Fix DeepRec titles; add levels under DeepRec docs (#1428)

recommenders-team · Jun 4, 2021 · c76d679 · c76d679
1 parent 628e88f
commit c76d679
Show file tree

Hide file tree

Showing 4 changed files with 66 additions and 48 deletions.
diff --git a/docs/source/recommender.rst b/docs/source/recommender.rst
@@ -1,89 +1,94 @@
 .. _recommender:
 
 Recommender algorithms module
-*******************************
+##############################
 
 Recommender system algorithms and utilities.
 
 Cornac
-===============================
+******************************
 
 .. automodule:: reco_utils.recommender.cornac.cornac_utils
     :members:
 
 DeepRec
-===============================
+******************************
 
-base_model
--------------------------------
+Base model
+==================
 .. automodule:: reco_utils.recommender.deeprec.models.base_model
     :members:
 
-dkn
--------------------------------
+DKN
+==================
 .. automodule:: reco_utils.recommender.deeprec.models.dkn
     :members:
 
-dkn_item2item
--------------------------------
+DKN item-to-item
+==================
 .. automodule:: reco_utils.recommender.deeprec.models.dkn_item2item
     :members:
 
+LightGCN
+==================
+.. automodule:: reco_utils.recommender.deeprec.models.graphrec.lightgcn
+    :members:
+
 xDeepFM
--------------------------------
+==============
 .. automodule:: reco_utils.recommender.deeprec.models.xDeepFM
     :members:
 
-graphrec
--------------------------------
-.. automodule:: reco_utils.recommender.deeprec.models.graphrec.lightgcn
-    :members:
+Sequential models
+==================
 
-sequential_base_model
--------------------------------
+Sequential base model
+---------------------------
 .. automodule:: reco_utils.recommender.deeprec.models.sequential.sequential_base_model
     :members:
 
-asvd
--------------------------------
+A2SVD
+-------------
 .. automodule:: reco_utils.recommender.deeprec.models.sequential.asvd
     :members:
 
-caser
--------------------------------
+Caser
+----------------
 .. automodule:: reco_utils.recommender.deeprec.models.sequential.caser
     :members:
 
-gru4rec
--------------------------------
+GRU4Rec
+--------------
 .. automodule:: reco_utils.recommender.deeprec.models.sequential.gru4rec
     :members:
 
-nextitnet
--------------------------------
+NextItNet
+--------------
 .. automodule:: reco_utils.recommender.deeprec.models.sequential.nextitnet
     :members:
 
-rnn_cell_implement
--------------------------------
+RNN Cells
+------------
 .. automodule:: reco_utils.recommender.deeprec.models.sequential.rnn_cell_implement
     :members:
 
-sli_rec
--------------------------------
-.. automodule:: reco_utils.recommender.deeprec.models.sequential.sli_rec
-    :members:
-
-sum_cells
--------------------------------
+Cell for Sequential User Matrix
+---------------------------------
 .. automodule:: reco_utils.recommender.deeprec.models.sequential.sum_cells
     :members:
 
-sum
--------------------------------
+Sequential User Matrix
+-------------------------
 .. automodule:: reco_utils.recommender.deeprec.models.sequential.sum
     :members:
 
+SLI Rec
+-------------
+.. automodule:: reco_utils.recommender.deeprec.models.sequential.sli_rec
+    :members:
+
+Iterators
+===========
 
 .. automodule:: reco_utils.recommender.deeprec.io.iterator
     :members:
@@ -100,20 +105,26 @@ sum
 .. automodule:: reco_utils.recommender.deeprec.io.sequential_iterator
     :members:
 
+Data processing utilities
+===========================
+
 .. automodule:: reco_utils.recommender.deeprec.DataModel.ImplicitCF
     :members:
 
+Utilities
+============
+
 .. automodule:: reco_utils.recommender.deeprec.deeprec_utils
     :members:
 
 FastAI
-===============================
+******************************
 
 .. automodule:: reco_utils.recommender.fastai.fastai_utils
     :members:
 
 Geoimc
-===============================
+******************************
 
 .. automodule:: reco_utils.recommender.geoimc.geoimc_algorithm
     :members:
@@ -128,19 +139,19 @@ Geoimc
     :members:
 
 LightFM
-===============================
+******************************
 
 .. automodule:: reco_utils.recommender.lightfm.lightfm_utils
     :members:
 
 LightGBM
-===============================
+******************************
 
 .. automodule:: reco_utils.recommender.lightgbm.lightgbm_utils
     :members:
 
 NCF
-===============================
+******************************
 
 .. automodule:: reco_utils.recommender.ncf.dataset
     :members:
@@ -149,7 +160,7 @@ NCF
     :members:
 
 Newsrec
-===============================
+******************************
 
 .. automodule:: reco_utils.recommender.newsrec.io.mind_all_iterator
     :members:
@@ -179,14 +190,14 @@ Newsrec
     :members:
 
 RBM
-===============================
+******************************
 
 .. automodule:: reco_utils.recommender.rbm.rbm
     :members:
 
 
 RLRMC
-===============================
+******************************
 
 .. automodule:: reco_utils.recommender.rlrmc.RLRMCalgorithm
     :members:
@@ -199,28 +210,28 @@ RLRMC
 
 
 SAR
-===============================
+******************************
 
 .. automodule:: reco_utils.recommender.sar.sar_singlenode
     :members:
 
 
 Surprise
-===============================
+******************************
 
 .. automodule:: reco_utils.recommender.surprise.surprise_utils
     :members:
 
 
 TF-IDF
-===============================
+******************************
 
 .. automodule:: reco_utils.recommender.tfidf.tfidf_utils
     :members:
 
 
 VAE
-===============================
+******************************
 
 .. automodule:: reco_utils.recommender.vae.multinomial_vae
     :members:
@@ -230,7 +241,7 @@ VAE
 
 
 Wide & Deep
-===============================
+******************************
 
 .. automodule:: reco_utils.recommender.wide_deep.wide_deep_utils
     :members:
diff --git a/reco_utils/recommender/deeprec/models/base_model.py b/reco_utils/recommender/deeprec/models/base_model.py
@@ -15,6 +15,8 @@
 
 
 class BaseModel:
+    """Base class for models"""
+
     def __init__(self, hparams, iterator_creator, graph=None, seed=None):
         """Initializing the model. Create common logics which are needed by all deeprec models, such as loss function,
         parameter set.

diff --git a/reco_utils/recommender/deeprec/models/dkn_item2item.py b/reco_utils/recommender/deeprec/models/dkn_item2item.py
@@ -13,6 +13,9 @@
 
 
 class DKNItem2Item(DKN):
+    """Class for item-to-item recommendations using DKN.
+    See https://github.com/microsoft/recommenders/blob/kdd2020_tutorial/scenarios/academic/KDD2020-tutorial/step4_run_dkn_item2item.ipynb"""
+
     def _compute_data_loss(self):
         logits = self.pred
         data_loss = -1 * tf.reduce_sum(tf.math.log(logits[:, 0] + 1e-10))

diff --git a/reco_utils/recommender/deeprec/models/sequential/sequential_base_model.py b/reco_utils/recommender/deeprec/models/sequential/sequential_base_model.py
@@ -15,6 +15,8 @@
 
 
 class SequentialBaseModel(BaseModel):
+    """Base class for sequential models"""
+
     def __init__(self, hparams, iterator_creator, graph=None, seed=None):
         """Initializing the model. Create common logics which are needed by all sequential models, such as loss function, 
         parameter set.