Remove llm argument from ToolCallAccuracy initialization (#1891)

sahusiddharth · web-flow · commit 964acfcb3e61 · 2025-02-02T10:28:05.000-08:00
Error Message
```
TypeError: ToolCallAccuracy.__init__() got an unexpected keyword argument 'llm'
```

This fix resolves the above TypeError caused by passing an unexpected
llm keyword argument to ToolCallAccuracy.__init__(). The llm argument
has been removed from all instances where it was being passed.
diff --git a/docs/concepts/metrics/available_metrics/agents.md b/docs/concepts/metrics/available_metrics/agents.md
@@ -103,7 +103,7 @@ sample = MultiTurnSample(
     ]
 )
 
-scorer = ToolCallAccuracy(llm = evaluator_llm)
+scorer = ToolCallAccuracy()
 await scorer.multi_turn_ascore(sample)
 ```
 Output
@@ -119,7 +119,7 @@ By default the tool names and arguments are compared using exact string matching
 from ragas.metrics._string import NonLLMStringSimilarity
 from ragas.metrics._tool_call_accuracy import ToolCallAccuracy
 
-metric = ToolCallAccuracy(llm = evaluator_llm)
+metric = ToolCallAccuracy()
 metric.arg_comparison_metric = NonLLMStringSimilarity()
 ```
 
diff --git a/docs/howtos/integrations/_langgraph_agent_evaluation.md b/docs/howtos/integrations/_langgraph_agent_evaluation.md
@@ -335,7 +335,6 @@ sample = MultiTurnSample(
 )
 
 tool_accuracy_scorer = ToolCallAccuracy()
-tool_accuracy_scorer.llm = ChatOpenAI(model="gpt-4o-mini")
 await tool_accuracy_scorer.multi_turn_ascore(sample)
 ```
 
diff --git a/docs/howtos/integrations/langgraph_agent_evaluation.ipynb b/docs/howtos/integrations/langgraph_agent_evaluation.ipynb
@@ -601,7 +601,6 @@
     ")\n",
     "\n",
     "tool_accuracy_scorer = ToolCallAccuracy()\n",
-    "tool_accuracy_scorer.llm = ChatOpenAI(model=\"gpt-4o-mini\")\n",
     "await tool_accuracy_scorer.multi_turn_ascore(sample)"
    ]
   },
diff --git a/docs/howtos/integrations/swarm_agent_evaluation.md b/docs/howtos/integrations/swarm_agent_evaluation.md
@@ -332,7 +332,6 @@ sample = MultiTurnSample(
 )
 
 tool_accuracy_scorer = ToolCallAccuracy()
-tool_accuracy_scorer.llm = ChatOpenAI(model="gpt-4o-mini")
 await tool_accuracy_scorer.multi_turn_ascore(sample)
 ```
 Output
@@ -381,7 +380,6 @@ sample = MultiTurnSample(
 )
 
 tool_accuracy_scorer = ToolCallAccuracy()
-tool_accuracy_scorer.llm = ChatOpenAI(model="gpt-4o-mini")
 await tool_accuracy_scorer.multi_turn_ascore(sample)
 ```
 Output

Original file line number	Diff line number	Diff line change
`@@ -103,7 +103,7 @@ sample = MultiTurnSample(`
`103`	`103`	`]`
`104`	`104`	`)`
`105`	`105`
`106`		`-scorer = ToolCallAccuracy(llm = evaluator_llm)`
	`106`	`+scorer = ToolCallAccuracy()`
`107`	`107`	`await scorer.multi_turn_ascore(sample)`
`108`	`108`	```
`109`	`109`	`Output`
`@@ -119,7 +119,7 @@ By default the tool names and arguments are compared using exact string matching`
`119`	`119`	`from ragas.metrics._string import NonLLMStringSimilarity`
`120`	`120`	`from ragas.metrics._tool_call_accuracy import ToolCallAccuracy`
`121`	`121`
`122`		`-metric = ToolCallAccuracy(llm = evaluator_llm)`
	`122`	`+metric = ToolCallAccuracy()`
`123`	`123`	`metric.arg_comparison_metric = NonLLMStringSimilarity()`
`124`	`124`	```
`125`	`125`
Original file line number	Diff line number	Diff line change
`@@ -335,7 +335,6 @@ sample = MultiTurnSample(`
`335`	`335`	`)`
`336`	`336`
`337`	`337`	`tool_accuracy_scorer = ToolCallAccuracy()`
`338`		`-tool_accuracy_scorer.llm = ChatOpenAI(model="gpt-4o-mini")`
`339`	`338`	`await tool_accuracy_scorer.multi_turn_ascore(sample)`
`340`	`339`	```
`341`	`340`
Original file line number	Diff line number	Diff line change
`@@ -601,7 +601,6 @@`
`601`	`601`	`")\n",`
`602`	`602`	`"\n",`
`603`	`603`	`"tool_accuracy_scorer = ToolCallAccuracy()\n",`
`604`		`- "tool_accuracy_scorer.llm = ChatOpenAI(model=\"gpt-4o-mini\")\n",`
`605`	`604`	`"await tool_accuracy_scorer.multi_turn_ascore(sample)"`
`606`	`605`	`]`
`607`	`606`	`},`
Original file line number	Diff line number	Diff line change
`@@ -332,7 +332,6 @@ sample = MultiTurnSample(`
`332`	`332`	`)`
`333`	`333`
`334`	`334`	`tool_accuracy_scorer = ToolCallAccuracy()`
`335`		`-tool_accuracy_scorer.llm = ChatOpenAI(model="gpt-4o-mini")`
`336`	`335`	`await tool_accuracy_scorer.multi_turn_ascore(sample)`
`337`	`336`	```
`338`	`337`	`Output`
`@@ -381,7 +380,6 @@ sample = MultiTurnSample(`
`381`	`380`	`)`
`382`	`381`
`383`	`382`	`tool_accuracy_scorer = ToolCallAccuracy()`
`384`		`-tool_accuracy_scorer.llm = ChatOpenAI(model="gpt-4o-mini")`
`385`	`383`	`await tool_accuracy_scorer.multi_turn_ascore(sample)`
`386`	`384`	```
`387`	`385`	`Output`