Skip to content

Commit 5e8856f

Browse files
DevinTDHaDevinTDHa
committed
Docs: Fix various Issues [skip test]
- Fix jekyll template issues for JanusForMultiModal and SmolVLMTransformer - Fix various sphinx python documentation generation issues
1 parent 47f96e6 commit 5e8856f

28 files changed

Lines changed: 146 additions & 157 deletions

‎docs/en/transformer_entries/JanusForMultiModal.md‎

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -17,6 +17,8 @@ val visualQA = JanusForMultiModal.pretrained()
1717
.setInputCols("image_assembler")
1818
.setOutputCol("answer")
1919
```
20+
{%- endcapture -%}
21+
2022
{%- capture input_anno -%}
2123
IMAGE
2224
{%- endcapture -%}

‎docs/en/transformer_entries/SmolVLMTransformer.md‎

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -16,6 +16,8 @@ val visualQA = SmolVLMTransformer.pretrained()
1616
.setInputCols("image_assembler")
1717
.setOutputCol("answer")
1818
```
19+
{%- endcapture -%}
20+
1921
{%- capture input_anno -%}
2022
IMAGE
2123
{%- endcapture -%}

‎python/docs/conf.py‎

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -99,6 +99,7 @@
9999
autoapi_options = [
100100
"members",
101101
"show-module-summary",
102+
"undoc-members"
102103
]
103104
autoapi_type = "python"
104105
autoapi_dirs = ["../sparknlp"]

‎python/sparknlp/annotator/cv/gemma3_for_multimodal.py‎

Lines changed: 5 additions & 10 deletions
Original file line numberDiff line numberDiff line change
@@ -83,23 +83,18 @@ class Gemma3ForMultiModal(AnnotatorModel,
8383
>>> from sparknlp.annotator import *
8484
>>> from pyspark.ml import Pipeline
8585
>>> from pyspark.sql.functions import lit
86-
>>>
8786
>>> imageDF = spark.read.format("image").load(images_path)
88-
>>> testDF = imageDF.withColumn("text", lit("<bos><start_of_turn>user\nYou are a helpful assistant.\n\n<start_of_image>Describe this image in detail.<end_of_turn>\n<start_of_turn>model\n"))
89-
>>>
90-
>>> imageAssembler = ImageAssembler() \
91-
... .setInputCol("image") \
87+
>>> testDF = imageDF.withColumn("text", lit("<bos><start_of_turn>user\\nYou are a helpful assistant.\\n\\n<start_of_image>Describe this image in detail.<end_of_turn>\\n<start_of_turn>model\\n"))
88+
>>> imageAssembler = ImageAssembler() \\
89+
... .setInputCol("image") \\
9290
... .setOutputCol("image_assembler")
93-
>>>
94-
>>> visualQA = Gemma3ForMultiModal.pretrained() \
95-
... .setInputCols("image_assembler") \
91+
>>> visualQA = Gemma3ForMultiModal.pretrained() \\
92+
... .setInputCols("image_assembler") \\
9693
... .setOutputCol("answer")
97-
>>>
9894
>>> pipeline = Pipeline().setStages([
9995
... imageAssembler,
10096
... visualQA
10197
... ])
102-
>>>
10398
>>> result = pipeline.fit(testDF).transform(testDF)
10499
>>> result.select("image_assembler.origin", "answer.result").show(truncate=False)
105100
"""

‎python/sparknlp/annotator/cv/internvl_for_multimodal.py‎

Lines changed: 7 additions & 11 deletions
Original file line numberDiff line numberDiff line change
@@ -20,8 +20,8 @@ class InternVLForMultiModal(AnnotatorModel,
2020
- Optimized for deployment with int4 quantization
2121
2222
Pretrained models can be loaded with :meth:`.pretrained` of the companion object:
23-
>>> visualQA = InternVLForMultiModal.pretrained() \
24-
... .setInputCols("image_assembler") \
23+
>>> visualQA = InternVLForMultiModal.pretrained() \\
24+
... .setInputCols("image_assembler") \\
2525
... .setOutputCol("answer")
2626
2727
The default model is `"internvl2_5_1b_int4"`, if no name is provided.
@@ -49,21 +49,17 @@ class InternVLForMultiModal(AnnotatorModel,
4949
>>> from sparknlp.annotator import *
5050
>>> from pyspark.ml import Pipeline
5151
>>> from pyspark.sql.functions import lit
52-
5352
>>> image_df = spark.read.format("image").load(path=images_path)
5453
>>> test_df = image_df.withColumn(
5554
... "text",
56-
... lit("<|im_start|><image>\nDescribe this image in detail.<|im_end|><|im_start|>assistant\n")
55+
... lit("<|im_start|><image>\\nDescribe this image in detail.<|im_end|><|im_start|>assistant\\n")
5756
... )
58-
59-
>>> imageAssembler = ImageAssembler() \
60-
... .setInputCol("image") \
57+
>>> imageAssembler = ImageAssembler() \\
58+
... .setInputCol("image") \\
6159
... .setOutputCol("image_assembler")
62-
63-
>>> visualQA = InternVLForMultiModal.pretrained() \
64-
... .setInputCols("image_assembler") \
60+
>>> visualQA = InternVLForMultiModal.pretrained() \\
61+
... .setInputCols("image_assembler") \\
6562
... .setOutputCol("answer")
66-
6763
>>> pipeline = Pipeline().setStages([
6864
... imageAssembler,
6965
... visualQA

‎python/sparknlp/annotator/cv/janus_for_multimodal.py‎

Lines changed: 8 additions & 13 deletions
Original file line numberDiff line numberDiff line change
@@ -36,8 +36,9 @@ class JanusForMultiModal(AnnotatorModel,
3636
and for image generation, it uses a tokenizer with a downsample rate of 16.
3737
3838
Pretrained models can be loaded with :meth:`.pretrained` of the companion object:
39-
>>> visualQAClassifier = JanusForMultiModal.pretrained() \
40-
... .setInputCols(["image_assembler"]) \
39+
40+
>>> visualQAClassifier = JanusForMultiModal.pretrained() \\
41+
... .setInputCols(["image_assembler"]) \\
4142
... .setOutputCol("answer")
4243
4344
The default model is `"janus_1_3b_int4"`, if no name is provided.
@@ -73,29 +74,23 @@ class JanusForMultiModal(AnnotatorModel,
7374
>>> from sparknlp.annotator import *
7475
>>> from pyspark.ml import Pipeline
7576
>>> from pyspark.sql.functions import lit
76-
7777
>>> image_df = SparkSessionForTest.spark.read.format("image").load(path=images_path)
7878
>>> test_df = image_df.withColumn(
7979
... "text",
80-
... lit("User: <image_placeholder>Describe image in details\n\nAssistant:")
80+
... lit("User: <image_placeholder>Describe image in details\\n\\nAssistant:")
8181
... )
82-
83-
>>> imageAssembler = ImageAssembler() \
84-
... .setInputCol("image") \
82+
>>> imageAssembler = ImageAssembler() \\
83+
... .setInputCol("image") \\
8584
... .setOutputCol("image_assembler")
86-
87-
>>> visualQAClassifier = JanusForMultiModal.pretrained() \
88-
... .setInputCols("image_assembler") \
85+
>>> visualQAClassifier = JanusForMultiModal.pretrained() \\
86+
... .setInputCols("image_assembler") \\
8987
... .setOutputCol("answer")
90-
9188
>>> pipeline = Pipeline().setStages([
9289
... imageAssembler,
9390
... visualQAClassifier
9491
... ])
95-
9692
>>> result = pipeline.fit(test_df).transform(test_df)
9793
>>> result.select("image_assembler.origin", "answer.result").show(truncate=False)
98-
9994
+--------------------------------------+----------------------------------------------------------------------+
10095
|origin |result |
10196
+--------------------------------------+----------------------------------------------------------------------+

‎python/sparknlp/annotator/cv/llava_for_multimodal.py‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -65,7 +65,7 @@ class LLAVAForMultiModal(AnnotatorModel,
6565
>>> from sparknlp.annotator import *
6666
>>> from pyspark.ml import Pipeline
6767
>>> image_df = SparkSessionForTest.spark.read.format("image").load(path=images_path)
68-
>>> test_df = image_df.withColumn("text", lit("USER: \n <|image|> \n What's this picture about? \n ASSISTANT:\n"))
68+
>>> test_df = image_df.withColumn("text", lit("USER: \\n <|image|> \\n What's this picture about? \\n ASSISTANT:\\n"))
6969
>>> imageAssembler = ImageAssembler() \\
7070
... .setInputCol("image") \\
7171
... .setOutputCol("image_assembler")

‎python/sparknlp/annotator/cv/paligemma_for_multimodal.py‎

Lines changed: 7 additions & 7 deletions
Original file line numberDiff line numberDiff line change
@@ -28,8 +28,8 @@ class PaliGemmaForMultiModal(AnnotatorModel,
2828
Pretrained models can be loaded with :meth:`.pretrained` of the companion
2929
object:
3030
31-
>>> visualQAClassifier = PaliGemmaForMultiModal.pretrained() \
32-
... .setInputCols(["image_assembler"]) \
31+
>>> visualQAClassifier = PaliGemmaForMultiModal.pretrained() \\
32+
... .setInputCols(["image_assembler"]) \\
3333
... .setOutputCol("answer")
3434
3535
The default model is ``"paligemma_3b_pt_224_int4"``, if no name is
@@ -59,12 +59,12 @@ class PaliGemmaForMultiModal(AnnotatorModel,
5959
>>> from sparknlp.annotator import *
6060
>>> from pyspark.ml import Pipeline
6161
>>> image_df = SparkSessionForTest.spark.read.format("image").load(path=images_path)
62-
>>> test_df = image_df.withColumn("text", lit("USER: \n <image> \nDescribe this image. \nASSISTANT:\n"))
63-
>>> imageAssembler = ImageAssembler() \
64-
... .setInputCol("image") \
62+
>>> test_df = image_df.withColumn("text", lit("USER: \\n <image> \\nDescribe this image. \\nASSISTANT:\\n"))
63+
>>> imageAssembler = ImageAssembler() \\
64+
... .setInputCol("image") \\
6565
... .setOutputCol("image_assembler")
66-
>>> visualQAClassifier = PaliGemmaForMultiModal.pretrained() \
67-
... .setInputCols("image_assembler") \
66+
>>> visualQAClassifier = PaliGemmaForMultiModal.pretrained() \\
67+
... .setInputCols("image_assembler") \\
6868
... .setOutputCol("answer")
6969
>>> pipeline = Pipeline().setStages([
7070
... imageAssembler,

‎python/sparknlp/annotator/cv/phi3_vision_for_multimodal.py‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -65,7 +65,7 @@ class Phi3Vision(AnnotatorModel,
6565
>>> from sparknlp.annotator import *
6666
>>> from pyspark.ml import Pipeline
6767
>>> image_df = SparkSessionForTest.spark.read.format("image").load(path=images_path)
68-
>>> test_df = image_df.withColumn("text", lit("<|user|> \n <|image_1|> \nWhat is unusual on this picture? <|end|>\n <|assistant|>\n"))
68+
>>> test_df = image_df.withColumn("text", lit("<|user|> \\n <|image_1|> \\nWhat is unusual on this picture? <|end|>\\n <|assistant|>\\n"))
6969
>>> imageAssembler = ImageAssembler() \\
7070
... .setInputCol("image") \\
7171
... .setOutputCol("image_assembler")

‎python/sparknlp/annotator/cv/qwen2vl_transformer.py‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -68,7 +68,7 @@ class Qwen2VLTransformer(AnnotatorModel,
6868
>>> from sparknlp.annotator import *
6969
>>> from pyspark.ml import Pipeline
7070
>>> image_df = SparkSessionForTest.spark.read.format("image").load(path=images_path)
71-
>>> test_df = image_df.withColumn("text", lit("<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\n<|vision_start|><|image_pad|><|vision_end|>Describe this image.<|im_end|>\n<|im_start|>assistant\n"))
71+
>>> test_df = image_df.withColumn("text", lit("<|im_start|>system\\nYou are a helpful assistant.<|im_end|>\\n<|im_start|>user\\n<|vision_start|><|image_pad|><|vision_end|>Describe this image.<|im_end|>\\n<|im_start|>assistant\\n"))
7272
>>> imageAssembler = ImageAssembler() \\
7373
... .setInputCol("image") \\
7474
... .setOutputCol("image_assembler")

0 commit comments

Comments
 (0)