Fix left bugs and release v1.4.5 (#883)

HYLcool · web-flow · commit 923faf3daac0 · 2026-01-13T14:35:30.000+08:00
* * update version

* * use forked version of transformers-stream-generator when checking it
diff --git a/data_juicer/__init__.py b/data_juicer/__init__.py
@@ -1,4 +1,4 @@
-__version__ = "1.4.4"
+__version__ = "1.4.5"
 
 import sys
 
diff --git a/data_juicer/ops/mapper/video_captioning_from_audio_mapper.py b/data_juicer/ops/mapper/video_captioning_from_audio_mapper.py
@@ -37,7 +37,7 @@ def __init__(self, keep_original_sample: bool = True, *args, **kwargs):
         LazyLoader.check_packages(
             [
                 "transformers",
-                "transformers-stream-generator",
+                "transformers-stream-generator @ git+https://github.com/datajuicer/transformers-stream-generator.git",
                 "einops",
                 "accelerate",
                 "tiktoken",
diff --git a/data_juicer/ops/mapper/video_captioning_from_summarizer_mapper.py b/data_juicer/ops/mapper/video_captioning_from_summarizer_mapper.py
@@ -90,7 +90,7 @@ def __init__(
             [
                 "torch",
                 "transformers",
-                "transformers_stream_generator",
+                "transformers-stream-generator @ git+https://github.com/datajuicer/transformers-stream-generator.git",
                 "einops",
                 "accelerate",
                 "tiktoken",  # by audio caption
diff --git a/tests/ops/mapper/test_text_tagging_by_prompt_mapper.py b/tests/ops/mapper/test_text_tagging_by_prompt_mapper.py
@@ -14,7 +14,7 @@ def check_string_in_list(string_list, output):
     assert False, f"没有字符串在输出中"
 
 
-class TextTaggingByPromptTest(DataJuicerTestCaseBase):
+class TextTaggingByPromptMapperTest(DataJuicerTestCaseBase):
     text_key = 'text'
 
     def _run_tagging(self, samples, enable_vllm=False, sampling_params={}, **kwargs):

Original file line number	Diff line number	Diff line change
`@@ -1,4 +1,4 @@`
`1`		`-__version__ = "1.4.4"`
	`1`	`+__version__ = "1.4.5"`
`2`	`2`
`3`	`3`	`import sys`
`4`	`4`
Original file line number	Diff line number	Diff line change
`@@ -37,7 +37,7 @@ def __init__(self, keep_original_sample: bool = True, args, *kwargs):`
`37`	`37`	`LazyLoader.check_packages(`
`38`	`38`	`[`
`39`	`39`	`"transformers",`
`40`		`- "transformers-stream-generator",`
	`40`	`+ "transformers-stream-generator @ git+https://github.com/datajuicer/transformers-stream-generator.git",`
`41`	`41`	`"einops",`
`42`	`42`	`"accelerate",`
`43`	`43`	`"tiktoken",`
Original file line number	Diff line number	Diff line change
`@@ -90,7 +90,7 @@ def __init__(`
`90`	`90`	`[`
`91`	`91`	`"torch",`
`92`	`92`	`"transformers",`
`93`		`- "transformers_stream_generator",`
	`93`	`+ "transformers-stream-generator @ git+https://github.com/datajuicer/transformers-stream-generator.git",`
`94`	`94`	`"einops",`
`95`	`95`	`"accelerate",`
`96`	`96`	`"tiktoken", # by audio caption`