Update

README.md

@@ -1,6 +1,9 @@
 ---
 license: apache-2.0
 ---
+## How to use the discriminator in `transformers`
+```python
+```
 
 ## How to use the discriminator in `transformers` on a custom dataset
 (Heavily based on: https://github.com/huggingface/notebooks/blob/master/examples/text_classification-tf.ipynb)
@@ -54,7 +57,7 @@ batches_per_epoch = math.ceil(len(train_dataset) / batch_size)
 total_train_steps = int(batches_per_epoch * num_epochs)
 
 optimizer, schedule = create_optimizer(
-    init_lr=1e-5, num_warmup_steps=1, num_train_steps=total_train_steps
+    init_lr=1e-6, num_warmup_steps=1, num_train_steps=total_train_steps
 )
 
 discriminator.compile(optimizer=optimizer, loss=loss)
@@ -72,3 +75,20 @@ print(text, ":", label)
 # ideally [v01214265 -> take.v.04 -> "get into one's hands, take physically"], but probably only with a better dataset
 
 ```
+
+## How to use in a Rasa pipeline
+The model can integrated in a Rasa pipeline through a [`LanguageModelFeaturizer`](https://rasa.com/docs/rasa/components#languagemodelfeaturizer)
+```yaml
+recipe: default.v1
+language: en
+
+pipeline:
+# See https://rasa.com/docs/rasa/tuning-your-model for more information.
+  ...
+  - name: "WhitespaceTokenizer"
+    ...
+  - name: LanguageModelFeaturizer
+    model_name: "distilbert"
+    model_weights: "Aureliano/distilbert-base-uncased-if"
+  ...
+```

config.json

@@ -1,8 +1,8 @@
 {
-  "_name_or_path": "wn_full_classifier-trainer",
+  "_name_or_path": "wn_classifier-trainer",
   "activation": "gelu",
   "architectures": [
-    "DistilBertModel"
+    "DistilBertForSequenceClassification"
   ],
   "attention_dropout": 0.1,
   "dim": 768,

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:12eaf2d9db88163a36fbbc5e45331658bc35d26421cb061f05c473c7708a30d4
-size 265487161
+oid sha256:90d57abc8b44939285b112eb10a677035d03d16c6441e3dd2537db9335c1f53e
+size 268111281

tf_model.h5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d789b0b93a1504985b62a35e46433ea14548042e4bdb2e9710619198d6a7b62e
-size 265571752
+oid sha256:92875108e5ce200e5c4bf672bc691f1bffe63ffe18cad7fb6ad4d773e5a99251
+size 268203800