fix video-temporal-localization

Dayof · Dayof · commit 8a8c5d1dacb1 · 2024-10-03T22:49:10.000-03:00
diff --git a/vision_agent/tools/tools.py b/vision_agent/tools/tools.py
@@ -899,7 +899,10 @@ def ixc25_temporal_localization(prompt: str, frames: List[np.ndarray]) -> List[b
         "function_name": "ixc25_temporal_localization",
     }
     data: List[int] = send_inference_request(
-        payload, "video-temporal-localization", files=files, v2=True
+        payload,
+        "video-temporal-localization?model=internlm-xcomposer",
+        files=files,
+        v2=True,
     )
     chunk_size = round(len(frames) / len(data))
     data_explode = [[elt] * chunk_size for elt in data]

Original file line number	Diff line number	Diff line change
`@@ -899,7 +899,10 @@ def ixc25_temporal_localization(prompt: str, frames: List[np.ndarray]) -> List[b`
`899`	`899`	`"function_name": "ixc25_temporal_localization",`
`900`	`900`	`}`
`901`	`901`	`data: List[int] = send_inference_request(`
`902`		`- payload, "video-temporal-localization", files=files, v2=True`
	`902`	`+ payload,`
	`903`	`+ "video-temporal-localization?model=internlm-xcomposer",`
	`904`	`+ files=files,`
	`905`	`+ v2=True,`
`903`	`906`	`)`
`904`	`907`	`chunk_size = round(len(frames) / len(data))`
`905`	`908`	`data_explode = [[elt] * chunk_size for elt in data]`