Remove config dir from tutorial

himkt · Nov 14, 2020 · 77661b8 · 77661b8
1 parent b2b0adc
commit 77661b8
Show file tree

Hide file tree

Showing 3 changed files with 15 additions and 17 deletions.
diff --git a/docs/source/tutorial/advanced_optuna_configuration.rst b/docs/source/tutorial/advanced_optuna_configuration.rst
@@ -2,10 +2,7 @@ Advanced configuration for Optuna
 =================================
 
 You can choose a pruner/sample implemented in Optuna.
-To specify a pruner/sampler, create a JSON config file
-
-The example of `optuna.json <./config/optuna.json>`_ looks like:
-
+To specify a pruner/sampler, create a JSON config file.
 
 - ``optuna.json``
 
@@ -90,6 +87,7 @@ Next, we have to add `optuna_pruner` to `epoch_callbacks`.
     },
     trainer: {
       cuda_device: cuda_device,
+      // NOTE add `optuna_pruner` here!
       epoch_callbacks: [
         {
           type: 'optuna_pruner',
@@ -110,8 +108,8 @@ Finally, you can run optimization with pruning:
 .. code-block:: bash
 
     poetry run allennlp tune \
-        config/imdb_optuna_with_pruning.jsonnet \
-        config/hparams.json \
-        --optuna-param-path config/optuna.json \
+        imdb_optuna_with_pruning.jsonnet \
+        hparams.json \
+        --optuna-param-path optuna.json \
         --serialization-dir result/hpo \
         --study-name test-with-pruning
diff --git a/docs/source/tutorial/hyperparameter_optimization_at_scale.rst b/docs/source/tutorial/hyperparameter_optimization_at_scale.rst
@@ -9,9 +9,9 @@ You can easily run distributed optimization by adding an option
 .. code-block:: bash
 
     poetry run allennlp tune \
-        config/imdb_optuna.jsonnet \
-        config/hparams.json \
-        --optuna-param-path config/optuna.json \
+        imdb_optuna.jsonnet \
+        hparams.json \
+        --optuna-param-path optuna.json \
         --serialization-dir result \
         --study-name test \
         --skip-if-exists
@@ -26,10 +26,10 @@ the command should be like following:
 .. code-block:: bash
 
     poetry run allennlp tune \
-        config/imdb_optuna.jsonnet \
-        config/hparams.json \
-        --optuna-param-path config/optuna.json \
-        --serialization-dir result \
+        imdb_optuna.jsonnet \
+        hparams.json \
+        --optuna-param-path optuna.json \
+        --serialization-dir result/distributed \
         --study-name test \
         --storage mysql://<user_name>:<passwd>@<db_host>/<db_name> \
         --skip-if-exists

diff --git a/docs/source/tutorial/run_allennlp_optuna.rst b/docs/source/tutorial/run_allennlp_optuna.rst
@@ -10,8 +10,8 @@ You can optimize hyperparameters by:
 .. code-block:: bash
 
     poetry run allennlp tune \
-        config/imdb_optuna.jsonnet \
-        config/hparams.json \
+        imdb_optuna.jsonnet \
+        hparams.json \
         --serialization-dir result \
         --study-name test
 
@@ -31,6 +31,6 @@ Retrain a model with optimized hyperparameters
 .. code-block:: bash
 
     poetry run allennlp retrain \
-        config/imdb_optuna.jsonnet \
+        imdb_optuna.jsonnet \
         --serialization-dir retrain_result \
         --study-name test