Multiple tab components

README.md

@@ -587,13 +587,111 @@ trainer.fit(
 )
 ```
 
-**7. Tabular with a multi-target loss**
+**7. A two-tower model**
+
+This is a popular model in the context of recommendation systems. Let's say we
+have a tabular dataset formed my triples (user features, item features,
+target). We can create a two-tower model where the user and item features are
+passed through two separate models and then "fused" via a dot product.
+
+<p align="center">
+  <img width="350" src="docs/figures/arch_7.png">
+</p>
+
+
+```python
+import numpy as np
+import pandas as pd
+
+from pytorch_widedeep import Trainer
+from pytorch_widedeep.preprocessing import TabPreprocessor
+from pytorch_widedeep.models import TabMlp, WideDeep, ModelFuser
+
+# Let's create the interaction dataset
+# user_features dataframe
+np.random.seed(42)
+user_ids = np.arange(1, 101)
+ages = np.random.randint(18, 60, size=100)
+genders = np.random.choice(["male", "female"], size=100)
+locations = np.random.choice(["city_a", "city_b", "city_c", "city_d"], size=100)
+user_features = pd.DataFrame(
+    {"id": user_ids, "age": ages, "gender": genders, "location": locations}
+)
+
+# item_features dataframe
+item_ids = np.arange(1, 101)
+prices = np.random.uniform(10, 500, size=100).round(2)
+colors = np.random.choice(["red", "blue", "green", "black"], size=100)
+categories = np.random.choice(["electronics", "clothing", "home", "toys"], size=100)
+
+item_features = pd.DataFrame(
+    {"id": item_ids, "price": prices, "color": colors, "category": categories}
+)
+
+# Interactions dataframe
+interaction_user_ids = np.random.choice(user_ids, size=1000)
+interaction_item_ids = np.random.choice(item_ids, size=1000)
+purchased = np.random.choice([0, 1], size=1000, p=[0.7, 0.3])
+interactions = pd.DataFrame(
+    {
+        "user_id": interaction_user_ids,
+        "item_id": interaction_item_ids,
+        "purchased": purchased,
+    }
+)
+user_item_purchased = interactions.merge(
+    user_features, left_on="user_id", right_on="id"
+).merge(item_features, left_on="item_id", right_on="id")
+
+# Users
+tab_preprocessor_user = TabPreprocessor(
+    cat_embed_cols=["gender", "location"],
+    continuous_cols=["age"],
+)
+X_user = tab_preprocessor_user.fit_transform(user_item_purchased)
+tab_mlp_user = TabMlp(
+    column_idx=tab_preprocessor_user.column_idx,
+    cat_embed_input=tab_preprocessor_user.cat_embed_input,
+    continuous_cols=["age"],
+    mlp_hidden_dims=[16, 8],
+    mlp_dropout=[0.2, 0.2],
+)
+
+# Items
+tab_preprocessor_item = TabPreprocessor(
+    cat_embed_cols=["color", "category"],
+    continuous_cols=["price"],
+)
+X_item = tab_preprocessor_item.fit_transform(user_item_purchased)
+tab_mlp_item = TabMlp(
+    column_idx=tab_preprocessor_item.column_idx,
+    cat_embed_input=tab_preprocessor_item.cat_embed_input,
+    continuous_cols=["price"],
+    mlp_hidden_dims=[16, 8],
+    mlp_dropout=[0.2, 0.2],
+)
+
+two_tower_model = ModelFuser([tab_mlp_user, tab_mlp_item], fusion_method="dot")
+
+model = WideDeep(deeptabular=two_tower_model)
+
+trainer = Trainer(model, objective="binary")
+
+trainer.fit(
+    X_tab=[X_user, X_item],
+    target=interactions.purchased.values,
+    n_epochs=1,
+    batch_size=32,
+)
+```
+
+**8. Tabular with a multi-target loss**
 
 This one is "a bonus" to illustrate the use of multi-target losses, more than
 actually a different architecture.
 
 <p align="center">
-  <img width="200" src="docs/figures/arch_7.png">
+  <img width="200" src="docs/figures/arch_8.png">
 </p>
 
 

VERSION

@@ -1 +1 @@
-1.6.2
+1.6.3

docs/examples.rst

@@ -17,5 +17,4 @@ them to address different problems
 * `HyperParameter Tuning With RayTune <https://github.com/jrzaurin/pytorch-widedeep/blob/master/examples/notebooks/10_hyperParameter_tuning_w_raytune_n_wnb.ipynb>`__
 * `Model Uncertainty Prediction <https://github.com/jrzaurin/pytorch-widedeep/blob/master/examples/notebooks/13_Model_Uncertainty_prediction.ipynb>`__
 * `Bayesian Models <https://github.com/jrzaurin/pytorch-widedeep/blob/master/examples/notebooks/14_bayesian_models.ipynb>`__
-* `Deep Imbalanced Regression <https://github.com/jrzaurin/pytorch-widedeep/blob/master/examples/notebooks/15_DIR-LDS_and_FDS.ipynb>`__
 

examples/scripts/readme_snippets.py

@@ -407,7 +407,90 @@ def output_dim(self):
 )
 
 
-# 7. Simply Tabular with a multi-target loss
+# 7. A Two tower model
+np.random.seed(42)
+
+# user_features dataframe
+user_ids = np.arange(1, 101)
+ages = np.random.randint(18, 60, size=100)
+genders = np.random.choice(["male", "female"], size=100)
+locations = np.random.choice(["city_a", "city_b", "city_c", "city_d"], size=100)
+user_features = pd.DataFrame(
+    {"id": user_ids, "age": ages, "gender": genders, "location": locations}
+)
+
+# item_features dataframe
+item_ids = np.arange(1, 101)
+prices = np.random.uniform(10, 500, size=100).round(2)
+colors = np.random.choice(["red", "blue", "green", "black"], size=100)
+categories = np.random.choice(["electronics", "clothing", "home", "toys"], size=100)
+
+item_features = pd.DataFrame(
+    {"id": item_ids, "price": prices, "color": colors, "category": categories}
+)
+
+# Interactions dataframe
+interaction_user_ids = np.random.choice(user_ids, size=1000)
+interaction_item_ids = np.random.choice(item_ids, size=1000)
+purchased = np.random.choice([0, 1], size=1000, p=[0.7, 0.3])
+interactions = pd.DataFrame(
+    {
+        "user_id": interaction_user_ids,
+        "item_id": interaction_item_ids,
+        "purchased": purchased,
+    }
+)
+user_item_purchased = interactions.merge(
+    user_features, left_on="user_id", right_on="id"
+).merge(item_features, left_on="item_id", right_on="id")
+
+
+# Users
+tab_preprocessor_user = TabPreprocessor(
+    cat_embed_cols=["gender", "location"],
+    continuous_cols=["age"],
+)
+X_user = tab_preprocessor_user.fit_transform(user_item_purchased)
+tab_mlp_user = TabMlp(
+    column_idx=tab_preprocessor_user.column_idx,
+    cat_embed_input=tab_preprocessor_user.cat_embed_input,
+    continuous_cols=["age"],
+    mlp_hidden_dims=[16, 8],
+    mlp_dropout=[0.2, 0.2],
+)
+
+# Items
+tab_preprocessor_item = TabPreprocessor(
+    cat_embed_cols=["color", "category"],
+    continuous_cols=["price"],
+)
+X_item = tab_preprocessor_item.fit_transform(user_item_purchased)
+tab_mlp_item = TabMlp(
+    column_idx=tab_preprocessor_item.column_idx,
+    cat_embed_input=tab_preprocessor_item.cat_embed_input,
+    continuous_cols=["price"],
+    mlp_hidden_dims=[16, 8],
+    mlp_dropout=[0.2, 0.2],
+)
+
+two_tower_model = ModelFuser([tab_mlp_user, tab_mlp_item], fusion_method="dot")
+
+model = WideDeep(deeptabular=two_tower_model)
+
+trainer = Trainer(
+    model,
+    objective="binary",
+)
+
+trainer.fit(
+    X_tab=[X_user, X_item],
+    target=interactions.purchased.values,
+    n_epochs=1,
+    batch_size=32,
+)
+
+
+# 8. Simply Tabular with a multi-target loss
 
 # let's add a second target to the dataframe
 df["target2"] = [random.choice([0, 1]) for _ in range(100)]

mkdocs/mkdocs.yml

@@ -52,15 +52,14 @@ nav:
         - 12_ZILNLoss_origkeras_vs_pytorch_widedeep: examples/12_ZILNLoss_origkeras_vs_pytorch_widedeep.ipynb
         - 13_model_uncertainty_prediction: examples/13_model_uncertainty_prediction.ipynb
         - 14_bayesian_models: examples/14_bayesian_models.ipynb
-        - 15_DIR-LDS_and_FDS: examples/15_DIR-LDS_and_FDS.ipynb
-        - 16_Self-Supervised Pre-Training pt 1: examples/16_Self_Supervised_Pretraning_pt1.ipynb
-        - 16_Self-Supervised Pre-Training pt 2: examples/16_Self_Supervised_Pretraning_pt2.ipynb
-        - 17_Usign-a-custom-hugging-face-model: examples/17_Usign_a_custom_hugging_face_model.ipynb
-        - 18_feature_importance_via_attention_weights: examples/18_feature_importance_via_attention_weights.ipynb
-        - 19_wide_and_deep_for_recsys_pt1: examples/19_wide_and_deep_for_recsys_pt1.ipynb
-        - 19_wide_and_deep_for_recsys_pt2: examples/19_wide_and_deep_for_recsys_pt2.ipynb
-        - 20_load_from_folder_functionality: examples/20_load_from_folder_functionality.ipynb
-        - 21-Using-huggingface-within-widedeep: examples/21_Using_huggingface_within_widedeep.ipynb
+        - 15_Self-Supervised Pre-Training pt 1: examples/16_Self_Supervised_Pretraning_pt1.ipynb
+        - 15_Self-Supervised Pre-Training pt 2: examples/16_Self_Supervised_Pretraning_pt2.ipynb
+        - 16_Usign-a-custom-hugging-face-model: examples/17_Usign_a_custom_hugging_face_model.ipynb
+        - 17_feature_importance_via_attention_weights: examples/18_feature_importance_via_attention_weights.ipynb
+        - 18_wide_and_deep_for_recsys_pt1: examples/19_wide_and_deep_for_recsys_pt1.ipynb
+        - 18_wide_and_deep_for_recsys_pt2: examples/19_wide_and_deep_for_recsys_pt2.ipynb
+        - 19_load_from_folder_functionality: examples/20_load_from_folder_functionality.ipynb
+        - 20-Using-huggingface-within-widedeep: examples/21_Using_huggingface_within_widedeep.ipynb
     - Contributing: contributing.md
 
 theme: