mirror of
https://github.com/gryf/coach.git
synced 2025-12-17 19:20:19 +01:00
Add documentation on distributed Coach. (#158)
* Added documentation on distributed Coach.
This commit is contained in:
Balaji Subramaniam
committed by
Gal Novik
Gal Novik
parent
e3ecf445e2
commit
d06197f663
@@ -85,6 +85,7 @@
|
||||
<p class="caption"><span class="caption-text">Intro</span></p>
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../usage.html">Usage</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../dist_usage.html">Usage - Distributed Coach</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../features/index.html">Features</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../selecting_an_algorithm.html">Selecting an Algorithm</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../dashboard.html">Coach Dashboard</a></li>
|
||||
@@ -93,6 +94,7 @@
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../design/control_flow.html">Control Flow</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../design/network.html">Network Design</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../design/horizontal_scaling.html">Distributed Coach - Horizontal Scale-Out</a></li>
|
||||
</ul>
|
||||
<p class="caption"><span class="caption-text">Contributing</span></p>
|
||||
<ul>
|
||||
@@ -103,10 +105,13 @@
|
||||
<ul>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../components/agents/index.html">Agents</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../components/architectures/index.html">Architectures</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../components/data_stores/index.html">Data Stores</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../components/environments/index.html">Environments</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../components/exploration_policies/index.html">Exploration Policies</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../components/filters/index.html">Filters</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../components/memories/index.html">Memories</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../components/memory_backends/index.html">Memory Backends</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../components/orchestrators/index.html">Orchestrators</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../components/core_types.html">Core Types</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../components/spaces.html">Spaces</a></li>
|
||||
<li class="toctree-l1"><a class="reference internal" href="../../components/additional_parameters.html">Additional Parameters</a></li>
|
||||
@@ -227,6 +232,43 @@
|
||||
<span class="c1">#ConcatDepthWise = 2</span>
|
||||
<span class="c1">#Multiply = 3</span>
|
||||
|
||||
<span class="k">class</span> <span class="nc">RunType</span><span class="p">(</span><span class="n">Enum</span><span class="p">):</span>
|
||||
<span class="n">ORCHESTRATOR</span> <span class="o">=</span> <span class="s2">"orchestrator"</span>
|
||||
<span class="n">TRAINER</span> <span class="o">=</span> <span class="s2">"trainer"</span>
|
||||
<span class="n">ROLLOUT_WORKER</span> <span class="o">=</span> <span class="s2">"rollout-worker"</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">__str__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">value</span>
|
||||
|
||||
|
||||
<span class="k">class</span> <span class="nc">DeviceType</span><span class="p">(</span><span class="n">Enum</span><span class="p">):</span>
|
||||
<span class="n">CPU</span> <span class="o">=</span> <span class="s1">'cpu'</span>
|
||||
<span class="n">GPU</span> <span class="o">=</span> <span class="s1">'gpu'</span>
|
||||
|
||||
|
||||
<span class="k">class</span> <span class="nc">Device</span><span class="p">(</span><span class="nb">object</span><span class="p">):</span>
|
||||
<span class="k">def</span> <span class="nf">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">device_type</span><span class="p">:</span> <span class="n">DeviceType</span><span class="p">,</span> <span class="n">index</span><span class="p">:</span> <span class="nb">int</span><span class="o">=</span><span class="mi">0</span><span class="p">):</span>
|
||||
<span class="sd">"""</span>
|
||||
<span class="sd"> :param device_type: type of device (CPU/GPU)</span>
|
||||
<span class="sd"> :param index: index of device (only used if device type is GPU)</span>
|
||||
<span class="sd"> """</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">_device_type</span> <span class="o">=</span> <span class="n">device_type</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">_index</span> <span class="o">=</span> <span class="n">index</span>
|
||||
|
||||
<span class="nd">@property</span>
|
||||
<span class="k">def</span> <span class="nf">device_type</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_device_type</span>
|
||||
|
||||
<span class="nd">@property</span>
|
||||
<span class="k">def</span> <span class="nf">index</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="k">return</span> <span class="bp">self</span><span class="o">.</span><span class="n">_index</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">__str__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="k">return</span> <span class="s2">"</span><span class="si">{}{}</span><span class="s2">"</span><span class="o">.</span><span class="n">format</span><span class="p">(</span><span class="bp">self</span><span class="o">.</span><span class="n">_device_type</span><span class="p">,</span> <span class="bp">self</span><span class="o">.</span><span class="n">_index</span><span class="p">)</span>
|
||||
|
||||
<span class="k">def</span> <span class="nf">__repr__</span><span class="p">(</span><span class="bp">self</span><span class="p">):</span>
|
||||
<span class="k">return</span> <span class="nb">str</span><span class="p">(</span><span class="bp">self</span><span class="p">)</span>
|
||||
|
||||
|
||||
<span class="c1"># DistributedCoachSynchronizationType provides the synchronization type for distributed Coach.</span>
|
||||
<span class="c1"># The default value is None, which means the algorithm or preset cannot be used with distributed Coach.</span>
|
||||
@@ -346,6 +388,9 @@
|
||||
<span class="c1"># Distributed Coach params</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">distributed_coach_synchronization_type</span> <span class="o">=</span> <span class="kc">None</span>
|
||||
|
||||
<span class="c1"># Should the workers wait for full episode</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">act_for_full_episodes</span> <span class="o">=</span> <span class="kc">False</span>
|
||||
|
||||
|
||||
<div class="viewcode-block" id="PresetValidationParameters"><a class="viewcode-back" href="../../components/additional_parameters.html#rl_coach.base_parameters.PresetValidationParameters">[docs]</a><span class="k">class</span> <span class="nc">PresetValidationParameters</span><span class="p">(</span><span class="n">Parameters</span><span class="p">):</span>
|
||||
<span class="k">def</span> <span class="nf">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span>
|
||||
@@ -670,7 +715,7 @@
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">input_filter</span> <span class="o">=</span> <span class="kc">None</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">output_filter</span> <span class="o">=</span> <span class="kc">None</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">pre_network_filter</span> <span class="o">=</span> <span class="n">NoInputFilter</span><span class="p">()</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">full_name_id</span> <span class="o">=</span> <span class="kc">None</span> <span class="c1"># TODO: do we really want to hold this parameter here?</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">full_name_id</span> <span class="o">=</span> <span class="kc">None</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">name</span> <span class="o">=</span> <span class="kc">None</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">is_a_highest_level_agent</span> <span class="o">=</span> <span class="kc">True</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">is_a_lowest_level_agent</span> <span class="o">=</span> <span class="kc">True</span>
|
||||
@@ -684,7 +729,8 @@
|
||||
<div class="viewcode-block" id="TaskParameters"><a class="viewcode-back" href="../../components/additional_parameters.html#rl_coach.base_parameters.TaskParameters">[docs]</a><span class="k">class</span> <span class="nc">TaskParameters</span><span class="p">(</span><span class="n">Parameters</span><span class="p">):</span>
|
||||
<span class="k">def</span> <span class="nf">__init__</span><span class="p">(</span><span class="bp">self</span><span class="p">,</span> <span class="n">framework_type</span><span class="p">:</span> <span class="n">Frameworks</span><span class="o">=</span><span class="n">Frameworks</span><span class="o">.</span><span class="n">tensorflow</span><span class="p">,</span> <span class="n">evaluate_only</span><span class="p">:</span> <span class="nb">bool</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">use_cpu</span><span class="p">:</span> <span class="nb">bool</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
|
||||
<span class="n">experiment_path</span><span class="o">=</span><span class="s1">'/tmp'</span><span class="p">,</span> <span class="n">seed</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">checkpoint_save_secs</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">checkpoint_restore_dir</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
|
||||
<span class="n">checkpoint_save_dir</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">export_onnx_graph</span><span class="p">:</span> <span class="nb">bool</span><span class="o">=</span><span class="kc">False</span><span class="p">):</span>
|
||||
<span class="n">checkpoint_save_dir</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">export_onnx_graph</span><span class="p">:</span> <span class="nb">bool</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">apply_stop_condition</span><span class="p">:</span> <span class="nb">bool</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
|
||||
<span class="n">num_gpu</span><span class="p">:</span> <span class="nb">int</span><span class="o">=</span><span class="mi">1</span><span class="p">):</span>
|
||||
<span class="sd">"""</span>
|
||||
<span class="sd"> :param framework_type: deep learning framework type. currently only tensorflow is supported</span>
|
||||
<span class="sd"> :param evaluate_only: the task will be used only for evaluating the model</span>
|
||||
@@ -695,6 +741,8 @@
|
||||
<span class="sd"> :param checkpoint_restore_dir: the directory to restore the checkpoints from</span>
|
||||
<span class="sd"> :param checkpoint_save_dir: the directory to store the checkpoints in</span>
|
||||
<span class="sd"> :param export_onnx_graph: If set to True, this will export an onnx graph each time a checkpoint is saved</span>
|
||||
<span class="sd"> :param apply_stop_condition: If set to True, this will apply the stop condition defined by reaching a target success rate</span>
|
||||
<span class="sd"> :param num_gpu: number of GPUs to use</span>
|
||||
<span class="sd"> """</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">framework_type</span> <span class="o">=</span> <span class="n">framework_type</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">task_index</span> <span class="o">=</span> <span class="mi">0</span> <span class="c1"># TODO: not really needed</span>
|
||||
@@ -705,7 +753,9 @@
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">checkpoint_restore_dir</span> <span class="o">=</span> <span class="n">checkpoint_restore_dir</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">checkpoint_save_dir</span> <span class="o">=</span> <span class="n">checkpoint_save_dir</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">seed</span> <span class="o">=</span> <span class="n">seed</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">export_onnx_graph</span> <span class="o">=</span> <span class="n">export_onnx_graph</span></div>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">export_onnx_graph</span> <span class="o">=</span> <span class="n">export_onnx_graph</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">apply_stop_condition</span> <span class="o">=</span> <span class="n">apply_stop_condition</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">num_gpu</span> <span class="o">=</span> <span class="n">num_gpu</span></div>
|
||||
|
||||
|
||||
<div class="viewcode-block" id="DistributedTaskParameters"><a class="viewcode-back" href="../../components/additional_parameters.html#rl_coach.base_parameters.DistributedTaskParameters">[docs]</a><span class="k">class</span> <span class="nc">DistributedTaskParameters</span><span class="p">(</span><span class="n">TaskParameters</span><span class="p">):</span>
|
||||
@@ -713,7 +763,7 @@
|
||||
<span class="n">task_index</span><span class="p">:</span> <span class="nb">int</span><span class="p">,</span> <span class="n">evaluate_only</span><span class="p">:</span> <span class="nb">bool</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">num_tasks</span><span class="p">:</span> <span class="nb">int</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
|
||||
<span class="n">num_training_tasks</span><span class="p">:</span> <span class="nb">int</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">use_cpu</span><span class="p">:</span> <span class="nb">bool</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">experiment_path</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">dnd</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
|
||||
<span class="n">shared_memory_scratchpad</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">seed</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">checkpoint_save_secs</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">checkpoint_restore_dir</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span>
|
||||
<span class="n">checkpoint_save_dir</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">export_onnx_graph</span><span class="p">:</span> <span class="nb">bool</span><span class="o">=</span><span class="kc">False</span><span class="p">):</span>
|
||||
<span class="n">checkpoint_save_dir</span><span class="o">=</span><span class="kc">None</span><span class="p">,</span> <span class="n">export_onnx_graph</span><span class="p">:</span> <span class="nb">bool</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">apply_stop_condition</span><span class="p">:</span> <span class="nb">bool</span><span class="o">=</span><span class="kc">False</span><span class="p">):</span>
|
||||
<span class="sd">"""</span>
|
||||
<span class="sd"> :param framework_type: deep learning framework type. currently only tensorflow is supported</span>
|
||||
<span class="sd"> :param evaluate_only: the task will be used only for evaluating the model</span>
|
||||
@@ -732,11 +782,13 @@
|
||||
<span class="sd"> :param checkpoint_restore_dir: the directory to restore the checkpoints from</span>
|
||||
<span class="sd"> :param checkpoint_save_dir: the directory to store the checkpoints in</span>
|
||||
<span class="sd"> :param export_onnx_graph: If set to True, this will export an onnx graph each time a checkpoint is saved</span>
|
||||
<span class="sd"> :param apply_stop_condition: If set to True, this will apply the stop condition defined by reaching a target success rate</span>
|
||||
|
||||
<span class="sd"> """</span>
|
||||
<span class="nb">super</span><span class="p">()</span><span class="o">.</span><span class="fm">__init__</span><span class="p">(</span><span class="n">framework_type</span><span class="o">=</span><span class="n">framework_type</span><span class="p">,</span> <span class="n">evaluate_only</span><span class="o">=</span><span class="n">evaluate_only</span><span class="p">,</span> <span class="n">use_cpu</span><span class="o">=</span><span class="n">use_cpu</span><span class="p">,</span>
|
||||
<span class="n">experiment_path</span><span class="o">=</span><span class="n">experiment_path</span><span class="p">,</span> <span class="n">seed</span><span class="o">=</span><span class="n">seed</span><span class="p">,</span> <span class="n">checkpoint_save_secs</span><span class="o">=</span><span class="n">checkpoint_save_secs</span><span class="p">,</span>
|
||||
<span class="n">checkpoint_restore_dir</span><span class="o">=</span><span class="n">checkpoint_restore_dir</span><span class="p">,</span> <span class="n">checkpoint_save_dir</span><span class="o">=</span><span class="n">checkpoint_save_dir</span><span class="p">,</span>
|
||||
<span class="n">export_onnx_graph</span><span class="o">=</span><span class="n">export_onnx_graph</span><span class="p">)</span>
|
||||
<span class="n">export_onnx_graph</span><span class="o">=</span><span class="n">export_onnx_graph</span><span class="p">,</span> <span class="n">apply_stop_condition</span><span class="o">=</span><span class="n">apply_stop_condition</span><span class="p">)</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">parameters_server_hosts</span> <span class="o">=</span> <span class="n">parameters_server_hosts</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">worker_hosts</span> <span class="o">=</span> <span class="n">worker_hosts</span>
|
||||
<span class="bp">self</span><span class="o">.</span><span class="n">job_type</span> <span class="o">=</span> <span class="n">job_type</span>
|
||||
|
||||
Reference in New Issue
Block a user