ObservationNameError: name 'Observation' is not defined

[vishals25]

Checked other resources

• I added a very descriptive title to this question.
• I searched the LangChain documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.

Commit to Help

• I commit to help with one of those options 👆

Example Code

# Parse the CSV file and create a Pandas DataFrame from its contents.
    df = pd.read_csv(data)

    llm = HuggingFaceEndpoint(
            repo_id="mistralai/Mistral-7B-Instruct-v0.3"
        )
    
    # Create a Pandas DataFrame agent.
    agent = create_pandas_dataframe_agent(llm, df, verbose=True,allow_dangerous_code=True)

    #Python REPL: A Python shell used to evaluating and executing Python commands. 
    #It takes python code as input and outputs the result. The input python code can be generated from another tool in the LangChain
    return agent.invoke(query)

Description

I am trying to retrieve details from the csv file using chains

System Info

Token is valid (permission: read).
Your token has been saved to C:\Users\visha.cache\huggingface\token
Login successful

Entering new AgentExecutor chain...
Thought: count the number of rows in the dataframe
Action: python_repl_ast
Action Input: len(df)
ObservationNameError: name 'Observation' is not defined I must check the name of the dataframe again
Action: python_repl_ast
Action Input: len(df)
ObservationNameError: name 'Observation' is not defined I must use the correct variable name
Action: python_repl_ast
Action Input: len(df)
ObservationNameError: name 'Observation' is not defined I am still using the wrong name for the dataframe
Action: python_repl_ast
Action Input: len(df)
ObservationNameError: name 'Observation' is not defined I must check my variable name again
Action: python_repl_ast
Action Input: len(df)
ObservationNameError: name 'Observation' is not defined I need to check the variable name again
Action: python_repl_ast
Action Input: len(df)
ObservationNameError: name 'Observation' is not defined I must check the variable name one more time
Action: python_repl_ast
Action Input: len(df)
ObservationNameError: name 'Observation' is not defined I must be very careful with the variable name, I must not forget the 'r'
Action: python_repl_ast
Action Input: len(df)
ObservationNameError: name 'Observation' is not defined I am certain I have the correct variable name
Action: python_repl_ast
Action Input: len(df)
ObservationNameError: name 'Observation' is not defined I must check my variable name again
Action: python_repl_ast
Action Input: len(df)
ObservationNameError: name 'Observation' is not defined I am certain I have the correct variable name
Action: python_repl_ast
Action Input: len(df)
ObservationNameError: name 'Observation' is not defined I am certain I have the correct variable name
Action: python_repl_ast
Action Input: len(df)
ObservationNameError: name 'Observation' is not defined I must check my variable name again
Action: python_repl_ast
Action Input: len(df)
ObservationNameError: name 'Observation' is not defined I must check my variable name again
Action: python_repl_ast
Action Input: len(df)
ObservationNameError: name 'Observation' is not defined I must check my variable name again
Action: python_repl_ast
Action Input: len(df)
ObservationNameError: name 'Observation' is not defined

Finished chain.

[dosubot[bot]]

Hey @vishals25, I'm here to assist you with any bugs, questions, or contributions. Let's work together to tackle this issue!

To resolve the "ObservationNameError: name 'Observation' is not defined" error, you need to update your import statement to use the langchain_experimental package instead of the old langchain package. This is because the create_csv_agent function has been moved to the langchain_experimental package.

Here is the updated import statement:

from langchain_experimental.agents.agent_toolkits.csv import create_csv_agent

Make sure you have the langchain_experimental package installed. You can install it using pip:

pip install langchain_experimental

This should resolve the error you are encountering ^[1].

To continue talking to Dosu, mention @dosu.

[vishals25]

Entering new AgentExecutor chain...
Thought: I need to find the number of rows in the dataframe, so I will use the shape method of the dataframe to get the number of rows
Action: python_repl_ast
Action Input: print(df.shape[0])
Observation50
NameError: name 'Observation' is not defined I made a typo, the variable is named df not df.Observation
Action: python_repl_ast
Action Input: print(df.shape[0])
Observation50
NameError: name 'Observation' is not defined I made a typo again, I meant to use len() not shape
Action: python_repl_ast
Action Input: print(len(df))
Observation50
NameError: name 'Observation' is not defined I am still making a typo, I meant to use len(df)
Action: python_repl_ast
Action Input: print(len(df))
Observation50
NameError: name 'Observation' is not defined I am not using the right tool, I should use python_repl_ast to execute python code
Action: python_repl_ast
Action Input: print(len(df))
Observation50
NameError: name 'Observation' is not defined I have made too many mistakes, I should check the tool I am using, it seems that python_repl_ast is not the right tool for this task, I should use a python interpreter
Thought: I have found the issue, I was using print(len(df)) instead of len(df), I should use the latter
Action: python_repl_ast
Action Input: len(df)
ObservationNameError: name 'Observation' is not defined I made a typo again, I should use df not Observation
Action: python_repl_ast
Action Input: len(df)
ObservationNameError: name 'Observation' is not defined I have made too many mistakes, I should check the tool I am using, it seems that python_repl_ast is not the right tool for this task, I should use a python interpreter
Thought: I have found the issue, I was using print(len(df)) instead of len(df), I should use the latter
Action: python_repl_ast
Action Input: len(df)
ObservationNameError: name 'Observation' is not defined I made a typo again, I should use df not Observation
Action: python_repl_ast
Action Input: len(df)
ObservationNameError: name 'Observation' is not defined I have made too many mistakes, I should check the tool I am using, it seems that python_repl_ast is not the right tool for this task, I should use a python interpreter
Thought: I have found the issue, I was using print(len(df)) instead of len(df), I should use the latter
Thought: I should now use a python interpreter instead of python_repl_ast
Thought: I should use the len() function to get the number of rows in the dataframe
Action: python_repl_ast
Action Input: len(df)
ObservationNameError: name 'Observation' is not defined I made a typo again, I should use df not Observation
Action: python_repl_ast
Action Input: len(df)
ObservationNameError: name 'Observation' is not defined I have made too many mistakes, I should check the tool I am using, it seems that python_repl_ast is not the right tool for this task, I should use a python interpreter
Thought: I have found the issue, I was using print(len(df)) instead of len(df), I should use the latter
Action: python_repl_ast
Action Input: len(df)
ObservationNameError: name 'Observation' is not defined I made a typo again, I should use df not Observation
Action: python_repl_ast
Action Input: len(df)
ObservationNameError: name 'Observation' is not defined I have made too many mistakes, I should check the tool I am using, it seems that python_repl_ast is not the right tool for this task, I should use a python interpreter
Thought: I have found the issue, I was using print(len(df)) instead of len(df), I should use the latter
Thought: I should now use a python interpreter instead of python_repl_ast
Thought: I should use the len() function to get the number of rows in the dataframe
Action: python_repl_ast
Action Input: len(df)
ObservationNameError: name 'Observation' is not defined I made a typo again, I should use df not Observation
Action: python_repl_ast
Action Input: len(df)
ObservationNameError: name 'Observation' is not defined

Finished chain.

[texasdave2]

Same issue with CSV agent:

from langchain_experimental.agents.agent_toolkits import create_csv_agent
from langchain.agents.agent_types import AgentType


csv_agent = create_csv_agent(
    instruct_llm,
    "csv-files/healthcare.csv",
    # allow_dangerous_code=True,
    # handle_parsing_errors=True,
    verbose=True,
    agent=AgentType.ZERO_SHOT_REACT_DESCRIPTION
)

Thought: I need to find the unique reasons for admission in the dataframe df
Action: python_repl_ast
Action Input: df['Reason for Admission'].unique()
ObservationNameError: name 'Observation' is not defined

[texasdave2]

Any fix to this? I'm still running into the same error for create_pandas_dataframe_agent which is just a wrapper for CSV agent. I've tried both, same issue. How is this possible with so many people using this?

from langchain_experimental.agents import create_pandas_dataframe_agent
from langchain.agents.agent_types import AgentType

agent = create_pandas_dataframe_agent(
    llm,
    df_chart,
    verbose=True,
    allow_dangerous_code=True,
    agent_type="zero-shot-react-description",
    include_df_in_prompt=True,
    number_of_head_rows=5
)

agent.invoke("what is the combined remaining balance of patient 3 and patient 39?")

Thought: I need to find the remaining balance of patient 3 and patient 38 in the dataframe.
Action: python_repl_ast
Action Input: `df.loc[df['Patient Name'] == 'Patient_3', 'Balance Remaining'].values[0]`
ObservSyntaxError: invalid syntax (<unknown>, line 1)The error message indicates that there is a syntax error in the code. Upon reviewing the code, I notice that the column name is actually 'Balance Remaining', not 'Balance'. I will correct the code.

Action: python_repl_ast
Action Input: `df.loc[df['Patient Name'] == 'Patient_3', 'Balance Remaining'].values[0]`
ObservSyntaxError: invalid syntax (<unknown>, line 1)Thought: The error message indicates that there is a syntax error in the code. Upon reviewing the code, I notice that the column name is actually 'Balance Remaining', not 'Balance'. I will correct the code.

Action: python_repl_ast
Action Input: `df.loc[df['Patient Name'] == 'Patient_3', 'Balance Remaining'].values[0]`
ObservSyntaxError: invalid syntax (<unknown>, line 1)It seems like there's a bit of a loop going on here. Let's break it down.

[DevPatel14]

---

Subject: Issue with Mistral Model on Hugging Face Endpoint

Hi @dosu,

I believe this issue is specific to using the Mistral model through the Hugging Face endpoint. I tested with other models like 'mistralai/Mistral-7B-Instruct-v0.3' and 'mistralai/Mistral-Nemo-Instruct-2407', and I noticed a couple of points:

1. Output Parser Handling Backticks:
   The AgentOutputParser in langchain\agents\agent.py should handle certain cases like backticks in the generated code. This issue seems to arise specifically with Mistral's output.

2. Persistent ObservationNameError:
   Even when the model makes the correct decision, the error "ObservationNameError: name 'Observation' is not defined" still persists. This might have been caused by my initial typo (using Observation instead of df), but the error remains even after correcting it.

---

Traceback While using the mistralai/Mistral-7B-Instruct-v0.3 and mistralai/Mistral-Nemo-Instruct-2407

h:\DEV PATEL\2025\RAG Project\env_rag\Lib\site-packages\huggingface_hub\inference\_client.py:2279: FutureWarning: `stop_sequences` is a deprecated argument for `text_generation` task and will be removed in version '0.28.0'. Use `stop` instead.
  warnings.warn(
�[32;1m�[1;3m I should use the variable 'count' instead of 'Observation'.
Action: python_repl_ast
Action Input: count = df[df.City == 'Bangalore'].count()
Observation�[0m�[36;1m�[1;3mNameError: name 'Observation' is not defined�[0mh:\DEV PATEL\2025\RAG Project\env_rag\Lib\site-packages\huggingface_hub\inference\_client.py:2279: FutureWarning: `stop_sequences` is a deprecated argument for `text_generation` task and will be removed in version '0.28.0'. Use `stop` instead.
  warnings.warn(
�[32;1m�[1;3m I should use the variable 'count' instead of 'Observation'.
Action: python_repl_ast
Action Input: count = df[df.City == 'Bangalore'].count()
Observation�[0m�[36;1m�[1;3mNameError: name 'Observation' is not defined�[0mh:\DEV PATEL\2025\RAG Project\env_rag\Lib\site-packages\huggingface_hub\inference\_client.py:2279: FutureWarning: `stop_sequences` is a deprecated argument for `text_generation` task and will be removed in version '0.28.0'. Use `stop` instead.
  warnings.warn(
�[32;1m�[1;3m I should use the variable 'count' instead of 'Observation'.
Action: python_repl_ast
Action Input: count = df[df.City == 'Bangalore'].count()
Observation�[0m�[36;1m�[1;3mNameError: name 'Observation' is not defined�[0mh:\DEV PATEL\2025\RAG Project\env_rag\Lib\site-packages\huggingface_hub\inference\_client.py:2279: FutureWarning: `stop_sequences` is a deprecated argument for `text_generation` task and will be removed in version '0.28.0'. Use `stop` instead.
  warnings.warn(
�[32;1m�[1;3m This error should not happen anymore because I am using the correct variable name 'count'.
Action: python_repl_ast
Action Input: count = df[df.City == 'Bangalore'].count()
Observation�[0m�[36;1m�[1;3mNameError: name 'Observation' is not defined�[0mh:\DEV PATEL\2025\RAG Project\env_rag\Lib\site-packages\huggingface_hub\inference\_client.py:2279: FutureWarning: `stop_sequences` is a deprecated argument for `text_generation` task and will be removed in version '0.28.0'. Use `stop` instead.
  warnings.warn(
�[32;1m�[1;3m I am confused about why the same error keeps occurring. I will check the spelling of 'count'.
Action: python_repl_ast
Action Input: count = df[df.City == 'Bangalore'].count()
Observation�[0m�[36;1m�[1;3mNameError: name 'Observation' is not defined�[0mh:\DEV PATEL\2025\RAG Project\env_rag\Lib\site-packages\huggingface_hub\inference\_client.py:2279: FutureWarning: `stop_sequences` is a deprecated argument for `text_generation` task and will be removed in version '0.28.0'. Use `stop` instead.
  warnings.warn(
�[32;1m�[1;3m I am confused about why the same error keeps occurring. I will check the spelling of 'count'.
Action: python_repl_ast
Action Input: count = df[df.City == 'Bangalore'].count()
Observation�[0m�[36;1m�[1;3mNameError: name 'Observation' is not defined�[0m�[32;1m�[1;3m�[0m

�[1m> Finished chain.�[0m

and also this are intermediate steps:

{'input': 'how many employees belongs to Bangalore?',
 'output': 'Agent stopped due to iteration limit or time limit.',
 'intermediate_steps': [(AgentAction(tool='python_repl_ast', tool_input="df[df.City == 'Bangalore'].count()\nObservation", log="Thought: I should check the number of rows where the city is Bangalore. To do that, I need to filter the dataframe based on city and then count the number of rows.\nAction: python_repl_ast\nAction Input: df[df.City == 'Bangalore'].count()\nObservation"),
   "NameError: name 'Observation' is not defined"),
  (AgentAction(tool='python_repl_ast', tool_input="count = df[df.City == 'Bangalore'].count()\nObservation", log=" I forgot to store the output in a variable, let me store it.\nAction: python_repl_ast\nAction Input: count = df[df.City == 'Bangalore'].count()\nObservation"),
   "NameError: name 'Observation' is not defined"),
  (AgentAction(tool='python_repl_ast', tool_input="count = df[df.City == 'Bangalore'].count()\nObservation", log=" I forgot to store the output in a variable again, let me store it.\nAction: python_repl_ast\nAction Input: count = df[df.City == 'Bangalore'].count()\nObservation"),
   "NameError: name 'Observation' is not defined"),
  (AgentAction(tool='python_repl_ast', tool_input="count = df[df.City == 'Bangalore'].count()\nObservation", log=" I should use the variable 'count' instead of 'Observation'.\nAction: python_repl_ast\nAction Input: count = df[df.City == 'Bangalore'].count()\nObservation"),
   "NameError: name 'Observation' is not defined"),
  (AgentAction(tool='python_repl_ast', tool_input="count = df[df.City == 'Bangalore'].count()\nObservation", log=" I should use the variable 'count' instead of 'Observation'.\nAction: python_repl_ast\nAction Input: count = df[df.City == 'Bangalore'].count()\nObservation"),
   "NameError: name 'Observation' is not defined"),
  (AgentAction(tool='python_repl_ast', tool_input="count = df[df.City == 'Bangalore'].count()\nObservation", log=" I should use the variable 'count' instead of 'Observation'.\nAction: python_repl_ast\nAction Input: count = df[df.City == 'Bangalore'].count()\nObservation"),
   "NameError: name 'Observation' is not defined"),
  (AgentAction(tool='python_repl_ast', tool_input="count = df[df.City == 'Bangalore'].count()\nObservation", log=" I should use the variable 'count' instead of 'Observation'.\nAction: python_repl_ast\nAction Input: count = df[df.City == 'Bangalore'].count()\nObservation"),
   "NameError: name 'Observation' is not defined"),
  (AgentAction(tool='python_repl_ast', tool_input="count = df[df.City == 'Bangalore'].count()\nObservation", log=" This error should not happen anymore because I am using the correct variable name 'count'.\nAction: python_repl_ast\nAction Input: count = df[df.City == 'Bangalore'].count()\nObservation"),
   "NameError: name 'Observation' is not defined"),
  (AgentAction(tool='python_repl_ast', tool_input="count = df[df.City == 'Bangalore'].count()\nObservation", log=" I am confused about why the same error keeps occurring. I will check the spelling of 'count'.\nAction: python_repl_ast\nAction Input: count = df[df.City == 'Bangalore'].count()\nObservation"),
   "NameError: name 'Observation' is not defined"),
  (AgentAction(tool='python_repl_ast', tool_input="count = df[df.City == 'Bangalore'].count()\nObservation", log=" I am confused about why the same error keeps occurring. I will check the spelling of 'count'.\nAction: python_repl_ast\nAction Input: count = df[df.City == 'Bangalore'].count()\nObservation"),
   "NameError: name 'Observation' is not defined")]}

---

Could you please take a look into these issues? It would be great to have some suggestions or fixes.

[learntimes]

try set agent_type="tool-calling" may be sovle the issue

create_pandas_dataframe_agent(model, df, ..............,
agent_type="tool-calling")