video datasets

Former-commit-id: 8cafc7b055

src/llamafactory/webui/chatter.py

@@ -133,6 +133,7 @@ class WebChatModel(ChatModel):
         system: str,
         tools: str,
         image: Optional[Any],
+        video: Optional[Any],
         max_new_tokens: int,
         top_p: float,
         temperature: float,
@@ -140,7 +141,7 @@ class WebChatModel(ChatModel):
         chatbot[-1][1] = ""
         response = ""
         for new_text in self.stream_chat(
-            messages, system, tools, image, max_new_tokens=max_new_tokens, top_p=top_p, temperature=temperature
+            messages, system, tools, image, video, max_new_tokens=max_new_tokens, top_p=top_p, temperature=temperature
         ):
             response += new_text
             if tools:

src/llamafactory/webui/components/chatbot.py

@@ -43,8 +43,12 @@ def create_chat_box(
                         system = gr.Textbox(show_label=False)
                         tools = gr.Textbox(show_label=False, lines=3)
 
-                    with gr.Column() as image_box:
-                        image = gr.Image(sources=["upload"], type="pil")
+                    with gr.Column() as mm_box:
+                        with gr.Tab("Image"):
+                            image = gr.Image(sources=["upload"], type="pil")
+
+                        with gr.Tab("Video"):
+                            video = gr.Video(sources=["upload"])
 
                 query = gr.Textbox(show_label=False, lines=8)
                 submit_btn = gr.Button(variant="primary")
@@ -63,7 +67,7 @@ def create_chat_box(
         [chatbot, messages, query],
     ).then(
         engine.chatter.stream,
-        [chatbot, messages, system, tools, image, max_new_tokens, top_p, temperature],
+        [chatbot, messages, system, tools, image, video, max_new_tokens, top_p, temperature],
         [chatbot, messages],
     )
     clear_btn.click(lambda: ([], []), outputs=[chatbot, messages])
@@ -76,8 +80,9 @@ def create_chat_box(
             role=role,
             system=system,
             tools=tools,
-            image_box=image_box,
+            mm_box=mm_box,
             image=image,
+            video=video,
             query=query,
             submit_btn=submit_btn,
             max_new_tokens=max_new_tokens,

src/llamafactory/webui/components/infer.py

@@ -68,7 +68,7 @@ def create_infer_tab(engine: "Engine") -> Dict[str, "Component"]:
     engine.manager.get_elem_by_id("top.model_name").change(
         lambda model_name: gr.Column(visible=get_visual(model_name)),
         [engine.manager.get_elem_by_id("top.model_name")],
-        [chat_elems["image_box"]],
+        [chat_elems["mm_box"]],
     )
 
     return elem_dict

src/llamafactory/webui/engine.py

@@ -59,7 +59,7 @@ class Engine:
             init_dict["train.output_dir"] = {"value": "train_{}".format(current_time)}
             init_dict["train.config_path"] = {"value": "{}.yaml".format(current_time)}
             init_dict["eval.output_dir"] = {"value": "eval_{}".format(current_time)}
-            init_dict["infer.image_box"] = {"visible": False}
+            init_dict["infer.mm_box"] = {"visible": False}
 
             if user_config.get("last_model", None):
                 init_dict["top.model_name"] = {"value": user_config["last_model"]}

src/llamafactory/webui/locales.py

@@ -1691,6 +1691,20 @@ LOCALES = {
             "label": "이미지 (선택 사항)",
         },
     },
+    "video": {
+        "en": {
+            "label": "Video (optional)",
+        },
+        "ru": {
+            "label": "Видео (по желанию)",
+        },
+        "zh": {
+            "label": "视频（非必填）",
+        },
+        "ko": {
+            "label": "비디오 (선택 사항)",
+        },
+    },
     "query": {
         "en": {
             "placeholder": "Input...",