Use Florence-2-base-ft

app.py

@@ -38,13 +38,13 @@ import subprocess
 device = get_device_type()
 if (device == "cuda"):
     subprocess.run('pip install flash-attn --no-build-isolation', env={'FLASH_ATTENTION_SKIP_CUDA_BUILD': "TRUE"}, shell=True)
-    model = AutoModelForCausalLM.from_pretrained("microsoft/Florence-2-large-ft", trust_remote_code=True)
-    processor = AutoProcessor.from_pretrained("microsoft/Florence-2-large-ft", trust_remote_code=True)
+    model = AutoModelForCausalLM.from_pretrained("microsoft/Florence-2-base-ft", trust_remote_code=True)
+    processor = AutoProcessor.from_pretrained("microsoft/Florence-2-base-ft", trust_remote_code=True)
 else:
-    #https://huggingface.co/microsoft/Florence-2-large-ft/discussions/4
+    #https://huggingface.co/microsoft/Florence-2-base-ft/discussions/4
     with patch("transformers.dynamic_module_utils.get_imports", fixed_get_imports):
-        model = AutoModelForCausalLM.from_pretrained("microsoft/Florence-2-large-ft", trust_remote_code=True)
-        processor = AutoProcessor.from_pretrained("microsoft/Florence-2-large-ft", trust_remote_code=True)
+        model = AutoModelForCausalLM.from_pretrained("microsoft/Florence-2-base-ft", trust_remote_code=True)
+        processor = AutoProcessor.from_pretrained("microsoft/Florence-2-base-ft", trust_remote_code=True)
 
 DESCRIPTION = "# [Florence-2 base-ft Demo with CPU inference support](https://huggingface.co/microsoft/Florence-2-base-ft)"