Add Dolly NER.v3 usage example. (#302)

usage_examples/ner_dolly/README.md

@@ -26,6 +26,9 @@ python run_pipeline.py \
     ./examples.yml
 ```
 
+By default this uses v3 if the NER recipe, which leverages a chain-of-thought prompt. If you want to run this with v2, 
+use `fewshot_v2.cfg`, `zeroshot_v2`.cfg and `examples_v2.yml` instead.
+
 By default, the pipeline assigns `PERSON`, `ORGANIZATION` or `LOCATION` labels
 for each entity. You can change these labels by updating the
 `zeroshot.cfg` configuration file.

usage_examples/ner_dolly/examples.yml

@@ -1,15 +1,18 @@
 - text: Jack and Jill went up the hill.
-  entities:
-    PERSON:
-      - Jack
-      - Jill
-    LOCATION:
-      - hill
-- text: Jack fell down and broke his crown.
-  entities:
-    PERSON:
-      - Jack
-- text: Jill came tumbling after.
-  entities:
-    PERSON:
-      - Jill
+  spans:
+    - text: Jack
+      is_entity: true
+      label: PERSON
+      reason: is the name of a person
+    - text: Jill
+      is_entity: true
+      label: PERSON
+      reason: is the name of a person
+    - text: went up
+      is_entity: false
+      label: ==NONE==
+      reason: is a verb
+    - text: hill
+      is_entity: true
+      label: LOCATION
+      reason: is a location

usage_examples/ner_dolly/examples_v2.yml

@@ -0,0 +1,15 @@
+- text: Jack and Jill went up the hill.
+  entities:
+    PERSON:
+      - Jack
+      - Jill
+    LOCATION:
+      - hill
+- text: Jack fell down and broke his crown.
+  entities:
+    PERSON:
+      - Jack
+- text: Jill came tumbling after.
+  entities:
+    PERSON:
+      - Jill

usage_examples/ner_dolly/fewshot.cfg

@@ -16,7 +16,7 @@ factory = "llm"
 name = "dolly-v2-3b"
 
 [components.llm.task]
-@llm_tasks = "spacy.NER.v2"
+@llm_tasks = "spacy.NER.v3"
 labels = PERSON,ORGANISATION,LOCATION
 
 [components.llm.task.examples]

usage_examples/ner_dolly/fewshot_v2.cfg

@@ -0,0 +1,27 @@
+[paths]
+examples = null
+
+[nlp]
+lang = "en"
+pipeline = ["llm"]
+batch_size = 128
+
+[components]
+
+[components.llm]
+factory = "llm"
+
+[components.llm.model]
+@llm_models = "spacy.Dolly.v1"
+name = "dolly-v2-3b"
+
+[components.llm.task]
+@llm_tasks = "spacy.NER.v2"
+labels = PERSON,ORGANISATION,LOCATION
+
+[components.llm.task.examples]
+@misc = "spacy.FewShotReader.v1"
+path = ${paths.examples}
+
+[components.llm.task.normalizer]
+@misc = "spacy.LowercaseNormalizer.v1"

usage_examples/ner_dolly/zeroshot.cfg

@@ -13,7 +13,7 @@ factory = "llm"
 name = "dolly-v2-3b"
 
 [components.llm.task]
-@llm_tasks = "spacy.NER.v2"
+@llm_tasks = "spacy.NER.v3"
 labels = PERSON,ORGANISATION,LOCATION
 examples = null
 

usage_examples/ner_dolly/zeroshot_v2.cfg

@@ -0,0 +1,21 @@
+[nlp]
+lang = "en"
+pipeline = ["llm"]
+batch_size = 128
+
+[components]
+
+[components.llm]
+factory = "llm"
+
+[components.llm.model]
+@llm_models = "spacy.Dolly.v1"
+name = "dolly-v2-3b"
+
+[components.llm.task]
+@llm_tasks = "spacy.NER.v2"
+labels = PERSON,ORGANISATION,LOCATION
+examples = null
+
+[components.llm.task.normalizer]
+@misc = "spacy.LowercaseNormalizer.v1"

usage_examples/tests/test_usage_examples.py

@@ -13,7 +13,7 @@
 
 @pytest.mark.gpu
 @pytest.mark.skipif(not has_torch_cuda_gpu, reason="needs GPU & CUDA")
-@pytest.mark.parametrize("config_name", ("fewshot.cfg", "zeroshot.cfg"))
+@pytest.mark.parametrize("config_name", ("fewshot_v2.cfg", "zeroshot_v2.cfg"))
 def test_ner_dolly(config_name: str):
     """Test NER Dolly usage example.
     config_name (str): Name of config file to use.
@@ -22,7 +22,23 @@ def test_ner_dolly(config_name: str):
     ner_dolly.run_pipeline(
         text="text",
         config_path=path / config_name,
-        examples_path=None if config_name == "zeroshot.cfg" else path / "examples.yml",
+        examples_path=None if "zeroshot" in config_name else path / "examples_v2.yml",
+        verbose=False,
+    )
+
+
+@pytest.mark.gpu
+@pytest.mark.skipif(not has_torch_cuda_gpu, reason="needs GPU & CUDA")
+@pytest.mark.parametrize("config_name", ("fewshot.cfg", "zeroshot.cfg"))
+def test_ner_v3_dolly(config_name: str):
+    """Test NER Dolly usage example.
+    config_name (str): Name of config file to use.
+    """
+    path = _USAGE_EXAMPLE_PATH / "ner_dolly"
+    ner_dolly.run_pipeline(
+        text="text",
+        config_path=path / config_name,
+        examples_path=None if "zeroshot" in config_name else path / "examples.yml",
         verbose=False,
     )