You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Traceback (most recent call last):
File "/root/spider_eval/Spider2/spider2-snow/evaluation_suite/evaluate.py", line 429, in <module>
evaluate_spider2sql(args)
File "/root/spider_eval/Spider2/spider2-snow/evaluation_suite/evaluate.py", line 345, in evaluate_spider2sql
exe_flag, dbms_error_info = get_snowflake_sql_result(pred_sql_query, database_id, True, "temp", f"{id}.csv")
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/root/spider_eval/Spider2/spider2-snow/evaluation_suite/evaluate.py", line 183, in get_snowflake_sql_result
conn = snowflake.connector.connect(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/root/spider_eval/Spider2/methods/spider-agent-snow/venv/lib/python3.12/site-packages/snowflake/connector/__init__.py", line 55, in Connect
return SnowflakeConnection(**kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/root/spider_eval/Spider2/methods/spider-agent-snow/venv/lib/python3.12/site-packages/snowflake/connector/connection.py", line 467, in __init__
self.connect(**kwargs)
File "/root/spider_eval/Spider2/methods/spider-agent-snow/venv/lib/python3.12/site-packages/snowflake/connector/connection.py", line 786, in connect
self.__open_connection()
File "/root/spider_eval/Spider2/methods/spider-agent-snow/venv/lib/python3.12/site-packages/snowflake/connector/connection.py", line 1116, in __open_connection
self.authenticate_with_retry(self.auth_class)
File "/root/spider_eval/Spider2/methods/spider-agent-snow/venv/lib/python3.12/site-packages/snowflake/connector/connection.py", line 1404, in authenticate_with_retry
self._authenticate(auth_instance)
File "/root/spider_eval/Spider2/methods/spider-agent-snow/venv/lib/python3.12/site-packages/snowflake/connector/connection.py", line 1432, in _authenticate
auth.authenticate(
File "/root/spider_eval/Spider2/methods/spider-agent-snow/venv/lib/python3.12/site-packages/snowflake/connector/auth/_auth.py", line 416, in authenticate
Error.errorhandler_wrapper(
File "/root/spider_eval/Spider2/methods/spider-agent-snow/venv/lib/python3.12/site-packages/snowflake/connector/errors.py", line 283, in errorhandler_wrapper
handed_over = Error.hand_to_other_handler(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/root/spider_eval/Spider2/methods/spider-agent-snow/venv/lib/python3.12/site-packages/snowflake/connector/errors.py", line 341, in hand_to_other_handler
connection.errorhandler(connection, cursor, error_class, error_value)
File "/root/spider_eval/Spider2/methods/spider-agent-snow/venv/lib/python3.12/site-packages/snowflake/connector/errors.py", line 214, in default_errorhandler
raise error_class(
snowflake.connector.errors.DatabaseError: 250001 (08001): Failed to connect to DB: RSRSBDK-YDB67606.snowflakecomputing.com:443. Incorrect username or password was specified.
I note that this would be because the evaluation_suite/snowflake_credential.json isn't populated correctly with the provided username and password given to access the snowflake database, but I specifically made sure that it was so.
The same credentials work when running the initial runs with the model as,
python3 run.py --model gpt-4o-mini -s test2
so I'm not sure what the problem is honestly.
I made sure credentials were being loaded properly into snowflake.connector.connect() in evaluation_suite/evaluate.py but even with some modifications,
def load_json(filename):
with open(filename, 'r') as file:
data = json.load(file)
return data
def get_snowflake_sql_result(sql_query, database_id, is_save, save_dir=None, file_name="result.csv"):
"""
is_save = True, output a 'result.csv'
if_save = False, output a string
"""
snowflake_credential = load_json('snowflake_credential.json')
# snowflake_credential = json.load(open('snowflake_credential.json'))
conn = snowflake.connector.connect(
**snowflake_credential,
database=database_id,
)
creds. aren't recognized and can't access the database for computing final evals based on SQL.
Perhaps you guys don't intend for us to eval this way given your update on the README.md
2024-12-26: Using Spider-Agent to benchmark your LLMs! Considering the widespread attention to the traditional text-to-SQL setting, we now recommend using [spider-agent-lite](https://github.com/xlang-ai/Spider2/tree/main/methods/spider-agent-lite) and [spider-agent-snow](https://github.com/xlang-ai/Spider2/tree/main/methods/spider-agent-snow) to work with [spider2-lite](https://github.com/xlang-ai/Spider2/tree/main/spider2-lite/evaluation_suite) and [spider2-snow](https://github.com/xlang-ai/Spider2/tree/main/spider2-snow/evaluation_suite) for benchmarking your LLMs. The final output should be CSV files, not SQLs.
but only 281 out of the 547 examples sampled from the LLM are rendered as CSV files and the rest are as SQL queries, so it's uncertain how to evaluate the 266 SQL queries if the above doesn't work.
The text was updated successfully, but these errors were encountered:
When trying to compute the final eval using,
I run across the following:
I note that this would be because the
evaluation_suite/snowflake_credential.json
isn't populated correctly with the provided username and password given to access the snowflake database, but I specifically made sure that it was so.The same credentials work when running the initial runs with the model as,
so I'm not sure what the problem is honestly.
I made sure credentials were being loaded properly into
snowflake.connector.connect()
inevaluation_suite/evaluate.py
but even with some modifications,creds. aren't recognized and can't access the database for computing final evals based on SQL.
Perhaps you guys don't intend for us to eval this way given your update on the
README.md
but only 281 out of the 547 examples sampled from the LLM are rendered as CSV files and the rest are as SQL queries, so it's uncertain how to evaluate the 266 SQL queries if the above doesn't work.
The text was updated successfully, but these errors were encountered: