/plot_kernels.rst.txt
index 8e77c2657..4dd4f1b46 100644
--- a/preview_pr/_sources/auto_examples/plot_kernels.rst.txt
+++ b/preview_pr/_sources/auto_examples/plot_kernels.rst.txt
@@ -69,7 +69,7 @@ This script requires matplotlib
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** (0 minutes 10.659 seconds)
+   **Total running time of the script:** (0 minutes 10.663 seconds)
 
 
 .. _sphx_glr_download_auto_examples_plot_kernels.py:
diff --git a/preview_pr/_sources/auto_examples/plot_smooth.rst.txt b/preview_pr/_sources/auto_examples/plot_smooth.rst.txt
index 208b9d252..dcc73df53 100644
--- a/preview_pr/_sources/auto_examples/plot_smooth.rst.txt
+++ b/preview_pr/_sources/auto_examples/plot_smooth.rst.txt
@@ -194,7 +194,7 @@ This script shows how to define a bandit environment and an UCB Index-based algo
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** (0 minutes 22.894 seconds)
+   **Total running time of the script:** (0 minutes 22.489 seconds)
 
 
 .. _sphx_glr_download_auto_examples_plot_smooth.py:
diff --git a/preview_pr/_sources/auto_examples/plot_writer_wrapper.rst.txt b/preview_pr/_sources/auto_examples/plot_writer_wrapper.rst.txt
index c115792ad..cc82abbab 100644
--- a/preview_pr/_sources/auto_examples/plot_writer_wrapper.rst.txt
+++ b/preview_pr/_sources/auto_examples/plot_writer_wrapper.rst.txt
@@ -63,9 +63,9 @@ during the fit of the agent and then use the plot utils.
 
  .. code-block:: none
 
-    [INFO] 10:01: ... trained! 
-    [INFO] 10:01: Saved ExperimentManager(UCBVIAgent) using pickle. 
-    [INFO] 10:01: The ExperimentManager was saved in : 'rlberry_data/temp/manager_data/UCBVIAgent_2024-10-16_10-01-31_7758b26c/manager_obj.pickle' 
+    [INFO] 12:44: ... trained! 
+    [INFO] 12:44: Saved ExperimentManager(UCBVIAgent) using pickle. 
+    [INFO] 12:44: The ExperimentManager was saved in : 'rlberry_data/temp/manager_data/UCBVIAgent_2024-10-16_12-44-21_f34c3498/manager_obj.pickle' 
 
 
 
@@ -141,7 +141,7 @@ during the fit of the agent and then use the plot utils.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** (0 minutes 3.353 seconds)
+   **Total running time of the script:** (0 minutes 3.395 seconds)
 
 
 .. _sphx_glr_download_auto_examples_plot_writer_wrapper.py:
diff --git a/preview_pr/_sources/auto_examples/sg_execution_times.rst.txt b/preview_pr/_sources/auto_examples/sg_execution_times.rst.txt
index f5f30e6a9..e929db9b2 100644
--- a/preview_pr/_sources/auto_examples/sg_execution_times.rst.txt
+++ b/preview_pr/_sources/auto_examples/sg_execution_times.rst.txt
@@ -6,18 +6,18 @@
 
 Computation times
 =================
-**00:39.728** total execution time for **auto_examples** files:
+**00:39.200** total execution time for **auto_examples** files:
 
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_auto_examples_plot_smooth.py` (``plot_smooth.py``)                 | 00:22.894 | 0.0 MB |
+| :ref:`sphx_glr_auto_examples_plot_smooth.py` (``plot_smooth.py``)                 | 00:22.489 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_auto_examples_plot_kernels.py` (``plot_kernels.py``)               | 00:10.659 | 0.0 MB |
+| :ref:`sphx_glr_auto_examples_plot_kernels.py` (``plot_kernels.py``)               | 00:10.663 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_auto_examples_plot_writer_wrapper.py` (``plot_writer_wrapper.py``) | 00:03.353 | 0.0 MB |
+| :ref:`sphx_glr_auto_examples_plot_writer_wrapper.py` (``plot_writer_wrapper.py``) | 00:03.395 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_auto_examples_plot_checkpointing.py` (``plot_checkpointing.py``)   | 00:01.589 | 0.0 MB |
+| :ref:`sphx_glr_auto_examples_plot_checkpointing.py` (``plot_checkpointing.py``)   | 00:01.500 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_auto_examples_plot_agent_manager.py` (``plot_agent_manager.py``)   | 00:01.233 | 0.0 MB |
+| :ref:`sphx_glr_auto_examples_plot_agent_manager.py` (``plot_agent_manager.py``)   | 00:01.153 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
 | :ref:`sphx_glr_auto_examples_adastop_example.py` (``adastop_example.py``)         | 00:00.000 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
diff --git a/preview_pr/_sources/basics/userguide/export_training_data.md.txt b/preview_pr/_sources/basics/userguide/export_training_data.md.txt
index cfe54c358..25c7bf1e0 100644
--- a/preview_pr/_sources/basics/userguide/export_training_data.md.txt
+++ b/preview_pr/_sources/basics/userguide/export_training_data.md.txt
@@ -195,7 +195,7 @@ model2_seed2.learn(total_timesteps=5_000, tb_log_name="A2C")
 ```
 
 
-Then, if you need to these logs in a pandas dataframe, you can use the tool `tensorboard_folder_to_dataframe`.
+Then, if you need to these logs in a pandas dataframe, you can use the tool [tensorboard_folder_to_dataframe](rlberry.manager.tensorboard_folder_to_dataframe).
 It will give you a `Dict` with all the scalar data from the tensorboad folder.
 
     - The `keys` will be the "tag" (the name of the measure)
diff --git a/preview_pr/api.html b/preview_pr/api.html
index 12b3a7dec..c85f8ad15 100644
--- a/preview_pr/api.html
+++ b/preview_pr/api.html
@@ -197,7 +197,7 @@ 
Evaluation and plot
manager.tensorboard_folder_to_dataframe(...)

-
path_to_tensorboard_data : path to the tensorboard data.

+
Function to convert 'tensorboard log' to 'Panda DataFrames'

 
 
 
diff --git a/preview_pr/auto_examples/demo_bandits/plot_TS_bandit.html b/preview_pr/auto_examples/demo_bandits/plot_TS_bandit.html
index 8de7bdb00..812129217 100644
--- a/preview_pr/auto_examples/demo_bandits/plot_TS_bandit.html
+++ b/preview_pr/auto_examples/demo_bandits/plot_TS_bandit.html
@@ -133,21 +133,21 @@
 
  • Cumulative Pseudo-Regret
    • 
 
  • Cumulative Pseudo-Regret
    • 
 
-
    [INFO] 10:02: ... trained!
-[INFO] 10:02: Saved ExperimentManager(Bounded UCB Agent) using pickle.
-[INFO] 10:02: The ExperimentManager was saved in : 'rlberry_data/temp/manager_data/Bounded UCB Agent_2024-10-16_10-02-15_bd80da50/manager_obj.pickle'
-[INFO] 10:02: Running ExperimentManager fit() for Bernoulli TS Agent with n_fit = 10 and max_workers = None.
-[INFO] 10:02: ... trained!
-[INFO] 10:02: Saved ExperimentManager(Bernoulli TS Agent) using pickle.
-[INFO] 10:02: The ExperimentManager was saved in : 'rlberry_data/temp/manager_data/Bernoulli TS Agent_2024-10-16_10-02-15_847030c5/manager_obj.pickle'
-[INFO] 10:02: Running ExperimentManager fit() for Gaussian UCB Agent with n_fit = 10 and max_workers = None.
-[INFO] 10:02: ... trained!
-[INFO] 10:02: Saved ExperimentManager(Gaussian UCB Agent) using pickle.
-[INFO] 10:02: The ExperimentManager was saved in : 'rlberry_data/temp/manager_data/Gaussian UCB Agent_2024-10-16_10-02-26_8ade7c97/manager_obj.pickle'
-[INFO] 10:02: Running ExperimentManager fit() for Gaussian TS Agent with n_fit = 10 and max_workers = None.
-[INFO] 10:02: ... trained!
-[INFO] 10:02: Saved ExperimentManager(Gaussian TS Agent) using pickle.
-[INFO] 10:02: The ExperimentManager was saved in : 'rlberry_data/temp/manager_data/Gaussian TS Agent_2024-10-16_10-02-26_ae7d0a45/manager_obj.pickle'
+
    [INFO] 12:45: ... trained!
+[INFO] 12:45: Saved ExperimentManager(Bounded UCB Agent) using pickle.
+[INFO] 12:45: The ExperimentManager was saved in : 'rlberry_data/temp/manager_data/Bounded UCB Agent_2024-10-16_12-45-04_319cb7e6/manager_obj.pickle'
+[INFO] 12:45: Running ExperimentManager fit() for Bernoulli TS Agent with n_fit = 10 and max_workers = None.
+[INFO] 12:45: ... trained!
+[INFO] 12:45: Saved ExperimentManager(Bernoulli TS Agent) using pickle.
+[INFO] 12:45: The ExperimentManager was saved in : 'rlberry_data/temp/manager_data/Bernoulli TS Agent_2024-10-16_12-45-04_279a356f/manager_obj.pickle'
+[INFO] 12:45: Running ExperimentManager fit() for Gaussian UCB Agent with n_fit = 10 and max_workers = None.
+[INFO] 12:45: ... trained!
+[INFO] 12:45: Saved ExperimentManager(Gaussian UCB Agent) using pickle.
+[INFO] 12:45: The ExperimentManager was saved in : 'rlberry_data/temp/manager_data/Gaussian UCB Agent_2024-10-16_12-45-15_f7d90f4f/manager_obj.pickle'
+[INFO] 12:45: Running ExperimentManager fit() for Gaussian TS Agent with n_fit = 10 and max_workers = None.
+[INFO] 12:45: ... trained!
+[INFO] 12:45: Saved ExperimentManager(Gaussian TS Agent) using pickle.
+[INFO] 12:45: The ExperimentManager was saved in : 'rlberry_data/temp/manager_data/Gaussian TS Agent_2024-10-16_12-45-15_9c0d4b7a/manager_obj.pickle'
 
    
 
    
 
    
@@ -295,7 +295,7 @@
 )
 
    
 
    
-
    Total running time of the script: (0 minutes 20.469 seconds)
    
+
    Total running time of the script: (0 minutes 20.350 seconds)
    -

    Total running time of the script: (0 minutes 41.147 seconds)

    +

    Total running time of the script: (0 minutes 41.099 seconds)

    -

    Total running time of the script: (0 minutes 7.648 seconds)

    +

    Total running time of the script: (0 minutes 7.662 seconds)