[Doc, Test] Add A2C script test and doc

.circleci/unittest/linux_examples/scripts/run_test.sh

@@ -38,6 +38,18 @@ coverage run examples/ddpg/ddpg.py \
   record_video=True \
   record_frames=4 \
   buffer_size=120
+coverage run examples/a2c/a2c.py \
+  total_frames=48 \
+  init_random_frames=10 \
+  batch_size=10 \
+  frames_per_batch=16 \
+  num_workers=2 \
+  env_per_collector=1 \
+  collector_devices=cuda:0 \
+  optim_steps_per_batch=1 \
+  record_video=True \
+  record_frames=4 \
+  buffer_size=120
 coverage run examples/dqn/dqn.py \
   total_frames=48 \
   init_random_frames=10 \

README.md

@@ -350,9 +350,11 @@ If you would like to contribute to new features, check our [call for contributio
 A series of [examples](examples/) are provided with an illustrative purpose:
 - [DQN (and add-ons up to Rainbow)](examples/dqn/dqn.py)
 - [DDPG](examples/ddpg/ddpg.py)
+- [A2C](examples/a2c/a2c.py)
 - [PPO](examples/ppo/ppo.py)
 - [SAC](examples/sac/sac.py)
 - [REDQ](examples/redq/redq.py)
+- [Dreamer](examples/dreamer/dreamer.py)
 
 and many more to come!
 

docs/source/reference/objectives.rst

@@ -51,6 +51,16 @@ PPO
     ClipPPOLoss
     KLPENPPOLoss
 
+A2C
+---
+
+.. autosummary::
+    :toctree: generated/
+    :template: rl_template_noinherit.rst
+
+    A2CLoss
+
+
 Returns
 -------
 .. currentmodule:: torchrl.objectives.value

docs/source/reference/trainers.rst

@@ -164,6 +164,7 @@ Builders
     transformed_env_constructor
     parallel_env_constructor
     make_sac_loss
+    make_a2c_loss
     make_dqn_loss
     make_ddpg_loss
     make_target_updater
@@ -172,6 +173,7 @@ Builders
     make_dqn_actor
     make_ddpg_actor
     make_ppo_model
+    make_a2c_model
     make_sac_model
     make_redq_model
     make_replay_buffer