Fix typos (#1328)

README.md

@@ -126,7 +126,7 @@ AI for Science:
 
 * [uni-fold-multimer](https://modelscope.cn/models/DPTech/uni-fold-multimer/summary)
 
-**Note:** Most models on ModelScope are public and can be downloaded without account registration on modelscope website([www.modelscope.cn](www.modelscope.cn)), please refer to instructions for [model download](https://modelscope.cn/docs/%E6%A8%A1%E5%9E%8B%E7%9A%84%E4%B8%8B%E8%BD%BD), for dowloading models with api provided by modelscope library or git.
+**Note:** Most models on ModelScope are public and can be downloaded without account registration on modelscope website([www.modelscope.cn](www.modelscope.cn)), please refer to instructions for [model download](https://modelscope.cn/docs/%E6%A8%A1%E5%9E%8B%E7%9A%84%E4%B8%8B%E8%BD%BD), for downloading models with api provided by modelscope library or git.
 
 # QuickTour
 
@@ -158,7 +158,7 @@ The output image with the background removed is:
 ![image](data/resource/portrait_output.png)
 
 
-Fine-tuning and evaluation can also be done with a few more lines of code to set up training dataset and trainer, with the heavy-lifting work of training and evaluation a model encapsulated in the implementation of  `traner.train()` and
+Fine-tuning and evaluation can also be done with a few more lines of code to set up training dataset and trainer, with the heavy-lifting work of training and evaluation a model encapsulated in the implementation of  `trainer.train()` and
 `trainer.evaluate()`  interfaces.
 
 For example, the gpt3 base model (1.3B) can be fine-tuned with the chinese-poetry dataset, resulting in a model that can be used for chinese-poetry generation.

README_zh.md

@@ -227,7 +227,7 @@ conda activate modelscope
 
 安装完前置依赖，你可以按照如下方式安装 ModelScope Library。
 
-ModelScope Libarary 由核心框架，以及不同领域模型的对接组件组成。如果只需要 ModelScope 模型和数据集访问等基础能力，可以只安装 ModelScope 的核心框架：
+ModelScope Library 由核心框架，以及不同领域模型的对接组件组成。如果只需要 ModelScope 模型和数据集访问等基础能力，可以只安装 ModelScope 的核心框架：
 
 ```shell
 pip install modelscope

examples/pytorch/llm/utils/utils.py

@@ -143,8 +143,8 @@ def print_example(example: Dict[str, Any], tokenizer) -> None:
     print(f'[INPUT] {tokenizer.decode(input_ids)}')
     print()
     n_mask = Counter(labels)[-100]
-    print(f'[LABLES_IDS] {labels}')
+    print(f'[LABELS_IDS] {labels}')
-    print(f'[LABLES] <-100 * {n_mask}>{tokenizer.decode(labels[n_mask:])}')
+    print(f'[LABELS] <-100 * {n_mask}>{tokenizer.decode(labels[n_mask:])}')
 
 
 def data_collate_fn(batch: List[Dict[str, Any]], tokenizer) -> Dict[str, Any]:

examples/pytorch/llm_agent/_common.py

@@ -202,7 +202,7 @@ def print_examples(examples: Dict[str, Any], tokenizer) -> None:
     print(f'[INPUT_IDS] {tokenizer.decode(input_ids)}')
     print()
     print(
-        f'[LABLES] {tokenizer.decode([lb if lb != -100 else 0 for lb in labels])}'
+        f'[LABELS] {tokenizer.decode([lb if lb != -100 else 0 for lb in labels])}'
     )
 
 

modelscope/metainfo.py

@@ -297,12 +297,12 @@ class Heads(object):
 class Pipelines(object):
     """ Names for different pipelines.
 
-        Holds the standard pipline name to use for identifying different pipeline.
+        Holds the standard pipeline name to use for identifying different pipeline.
     This should be used to register pipelines.
 
         For pipeline which support different models and implements the common function, we
     should use task name for this pipeline.
-        For pipeline which suuport only one model, we should use ${Model}-${Task} as its name.
+        For pipeline which support only one model, we should use ${Model}-${Task} as its name.
     """
     pipeline_template = 'pipeline-template'
     # vision tasks
@@ -1105,7 +1105,7 @@ class Preprocessors(object):
 
         For a general preprocessor, just use the function name as preprocessor name such as
     resize-image, random-crop
-        For a model-specific preprocessor, use ${modelname}-${fuction}
+        For a model-specific preprocessor, use ${modelname}-${function}
     """
 
     # cv preprocessor

modelscope/metrics/ciderD/ciderD_scorer.py

@@ -20,7 +20,7 @@ def precook(s, n=4, out=False):
     can take string arguments as well.
     :param s: string : sentence to be converted into ngrams
     :param n: int    : number of ngrams for which representation is calculated
-    :return: term frequency vector for occuring ngrams
+    :return: term frequency vector for occurring ngrams
     """
     words = s.split()
     counts = defaultdict(int)

modelscope/metrics/image_colorization_metric.py

@@ -98,7 +98,7 @@ class INCEPTION_V3_FID(nn.Module):
     # Maps feature dimensionality to their output blocks indices
     BLOCK_INDEX_BY_DIM = {
         64: 0,  # First max pooling features
-        192: 1,  # Second max pooling featurs
+        192: 1,  # Second max pooling features
         768: 2,  # Pre-aux classifier features
         2048: 3  # Final average pooling features
     }
@@ -295,7 +295,7 @@ def calculate_frechet_distance(mu1, sigma1, mu2, sigma2, eps=1e-6):
             nception net (like returned by the function 'get_predictions')
             or generated samples.
         mu2: The sample mean over activations, precalculated on an
-            representive data set.
+            representative data set.
         sigma1: The covariance matrix over activations for generated samples.
         sigma2: The covariance matrix over activations, precalculated on an
             epresentive data set.

modelscope/pipelines/audio/inverse_text_processing_pipeline.py

@@ -37,7 +37,7 @@ class InverseTextProcessingPipeline(Pipeline):
     >>> sentence = 'sembilan ribu sembilan ratus sembilan puluh sembilan'
     >>> print(pipeline_itn(sentence))
 
-    To view other examples plese check tests/pipelines/test_inverse_text_processing.py.
+    To view other examples please check tests/pipelines/test_inverse_text_processing.py.
     """
 
     def __init__(self, model: Union[Model, str] = None, **kwargs):

modelscope/pipelines/multi_modal/disco_guided_diffusion_pipeline/utils.py

@@ -84,8 +84,8 @@ def resize(input,
     eps = fw.finfo(fw.float32).eps
     device = input.device if fw is torch else None
 
-    # set missing scale factors or output shapem one according to another,
+    # set missing scale factors or output shape one according to another,
-    # scream if both missing. this is also where all the defults policies
+    # scream if both missing. this is also where all the defaults policies
     # take place. also handling the by_convs attribute carefully.
     scale_factors, out_shape, by_convs = set_scale_and_out_sz(
         in_shape, out_shape, scale_factors, by_convs, scale_tolerance,
@@ -155,15 +155,15 @@ def resize(input,
 
 
 def get_projected_grid(in_sz, out_sz, scale_factor, fw, by_convs, device=None):
-    # we start by having the ouput coordinates which are just integer locations
+    # we start by having the output coordinates which are just integer locations
-    # in the special case when usin by_convs, we only need two cycles of grid
+    # in the special case when using by_convs, we only need two cycles of grid
     # points. the first and last.
     grid_sz = out_sz if not by_convs else scale_factor.numerator
     out_coordinates = fw_arange(grid_sz, fw, device)
 
-    # This is projecting the ouput pixel locations in 1d to the input tensor,
+    # This is projecting the output pixel locations in 1d to the input tensor,
     # as non-integer locations.
-    # the following fomrula is derived in the paper
+    # the following formula is derived in the paper
     # "From Discrete to Continuous Convolutions" by Shocher et al.
     v1 = out_coordinates / float(scale_factor) + (in_sz - 1) / 2
     v2 = (out_sz - 1) / (2 * float(scale_factor))

modelscope/pipelines/multi_modal/gridvlp_pipeline.py

@@ -155,7 +155,7 @@ class GridVlpPipeline(Pipeline):
     Tasks.visual_question_answering,
     module_name=Pipelines.gridvlp_multi_modal_classification)
 class GridVlpClassificationPipeline(GridVlpPipeline):
-    """ Pipeline for gridvlp classification, including cate classfication and
+    """ Pipeline for gridvlp classification, including cate classification and
     brand classification.
 
     Example:
@@ -174,7 +174,7 @@ class GridVlpClassificationPipeline(GridVlpPipeline):
     """
 
     def __init__(self, model_name_or_path: str, **kwargs):
-        """ Pipeline for gridvlp classification, including cate classfication and
+        """ Pipeline for gridvlp classification, including cate classification and
     brand classification.
         Args:
             model: path to local model directory.

modelscope/pipelines/nlp/canmt_translation_pipeline.py

@@ -49,7 +49,7 @@ class CanmtTranslationPipeline(Pipeline):
             >>> # Or use the list input:
             >>> print(pipeline_ins([sentence1])
 
-            To view other examples plese check tests/pipelines/test_canmt_translation.py.
+            To view other examples please check tests/pipelines/test_canmt_translation.py.
         """
         super().__init__(
             model=model,

modelscope/pipelines/nlp/document_grounded_dialog_rerank_pipeline.py

@@ -415,7 +415,7 @@ def get_rank(guess_item, gold_item, k, rank_keys, verbose=False):
                 f'for a robust recall@{k} computation (you provided {len(guess_ids)} item(s)).'
             )
 
-        # 3. rank by gruping pages in each evidence set (each evidence set count as 1),
+        # 3. rank by grouping pages in each evidence set (each evidence set count as 1),
         # the position in the rank of each evidence set is given by the last page in guess_ids
         # non evidence pages counts as 1
         rank = []

modelscope/pipelines/nlp/fill_mask_pipeline.py

@@ -55,7 +55,7 @@ class FillMaskPipeline(Pipeline):
         NOTE2: Please pay attention to the model's special tokens.
         If bert based model(bert, structbert, etc.) is used, the mask token is '[MASK]'.
         If the xlm-roberta(xlm-roberta, veco, etc.) based model is used, the mask token is '<mask>'.
-        To view other examples plese check tests/pipelines/test_fill_mask.py.
+        To view other examples please check tests/pipelines/test_fill_mask.py.
         """
         super().__init__(
             model=model,

modelscope/pipelines/nlp/named_entity_recognition_pipeline.py

@@ -48,7 +48,7 @@ class NamedEntityRecognitionPipeline(TokenClassificationPipeline):
             >>> input = '这与温岭市新河镇的一个神秘的传说有关。'
             >>> print(pipeline_ins(input))
 
-            To view other examples plese check the tests/pipelines/test_plugin_model.py.
+            To view other examples please check the tests/pipelines/test_plugin_model.py.
         """
         super().__init__(
             model=model,

modelscope/pipelines/nlp/siamese_uie_pipeline.py

@@ -60,7 +60,7 @@ class SiameseUiePipeline(Pipeline):
             >>> sentence = '1944年毕业于北大的名古屋铁道会长谷口清太郎等人在日本积极筹资，共筹款2.7亿日元，参加捐款的日本企业有69家。'
             >>> print(pipeline_ins(sentence, schema={'人物': None, '地理位置': None, '组织机构': None}))
 
-            To view other examples plese check tests/pipelines/test_siamese_uie.py.
+            To view other examples please check tests/pipelines/test_siamese_uie.py.
         """
         super().__init__(
             model=model,

modelscope/pipelines/nlp/text_error_correction_pipeline.py

@@ -43,7 +43,7 @@ class TextErrorCorrectionPipeline(Pipeline):
             >>> sentence1 = '随着中国经济突飞猛近，建造工业与日俱增'
             >>> print(pipeline_ins(sentence1))
 
-        To view other examples plese check tests/pipelines/test_text_error_correction.py.
+        To view other examples please check tests/pipelines/test_text_error_correction.py.
         """
         super().__init__(
             model=model,

modelscope/pipelines/nlp/text_generation_pipeline.py

@@ -64,7 +64,7 @@ class TextGenerationPipeline(Pipeline, PipelineStreamingOutputMixin):
             >>> # Or use the dict input:
             >>> print(pipeline_ins({'sentence': sentence1}))
 
-            To view other examples plese check tests/pipelines/test_text_generation.py.
+            To view other examples please check tests/pipelines/test_text_generation.py.
         """
         super().__init__(
             model=model,
@@ -517,7 +517,7 @@ class Llama2TaskPipeline(TextGenerationPipeline):
             >>>     temperature=1.0, repetition_penalty=1., eos_token_id=2, bos_token_id=1, pad_token_id=0)
             >>> print(result['text'])
 
-            To view other examples plese check tests/pipelines/test_llama2_text_generation_pipeline.py.
+            To view other examples please check tests/pipelines/test_llama2_text_generation_pipeline.py.
         """
         self.model = Model.from_pretrained(
             model, device_map='auto', torch_dtype=torch.float16)
@@ -604,7 +604,7 @@ class Llama2chatTaskPipeline(Pipeline):
             >>> pad_token_id=0, history=history_demo)
             >>> print(result['response'])
 
-            To view other examples plese check tests/pipelines/test_llama2_text_generation_pipeline.py.
+            To view other examples please check tests/pipelines/test_llama2_text_generation_pipeline.py.
         """
 
     def __init__(self,

modelscope/pipelines/nlp/word_segmentation_pipeline.py

@@ -35,7 +35,7 @@ class WordSegmentationPipeline(TokenClassificationPipeline):
         >>> sentence1 = '今天天气不错，适合出去游玩'
         >>> print(pipeline_ins(sentence1))
 
-    To view other examples plese check tests/pipelines/test_word_segmentation.py.
+    To view other examples please check tests/pipelines/test_word_segmentation.py.
     """
 
     def postprocess(self,

modelscope/pipelines/nlp/zero_shot_classification_pipeline.py

@@ -59,7 +59,7 @@ class ZeroShotClassificationPipeline(Pipeline):
             >>> template = '这篇文章的标题是{}'
             >>> print(pipeline_ins(sentence1, candidate_labels=labels, hypothesis_template=template))
 
-            To view other examples plese check tests/pipelines/test_zero_shot_classification.py.
+            To view other examples please check tests/pipelines/test_zero_shot_classification.py.
         """
         super().__init__(
             model=model,