【问题标题】:Unable to run command rasa train无法运行命令 rasa train
【发布时间】:2021-05-12 14:49:12
【问题描述】:

我正在尝试安装我的第一个 rasa 项目,但是当我运行 rasa train 时,我收到了这些错误消息。有人能帮我吗? (对不起,如果太乱了,这是我第一次在 StackOverflow 中)

(venv) ➜ rasa-init-demo rasa train 策略和管道的配置是自动选择的。它被写入 'config.yml' 的配置文件中。 训练 NLU 模型... 2021-02-08 18:38:15 信息 rasa.shared.nlu.training_data.training_data - 训练数据统计: 2021-02-08 18:38:15 INFO rasa.shared.nlu.training_data.training_data - 意图示例数:69(7 个不同的意图)

2021-02-08 18:38:15 信息 rasa.shared.nlu.training_data.training_data - 发现意图:“拒绝”、“再见”、“mood_unhappy”、“肯定”、“问候”、“mood_great” , '机器人挑战' 2021-02-08 18:38:15 信息 rasa.shared.nlu.training_data.training_data - 响应示例数:0(0 个不同响应) 2021-02-08 18:38:15 信息 rasa.shared.nlu.training_data.training_data - 实体示例数:0(0 个不同的实体) 2021-02-08 18:38:15 INFO rasa.nlu.model - 开始训练组件 WhitespaceTokenizer 2021-02-08 18:38:15 INFO rasa.nlu.model - 完成训练组件。 2021-02-08 18:38:15 INFO rasa.nlu.model - 开始训练组件 RegexFeaturizer 2021-02-08 18:38:15 INFO rasa.nlu.model - 完成训练组件。 2021-02-08 18:38:15 INFO rasa.nlu.model - 开始训练组件 LexicalSyntacticFeaturizer 2021-02-08 18:38:15 INFO rasa.nlu.model - 完成训练组件。 2021-02-08 18:38:15 INFO rasa.nlu.model - 开始训练组件 CountVectorsFeaturizer 2021-02-08 18:38:15 信息 rasa.nlu.featurizers.sparse_featurizer.count_vectors_featurizer - 为文本属性配置的 1080 个插槽中消耗了 80 个词汇插槽。 2021-02-08 18:38:15 INFO rasa.nlu.model - 完成训练组件。 2021-02-08 18:38:15 INFO rasa.nlu.model - 开始训练组件 CountVectorsFeaturizer 2021-02-08 18:38:15 信息 rasa.nlu.featurizers.sparse_featurizer.count_vectors_featurizer - 为文本属性配置的 1697 个插槽中消耗了 697 个词汇插槽。 2021-02-08 18:38:15 INFO rasa.nlu.model - 完成训练组件。 2021-02-08 18:38:15 INFO rasa.nlu.model - 开始训练组件 DIETClassifier 回溯(最近一次通话最后): 文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/tensorflow/python/autograph/impl/api.py”,第584行,在converted_call中 convert_f = conversion.convert(target_entity, program_ctx) 文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/tensorflow/python/autograph/impl/conversion.py”,第 118 行,转换 转换,模块,source_map = _TRANSPILER.transform_function( 文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/tensorflow/python/autograph/pyct/transpiler.py”,第 411 行,在 transform_function 工厂 = self._transformed_factory(fn, caching_subkey, user_context, _transformed_factory 中的文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/tensorflow/python/autograph/pyct/transpiler.py”,第 373 行 节点,ctx = self._transform_function(fn, user_context) _transform_function 中的文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/tensorflow/python/autograph/pyct/transpiler.py”,第 339 行 节点 = self.transform_ast(节点,上下文) 文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/tensorflow/python/autograph/impl/conversion.py”,第 69 行,在 transform_ast 节点 = qual_names.resolve(节点) 文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/tensorflow/python/autograph/pyct/qual_names.py”,第252行,解决 返回 QnResolver().visit(node) 文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/ast.py”,第371行,访问 回访者(节点) 文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/ast.py”,第 456 行,位于 generic_visit new_node = self.visit(old_value) 文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/ast.py”,第371行,访问 回访者(节点) 文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/ast.py”,第 447 行,位于 generic_visit 价值 = self.visit(价值) 文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/ast.py”,第371行,访问 回访者(节点) 文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/tensorflow/python/autograph/pyct/qual_names.py”,第 217 行,在 visit_Name 节点 = self.generic_visit(节点) 文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/ast.py”,第 456 行,位于 generic_visit new_node = self.visit(old_value) 文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/ast.py”,第371行,访问 回访者(节点) 文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/tensorflow/python/autograph/pyct/qual_names.py”,第 230 行,在 visit_Subscript 节点 = self.generic_visit(节点) 文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/ast.py”,第 456 行,位于 generic_visit new_node = self.visit(old_value) 文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/ast.py”,第371行,访问 回访者(节点) 文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/ast.py”,第 447 行,位于 generic_visit 价值 = self.visit(价值) 文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/ast.py”,第371行,访问 回访者(节点) 文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/tensorflow/python/autograph/pyct/qual_names.py”,第232行,在visit_Subscript 如果不是 isinstance(s, gast.Index): AttributeError:模块“gast”没有属性“索引”

在处理上述异常的过程中,又发生了一个异常:

Traceback(最近一次调用最后一次): 文件“/Library/Frameworks/Python.framework/Versions/3.8/bin/rasa”,第 8 行,在 sys.exit(main()) 文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/rasa/ma​​in.py”,第 116 行,在 main cmdline_arguments.func(cmdline_arguments) 文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/rasa/cli/train.py”,第 58 行,在 train_parser.set_defaults(func=lambda args: train(args, can_exit=True)) 文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/rasa/cli/train.py”,第 90 行,在火车中 training_result = rasa.train( 文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/rasa/train.py”,第 94 行,在火车中 返回 rasa.utils.common.run_in_loop( 文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/rasa/utils/common.py”,第 308 行,在 run_in_loop 结果 = loop.run_until_complete(f) 文件“uvloop/loop.pyx”,第 1456 行,在 uvloop.loop.Loop.run_until_complete train_async 中的文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/rasa/train.py”,第 163 行 返回等待_train_async_internal( _train_async_internal 中的文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/rasa/train.py”,第 342 行 等待_do_training( _do_training 中的文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/rasa/train.py”,第 388 行 模型路径=等待_train_nlu_with_validated_data( _train_nlu_with_validated_data 中的文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/rasa/train.py”,第 811 行 等待 rasa.nlu.train( 文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/rasa/nlu/train.py”,第 116 行,在火车中 解释器 = trainer.train(training_data, **kwargs) 文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/rasa/nlu/model.py”,第 209 行,在火车中 更新 = component.train(working_data, self.config, **context) 文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/rasa/nlu/classifiers/diet_classifier.py”,第 816 行,在火车中 self.model.fit( 文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/rasa/utils/tensorflow/models.py”,第 220 行,适合 ) = self._get_tf_train_functions(eager, model_data, batch_strategy) _get_tf_train_functions 中的文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/rasa/utils/tensorflow/models.py”,第 481 行 self._get_tf_call_model_function( _get_tf_call_model_function 中的文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/rasa/utils/tensorflow/models.py”,第 464 行 tf_call_model_function(next(iter(init_dataset))) 调用中的文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/tensorflow/python/eager/def_function.py”,第 780 行 结果 = self._call(*args, **kwds) _call 中的文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/tensorflow/python/eager/def_function.py”,第 823 行 self._initialize(args, kwds, add_initializers_to=initializers) _initialize 中的文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/tensorflow/python/eager/def_function.py”,第 696 行 self._stateful_fn._get_concrete_function_internal_garbage_collected(#pylint: disable=protected-access _get_concrete_function_internal_garbage_collected 中的文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/tensorflow/python/eager/function.py”,第 2855 行 图函数,_,_ = self._maybe_define_function(args,kwargs) _maybe_define_function 中的文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/tensorflow/python/eager/function.py”,第 3213 行 graph_function = self._create_graph_function(args, kwargs) _create_graph_function 中的文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/tensorflow/python/eager/function.py”,第 3065 行 func_graph_module.func_graph_from_py_func( func_graph_from_py_func 中的文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/tensorflow/python/framework/func_graph.py”,第 986 行 func_outputs = python_func(*func_args, **func_kwargs) 文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/tensorflow/python/eager/def_function.py”,第 600 行,位于 Wrapped_fn return weak_wrapped_fn().wrapped(*args, **kwds) 包装器中的文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/tensorflow/python/framework/func_graph.py”,第 962 行 返回 autograph.converted_call( 文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/tensorflow/python/autograph/impl/api.py”,第591行,在converted_call中 return _fall_back_unconverted(f, args, kwargs, options, e) _fall_back_unconverted 中的文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/tensorflow/python/autograph/impl/api.py”,第 398 行 return _call_unconverted(f, args, kwargs, 选项) _call_unconverted 中的文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/tensorflow/python/autograph/impl/api.py”,第 339 行 返回 f(*args, **kwargs) 文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/rasa/utils/tensorflow/models.py”,第 293 行,在 train_on_batch prediction_loss = self.batch_loss(batch_in) 文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/rasa/nlu/classifiers/diet_classifier.py”,第 1417 行,在 batch_loss sequence_lengths = self._get_sequence_lengths( _get_sequence_lengths 中的文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/rasa/utils/tensorflow/models.py”,第 1112 行 sequence_lengths = tf.ones([batch_dim], dtype=tf.int32) 包装器中的文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/tensorflow/python/util/dispatch.py​​”,第 201 行 返回目标(*args,**kwargs) 文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/tensorflow/python/ops/array_ops.py”,第 3041 行,在一个 输出 = _constant_if_small(一,形状,数据类型,名称) _constant_if_small 中的文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/tensorflow/python/ops/array_ops.py”,第 2732 行 如果 np.prod(shape) array_function internals>”,第 5 行,在 prod 文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/numpy/core/fromnumeric.py”,第 3030 行,在产品中 return _wrapreduction(a, np.multiply, 'prod', 轴, dtype, out, _wrapreduction 中的文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/numpy/core/fromnumeric.py”,第 87 行 return ufunc.reduce(obj, axis, dtype, out, **passkwargs) array 中的文件“/Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages/tensorflow/python/framework/ops.py”,第 845 行 引发 NotImplementedError( NotImplementedError:无法将符号张量 (strided_slice_6:0) 转换为 numpy 数组。此错误可能表明您正在尝试将张量传递给 NumPy 调用,这是不受支持的

Rasa 版本:2.2.9 Rasa SDK 版本:2.2.0 Rasa X 版本:无 Python版本:3.8.7 操作系统:macOS-10.15.7-x86_64-i386-64bit Python 路径:/usr/local/bin/python3.8

【问题讨论】:

    标签: python-3.x macos rasa


    【解决方案1】:

    这应该在今天发布的 2.2.10 中修复:https://github.com/RasaHQ/rasa/releases/tag/2.2.10

    【讨论】:

      猜你喜欢
      • 1970-01-01
      • 1970-01-01
      • 1970-01-01
      • 1970-01-01
      • 1970-01-01
      • 1970-01-01
      • 2020-11-30
      • 1970-01-01
      • 2020-12-20
      相关资源
      最近更新 更多