ValueError: Error raised by bedrock service: 'str' object has no attribute 'invoke_model'

[anth0nyhak1m]

Here is the example code I am using, with some minor adjustments. Bedrock will not instantiate with a temperature parameter, so I deleted that, and I went into the code for abstract_graph.py and deleted any default instantiation of a temperature parameter.

"""
Basic example of scraping pipeline using SmartScraper
"""

import os
from dotenv import load_dotenv
from scrapegraphai.graphs import SmartScraperGraph
from scrapegraphai.utils import prettify_exec_info

load_dotenv()

************************************************

Define the configuration for the graph

************************************************

graph_config = {
"llm": {
"client": "bedrock",
"model": "bedrock/anthropic.claude-3-sonnet-20240229-v1:0"
}
}

************************************************

Create the SmartScraperGraph instance and run it

************************************************

smart_scraper_graph = SmartScraperGraph(
prompt="List me all the projects with their description",
# also accepts a string with the already downloaded HTML code
source="https://perinim.github.io/projects/",
config=graph_config
)

result = smart_scraper_graph.run()
print(result)

************************************************

Get graph execution info

************************************************

graph_exec_info = smart_scraper_graph.get_execution_info()
print(prettify_exec_info(graph_exec_info))

Here is the error that I get that I cannot solve, I think it's an issue deep within the langchain library, however I am not sure if there are any workarounds:

Traceback (most recent call last):
File "C:\Users\AnthonyHakim\anaconda3\envs\Lib\site-packages\langchain_aws\llms\bedrock.py", line 715, in _prepare_input_and_invoke
response = self.client.invoke_model(**request_options)
^^^^^^^^^^^^^^^^^^^^^^^^
AttributeError: 'str' object has no attribute 'invoke_model'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "C:\Users\AnthonyHakim\\test.py", line 72, in
result = smart_scraper_graph.run()
^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\AnthonyHakim\anaconda3\envs\Lib\site-packages\scrapegraphai\graphs\smart_scraper_graph.py", line 114, in run
self.final_state, self.execution_info = self.graph.execute(inputs)
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\AnthonyHakim\anaconda3\envs\Lib\site-packages\scrapegraphai\graphs\base_graph.py", line 263, in execute
return self._execute_standard(initial_state)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\AnthonyHakim\anaconda3\envs\Lib\site-packages\scrapegraphai\graphs\base_graph.py", line 184, in _execute_standard
raise e
File "C:\Users\AnthonyHakim\anaconda3\envs\Lib\site-packages\scrapegraphai\graphs\base_graph.py", line 168, in _execute_standard
result = current_node.execute(state)
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\AnthonyHakim\anaconda3\envs\Lib\site-packages\scrapegraphai\nodes\generate_answer_node.py", line 134, in execute
answer = chain.invoke({"question": user_prompt})
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\AnthonyHakim\anaconda3\envs\Lib\site-packages\langchain_core\runnables\base.py", line 2878, in invoke
input = context.run(step.invoke, input, config)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\AnthonyHakim\anaconda3\envs\Lib\site-packages\langchain_core\language_models\chat_models.py", line 276, in invoke
self.generate_prompt(
File "C:\Users\AnthonyHakim\anaconda3\envs\Lib\site-packages\langchain_core\language_models\chat_models.py", line 776, in generate_prompt
return self.generate(prompt_messages, stop=stop, callbacks=callbacks, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\AnthonyHakim\anaconda3\envs\Lib\site-packages\langchain_core\language_models\chat_models.py", line 633, in generate
raise e
File "C:\Users\AnthonyHakim\anaconda3\envs\Lib\site-packages\langchain_core\language_models\chat_models.py", line 623, in generate
self._generate_with_cache(
File "C:\Users\AnthonyHakim\anaconda3\envs\Lib\site-packages\langchain_core\language_models\chat_models.py", line 845, in _generate_with_cache
result = self._generate(
^^^^^^^^^^^^^^^
File "C:\Users\AnthonyHakim\anaconda3\envs\Lib\site-packages\langchain_aws\chat_models\bedrock.py", line 552, in _generate
completion, tool_calls, llm_output = self._prepare_input_and_invoke(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\AnthonyHakim\anaconda3\envs\Lib\site-packages\langchain_aws\llms\bedrock.py", line 726, in _prepare_input_and_invoke
raise ValueError(f"Error raised by bedrock service: {e}")
ValueError: Error raised by bedrock service: 'str' object has no attribute 'invoke_model'

(I removed project names from paths)

[VinciGit00]

Does the smart scraper works?

[f-aguzzi]

@anth0nyhak1m Thanks for reporting this.

The bug should be fixed by pull request #636, along with issue #634.

[anth0nyhak1m]

So the way that I got this to work around the error, and this should be mentioned in the comments or docs, is that in graph_config, "client" should map to an instance of a boto3 'bedrock-runtime' client.

If you try to instantiate a 'bedrock' client you will get an error that says bedrock object has no attribute called invoke model. If you pass in a string, it will say string object has no attribute invoke model.

Must be a 'bedrock-runtime' client.

I found this fix directly in the issues of langchain:
langchain-ai/langchain#2828 (comment)

[f-aguzzi]

Thanks for your work and your research.

Meanwhile I found out that there is a different way to pass the temperature parameter to Bedrock models, so I added a workaround in the beta. Let's wait for the LangChain to fix their thing, and then the Bedrock support in ScrapeGraph will be operational again.

[VinciGit00]

Hi, can you add in the examples the bedrock runtime? Please

[VinciGit00]

please update to the new version