auto-generating sphinx docs

main/_modules/torchrl/objectives/utils.html

@@ -810,11 +810,18 @@ <h1>Source code for torchrl.objectives.utils</h1><div class="highlight"><pre>
 
     <span class="k">def</span> <span class="fm">__enter__</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
         <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">mode</span><span class="p">:</span>
-            <span class="bp">self</span><span class="o">.</span><span class="n">network</span><span class="o">.</span><span class="n">requires_grad_</span><span class="p">(</span><span class="kc">False</span><span class="p">)</span>
+            <span class="k">if</span> <span class="n">is_dynamo_compiling</span><span class="p">():</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">_params</span> <span class="o">=</span> <span class="n">TensorDict</span><span class="o">.</span><span class="n">from_module</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">network</span><span class="p">)</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">_params</span><span class="o">.</span><span class="n">data</span><span class="o">.</span><span class="n">to_module</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">network</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">network</span><span class="o">.</span><span class="n">requires_grad_</span><span class="p">(</span><span class="kc">False</span><span class="p">)</span>
 
     <span class="k">def</span> <span class="fm">__exit__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">exc_type</span><span class="p">,</span> <span class="n">exc_val</span><span class="p">,</span> <span class="n">exc_tb</span><span class="p">)</span> <span class="o">-&gt;</span> <span class="kc">None</span><span class="p">:</span>
         <span class="k">if</span> <span class="bp">self</span><span class="o">.</span><span class="n">mode</span><span class="p">:</span>
-            <span class="bp">self</span><span class="o">.</span><span class="n">network</span><span class="o">.</span><span class="n">requires_grad_</span><span class="p">()</span></div>
+            <span class="k">if</span> <span class="n">is_dynamo_compiling</span><span class="p">():</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">_params</span><span class="o">.</span><span class="n">to_module</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">network</span><span class="p">)</span>
+            <span class="k">else</span><span class="p">:</span>
+                <span class="bp">self</span><span class="o">.</span><span class="n">network</span><span class="o">.</span><span class="n">requires_grad_</span><span class="p">()</span></div>
 
 
 <div class="viewcode-block" id="hold_out_params"><a class="viewcode-back" href="../../../reference/generated/torchrl.objectives.hold_out_params.html#torchrl.objectives.hold_out_params">[docs]</a><span class="k">class</span> <span class="nc">hold_out_params</span><span class="p">(</span><span class="n">_context_manager</span><span class="p">):</span>

main/_sources/sg_execution_times.rst.txt

@@ -6,7 +6,7 @@
 
 Computation times
 =================
-**27:50.593** total execution time for 18 files **from all galleries**:
+**27:35.002** total execution time for 18 files **from all galleries**:
 
 .. container::
 
@@ -33,56 +33,56 @@ Computation times
      - Time
      - Mem (MB)
    * - :ref:`sphx_glr_tutorials_torchrl_demo.py` (``reference/generated/tutorials/torchrl_demo.py``)
-     - 03:44.258
-     - 350.1
+     - 03:43.142
+     - 328.6
    * - :ref:`sphx_glr_tutorials_torchrl_envs.py` (``reference/generated/tutorials/torchrl_envs.py``)
-     - 03:28.778
-     - 4750.9
+     - 03:27.697
+     - 3191.4
    * - :ref:`sphx_glr_tutorials_dqn_with_rnn.py` (``reference/generated/tutorials/dqn_with_rnn.py``)
-     - 03:09.560
-     - 2201.1
+     - 03:06.208
+     - 2219.5
    * - :ref:`sphx_glr_tutorials_rb_tutorial.py` (``reference/generated/tutorials/rb_tutorial.py``)
-     - 02:56.498
-     - 539.2
+     - 02:56.053
+     - 528.8
    * - :ref:`sphx_glr_tutorials_coding_dqn.py` (``reference/generated/tutorials/coding_dqn.py``)
-     - 02:40.706
-     - 1101.4
+     - 02:38.515
+     - 1361.2
    * - :ref:`sphx_glr_tutorials_coding_ddpg.py` (``reference/generated/tutorials/coding_ddpg.py``)
-     - 01:53.236
-     - 331.2
+     - 01:52.093
+     - 332.4
    * - :ref:`sphx_glr_tutorials_multiagent_ppo.py` (``reference/generated/tutorials/multiagent_ppo.py``)
-     - 01:52.901
-     - 320.9
+     - 01:50.887
+     - 322.9
    * - :ref:`sphx_glr_tutorials_multiagent_competitive_ddpg.py` (``reference/generated/tutorials/multiagent_competitive_ddpg.py``)
-     - 01:31.871
-     - 322.4
+     - 01:30.707
+     - 323.6
    * - :ref:`sphx_glr_tutorials_coding_ppo.py` (``reference/generated/tutorials/coding_ppo.py``)
-     - 01:26.437
-     - 319.9
+     - 01:25.860
+     - 321.0
    * - :ref:`sphx_glr_tutorials_pretrained_models.py` (``reference/generated/tutorials/pretrained_models.py``)
-     - 00:55.856
-     - 4235.3
+     - 00:54.184
+     - 2642.4
    * - :ref:`sphx_glr_tutorials_multi_task.py` (``reference/generated/tutorials/multi_task.py``)
-     - 00:51.256
-     - 320.6
+     - 00:51.038
+     - 322.4
    * - :ref:`sphx_glr_tutorials_getting-started-1.py` (``reference/generated/tutorials/getting-started-1.py``)
-     - 00:46.472
-     - 319.1
+     - 00:46.351
+     - 320.2
    * - :ref:`sphx_glr_tutorials_getting-started-0.py` (``reference/generated/tutorials/getting-started-0.py``)
-     - 00:46.344
-     - 319.1
+     - 00:46.183
+     - 320.2
    * - :ref:`sphx_glr_tutorials_getting-started-2.py` (``reference/generated/tutorials/getting-started-2.py``)
-     - 00:26.881
-     - 322.3
+     - 00:26.768
+     - 320.2
    * - :ref:`sphx_glr_tutorials_getting-started-5.py` (``reference/generated/tutorials/getting-started-5.py``)
-     - 00:22.901
-     - 323.8
+     - 00:22.841
+     - 324.0
    * - :ref:`sphx_glr_tutorials_getting-started-3.py` (``reference/generated/tutorials/getting-started-3.py``)
-     - 00:22.019
-     - 320.7
+     - 00:21.930
+     - 322.4
    * - :ref:`sphx_glr_tutorials_pendulum.py` (``reference/generated/tutorials/pendulum.py``)
-     - 00:19.899
-     - 5083.2
+     - 00:19.866
+     - 3635.4
    * - :ref:`sphx_glr_tutorials_getting-started-4.py` (``reference/generated/tutorials/getting-started-4.py``)
-     - 00:14.722
-     - 350.7
+     - 00:14.680
+     - 351.6

main/_sources/tutorials/coding_ddpg.rst.txt

@@ -1634,7 +1634,7 @@ modules we need.
 
  .. code-block:: none
 
-      0%|          | 0/10000 [00:00<?, ?it/s]      8%|▊         | 800/10000 [00:00<00:03, 2533.63it/s]     16%|█▌        | 1600/10000 [00:01<00:10, 827.75it/s]     24%|██▍       | 2400/10000 [00:02<00:06, 1184.15it/s]     32%|███▏      | 3200/10000 [00:02<00:04, 1482.38it/s]     40%|████      | 4000/10000 [00:02<00:03, 1727.05it/s]     48%|████▊     | 4800/10000 [00:03<00:02, 1932.28it/s]     56%|█████▌    | 5600/10000 [00:03<00:02, 2091.49it/s]    reward: -2.23 (r0 = -1.88), reward eval: reward:  0.00, reward normalized=-1.85/5.95, grad norm= 115.03, loss_value= 223.69, loss_actor= 14.65, target value: -12.77:  56%|█████▌    | 5600/10000 [00:05<00:02, 2091.49it/s]    reward: -2.23 (r0 = -1.88), reward eval: reward:  0.00, reward normalized=-1.85/5.95, grad norm= 115.03, loss_value= 223.69, loss_actor= 14.65, target value: -12.77:  64%|██████▍   | 6400/10000 [00:05<00:04, 837.12it/s]     reward: -2.76 (r0 = -1.88), reward eval: reward:  0.00, reward normalized=-2.45/5.53, grad norm= 49.43, loss_value= 213.72, loss_actor= 15.69, target value: -16.02:  64%|██████▍   | 6400/10000 [00:07<00:04, 837.12it/s]     reward: -2.76 (r0 = -1.88), reward eval: reward:  0.00, reward normalized=-2.45/5.53, grad norm= 49.43, loss_value= 213.72, loss_actor= 15.69, target value: -16.02:  72%|███████▏  | 7200/10000 [00:07<00:04, 595.56it/s]    reward: -4.17 (r0 = -1.88), reward eval: reward:  0.00, reward normalized=-2.85/5.99, grad norm= 60.16, loss_value= 295.90, loss_actor= 18.57, target value: -19.15:  72%|███████▏  | 7200/10000 [00:09<00:04, 595.56it/s]    reward: -4.17 (r0 = -1.88), reward eval: reward:  0.00, reward normalized=-2.85/5.99, grad norm= 60.16, loss_value= 295.90, loss_actor= 18.57, target value: -19.15:  80%|████████  | 8000/10000 [00:09<00:04, 495.01it/s]    reward: -3.54 (r0 = -1.88), reward eval: reward:  0.00, reward normalized=-2.78/5.50, grad norm= 93.13, loss_value= 222.20, loss_actor= 17.51, target value: -19.17:  80%|████████  | 8000/10000 [00:11<00:04, 495.01it/s]    reward: -3.54 (r0 = -1.88), reward eval: reward:  0.00, reward normalized=-2.78/5.50, grad norm= 93.13, loss_value= 222.20, loss_actor= 17.51, target value: -19.17:  88%|████████▊ | 8800/10000 [00:12<00:02, 447.78it/s]    reward: -3.94 (r0 = -1.88), reward eval: reward: -5.05, reward normalized=-3.18/5.21, grad norm= 73.80, loss_value= 216.62, loss_actor= 21.77, target value: -23.01:  88%|████████▊ | 8800/10000 [00:14<00:02, 447.78it/s]    reward: -3.94 (r0 = -1.88), reward eval: reward: -5.05, reward normalized=-3.18/5.21, grad norm= 73.80, loss_value= 216.62, loss_actor= 21.77, target value: -23.01:  96%|█████████▌| 9600/10000 [00:15<00:01, 356.23it/s]    reward: -1.04 (r0 = -1.88), reward eval: reward: -5.05, reward normalized=-3.08/5.53, grad norm= 191.09, loss_value= 228.22, loss_actor= 18.82, target value: -21.64:  96%|█████████▌| 9600/10000 [00:17<00:01, 356.23it/s]    reward: -1.04 (r0 = -1.88), reward eval: reward: -5.05, reward normalized=-3.08/5.53, grad norm= 191.09, loss_value= 228.22, loss_actor= 18.82, target value: -21.64: : 10400it [00:18, 327.83it/s]                            reward: -4.56 (r0 = -1.88), reward eval: reward: -5.05, reward normalized=-3.14/5.20, grad norm= 203.16, loss_value= 257.88, loss_actor= 20.11, target value: -22.75: : 10400it [00:20, 327.83it/s]
+      0%|          | 0/10000 [00:00<?, ?it/s]      8%|▊         | 800/10000 [00:00<00:03, 2826.97it/s]     16%|█▌        | 1600/10000 [00:01<00:09, 876.90it/s]     24%|██▍       | 2400/10000 [00:01<00:06, 1263.22it/s]     32%|███▏      | 3200/10000 [00:02<00:04, 1597.19it/s]     40%|████      | 4000/10000 [00:02<00:03, 1865.86it/s]     48%|████▊     | 4800/10000 [00:02<00:02, 2075.87it/s]     56%|█████▌    | 5600/10000 [00:03<00:01, 2232.59it/s]    reward: -2.07 (r0 = -2.11), reward eval: reward: -0.01, reward normalized=-2.73/6.82, grad norm= 357.93, loss_value= 599.35, loss_actor= 13.16, target value: -17.71:  56%|█████▌    | 5600/10000 [00:04<00:01, 2232.59it/s]    reward: -2.07 (r0 = -2.11), reward eval: reward: -0.01, reward normalized=-2.73/6.82, grad norm= 357.93, loss_value= 599.35, loss_actor= 13.16, target value: -17.71:  64%|██████▍   | 6400/10000 [00:05<00:04, 873.31it/s]     reward: -1.95 (r0 = -2.11), reward eval: reward: -0.01, reward normalized=-2.58/5.85, grad norm= 157.59, loss_value= 272.08, loss_actor= 12.85, target value: -16.15:  64%|██████▍   | 6400/10000 [00:06<00:04, 873.31it/s]    reward: -1.95 (r0 = -2.11), reward eval: reward: -0.01, reward normalized=-2.58/5.85, grad norm= 157.59, loss_value= 272.08, loss_actor= 12.85, target value: -16.15:  72%|███████▏  | 7200/10000 [00:07<00:04, 612.34it/s]    reward: -4.90 (r0 = -2.11), reward eval: reward: -0.01, reward normalized=-3.03/6.39, grad norm= 133.85, loss_value= 380.05, loss_actor= 18.30, target value: -20.30:  72%|███████▏  | 7200/10000 [00:09<00:04, 612.34it/s]    reward: -4.90 (r0 = -2.11), reward eval: reward: -0.01, reward normalized=-3.03/6.39, grad norm= 133.85, loss_value= 380.05, loss_actor= 18.30, target value: -20.30:  80%|████████  | 8000/10000 [00:09<00:03, 513.24it/s]    reward: -4.33 (r0 = -2.11), reward eval: reward: -0.01, reward normalized=-3.11/5.59, grad norm= 95.87, loss_value= 247.86, loss_actor= 19.65, target value: -20.67:  80%|████████  | 8000/10000 [00:11<00:03, 513.24it/s]     reward: -4.33 (r0 = -2.11), reward eval: reward: -0.01, reward normalized=-3.11/5.59, grad norm= 95.87, loss_value= 247.86, loss_actor= 19.65, target value: -20.67:  88%|████████▊ | 8800/10000 [00:11<00:02, 464.24it/s]    reward: -5.32 (r0 = -2.11), reward eval: reward: -5.97, reward normalized=-2.78/4.76, grad norm= 127.61, loss_value= 137.52, loss_actor= 22.26, target value: -20.93:  88%|████████▊ | 8800/10000 [00:14<00:02, 464.24it/s]    reward: -5.32 (r0 = -2.11), reward eval: reward: -5.97, reward normalized=-2.78/4.76, grad norm= 127.61, loss_value= 137.52, loss_actor= 22.26, target value: -20.93:  96%|█████████▌| 9600/10000 [00:14<00:01, 369.08it/s]    reward: -5.01 (r0 = -2.11), reward eval: reward: -5.97, reward normalized=-3.93/4.97, grad norm= 213.26, loss_value= 254.96, loss_actor= 24.40, target value: -26.34:  96%|█████████▌| 9600/10000 [00:16<00:01, 369.08it/s]    reward: -5.01 (r0 = -2.11), reward eval: reward: -5.97, reward normalized=-3.93/4.97, grad norm= 213.26, loss_value= 254.96, loss_actor= 24.40, target value: -26.34: : 10400it [00:17, 339.95it/s]                            reward: -4.92 (r0 = -2.11), reward eval: reward: -5.97, reward normalized=-3.35/4.84, grad norm= 116.39, loss_value= 227.09, loss_actor= 22.01, target value: -23.85: : 10400it [00:19, 339.95it/s]
 
 
 
@@ -1704,9 +1704,9 @@ To iterate further on this loss module we might consider:
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** (1 minutes 53.236 seconds)
+   **Total running time of the script:** (1 minutes 52.093 seconds)
 
-**Estimated memory usage:**  331 MB
+**Estimated memory usage:**  332 MB
 
 
 .. _sphx_glr_download_tutorials_coding_ddpg.py: