Merge pull request #18 from BDonnot/gym_integration

fix bugs introduced in 0.4.2 previously
rte-france · Jan 9, 2020 · a2c19cc · a2c19cc
2 parents f7996ee + b1117e1
commit a2c19cc
Show file tree

Hide file tree

Showing 10 changed files with 125 additions and 2,632 deletions.
diff --git a/.gitignore b/.gitignore
@@ -157,3 +157,4 @@ grid2op/data/*/*.csv
 **token.json
 getting_started/study_agent_getting_started/
 **Untitled.ipynb
+grid2op/tests/test_agent/
diff --git a/CHANGELOG.rst b/CHANGELOG.rst
@@ -3,6 +3,10 @@ Change Log
 
 [0.4.2] - 2020-01-08
 --------------------
+- [FIXED] Runner cannot save properly action and observation (sizes are not computed properly)
+  **now fixed and unit test added**
+- [FIXED] Plot utility has a bug in extracting grid information.
+  **now fixed**
 - [FIXED] gym compatibility issue for environment
 - [FIXED] checking key-word arguments in "make" function: if an invalid argument is provided,
   it now raises an error.
@@ -38,6 +42,8 @@ Action / Observation  _lines_ex_pos_topo_vect  lines_ex_pos_topo_vect
 GridValue             n_lines                  n_line
 ====================  =======================  =======================
 
+- [UPDATE] the getting_started notebooks to reflect these changes
+
 [0.4.1] - 2019-12-17
 --------------------
 - [FIXED] Bug#14 : Nan in the observation space after switching one powerline [PandaPowerBackend]

diff --git a/getting_started/1_Observation_Agents.ipynb b/getting_started/1_Observation_Agents.ipynb
@@ -338,8 +338,8 @@
    "source": [
     "print(\"Number of generators of the powergrid: {}\".format(obs.n_gen))\n",
     "print(\"Number of loads of the powergrid: {}\".format(obs.n_load))\n",
-    "print(\"Number of powerline of the powergrid: {}\".format(obs.n_lines))\n",
-    "print(\"Number of elements connected to each substations in the powergrid: {}\".format(obs.subs_info))\n",
+    "print(\"Number of powerline of the powergrid: {}\".format(obs.n_line))\n",
+    "print(\"Number of elements connected to each substations in the powergrid: {}\".format(obs.sub_info))\n",
     "print(\"Total number of elements: {}\".format(obs.dim_topo))"
    ]
   },
@@ -566,8 +566,8 @@
     {
      "data": {
       "text/plain": [
-       "array([ 2.01900000e+03,  1.00000000e+00,  1.00000000e+00,  1.00000000e+00,\n",
-       "        0.00000000e+00,  1.00000000e+01,  0.00000000e+00,  2.33000000e+01,\n",
+       "array([ 2.01900000e+03,  1.00000000e+00,  1.00000000e+00,  0.00000000e+00,\n",
+       "        1.00000000e+01,  1.00000000e+00,  0.00000000e+00,  2.33000000e+01,\n",
        "        6.91000000e+01,  0.00000000e+00,  1.49484728e+02,  4.71334079e+01,\n",
        "        5.86085309e+01,  1.46841707e+01,  2.35979474e+01, -4.99757939e-01,\n",
        "        1.41075000e+02,  1.36350000e+02,  2.22560000e-01,  1.30800000e+01,\n",
@@ -648,7 +648,35 @@
        "        1.00000000e+00,  1.00000000e+00,  1.00000000e+00,  1.00000000e+00,\n",
        "        1.00000000e+00,  1.00000000e+00,  1.00000000e+00,  1.00000000e+00,\n",
        "        1.00000000e+00,  1.00000000e+00,  1.00000000e+00,  1.00000000e+00,\n",
-       "        1.00000000e+00,  1.00000000e+00])"
+       "        1.00000000e+00,  1.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
+       "        0.00000000e+00,  0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
+       "        0.00000000e+00,  0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
+       "        0.00000000e+00,  0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
+       "        0.00000000e+00,  0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
+       "        0.00000000e+00,  0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
+       "        0.00000000e+00,  0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
+       "        0.00000000e+00,  0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
+       "        0.00000000e+00,  0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
+       "        0.00000000e+00,  0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
+       "        0.00000000e+00,  0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
+       "        0.00000000e+00,  0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
+       "        0.00000000e+00,  0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
+       "        0.00000000e+00,  0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
+       "        0.00000000e+00,  0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
+       "        0.00000000e+00,  0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
+       "        0.00000000e+00,  0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
+       "        0.00000000e+00,  0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
+       "        0.00000000e+00,  0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
+       "       -1.00000000e+00, -1.00000000e+00, -1.00000000e+00, -1.00000000e+00,\n",
+       "       -1.00000000e+00, -1.00000000e+00, -1.00000000e+00, -1.00000000e+00,\n",
+       "       -1.00000000e+00, -1.00000000e+00, -1.00000000e+00, -1.00000000e+00,\n",
+       "       -1.00000000e+00, -1.00000000e+00, -1.00000000e+00, -1.00000000e+00,\n",
+       "       -1.00000000e+00, -1.00000000e+00, -1.00000000e+00, -1.00000000e+00,\n",
+       "        0.00000000e+00,  0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
+       "        0.00000000e+00,  0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
+       "        0.00000000e+00,  0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
+       "        0.00000000e+00,  0.00000000e+00,  0.00000000e+00,  0.00000000e+00,\n",
+       "        0.00000000e+00,  0.00000000e+00,  0.00000000e+00,  0.00000000e+00])"
       ]
      },
      "execution_count": 12,
@@ -693,7 +721,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "None\n"
+      "[nan nan nan nan nan]\n"
      ]
     }
    ],
@@ -1049,8 +1077,8 @@
      "text": [
       "The results for DoNothing agent are:\n",
       "\tFor chronics located at /home/donnotben/.local/lib/python3.6/site-packages/grid2op/data/test_multi_chronics/1\n",
-      "\t\t - cumulative reward: 999.991933\n",
-      "\t\t - number of time steps completed: 50 / 50\n"
+      "\t\t - cumulative reward: 199.998003\n",
+      "\t\t - number of time steps completed: 10 / 10\n"
      ]
     }
    ],
@@ -1089,8 +1117,8 @@
      "text": [
       "The results for the custom agent are:\n",
       "\tFor chronics located at /home/donnotben/.local/lib/python3.6/site-packages/grid2op/data/test_multi_chronics/1\n",
-      "\t\t - cumulative reward: 999.991933\n",
-      "\t\t - number of time steps completed: 50 / 50\n"
+      "\t\t - cumulative reward: 199.998003\n",
+      "\t\t - number of time steps completed: 10 / 10\n"
      ]
     }
    ],
@@ -1115,7 +1143,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "As we can see, there is a slight degradation of the performance (cumulative reward) for the second (custom) Agent. The cumulative reward in this case is 5739.95***1012*** the baseline (do nothing) was 5739.95***1023***.\n",
+    "As we can see, there is no change in the performance for both agent (there would be if we didn't limit the episode length to 10 time steps)\n",
     "\n",
     "This agent is NOT recommended.\n",
     "\n",
@@ -1133,8 +1161,8 @@
      "text": [
       "The results for the PowerLineSwitch agent are:\n",
       "\tFor chronics located at /home/donnotben/.local/lib/python3.6/site-packages/grid2op/data/test_multi_chronics/1\n",
-      "\t\t - cumulative reward: 999.992511\n",
-      "\t\t - number of time steps completed: 50 / 50\n"
+      "\t\t - cumulative reward: 199.998134\n",
+      "\t\t - number of time steps completed: 10 / 10\n"
      ]
     }
    ],
@@ -1160,7 +1188,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "We want however to emphasize that do nothing is NOT the best solution, even in this simple case. For example, an agent choosing at each time step to disconnect / reconnect as to greedily maximize the anticipated reward will have a cumulative reward of 5739.95***2376*** in this situation.\n",
+    "We want however to emphasize that do nothing is NOT the best solution, even in this simple case. For example, an agent choosing at each time step to disconnect / reconnect as to greedily maximize the anticipated reward will have a cumulative reward of 199.998***134*** in this situation.\n",
     "\n",
     "**NB** For these simulations, the score is completely irrealistic. Indeed, no special care has been taken to set the thermal limits to plausible values. This explain the very little different observed between the three agents above.\n",
     "\n",