Update ig.py
Browse files
ig.py
CHANGED
@@ -1,10 +1,21 @@
|
|
1 |
from transformers import AutoProcessor, AutoModelForCausalLM
|
|
|
2 |
import gradio as gr
|
3 |
import torch
|
4 |
|
5 |
-
# Load the processor
|
6 |
processor = AutoProcessor.from_pretrained("microsoft/git-base")
|
7 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
8 |
|
9 |
def predict(image):
|
10 |
try:
|
@@ -28,7 +39,7 @@ def predict(image):
|
|
28 |
print("Error during prediction:", str(e))
|
29 |
return "Error: " + str(e)
|
30 |
|
31 |
-
#
|
32 |
with gr.Blocks() as demo:
|
33 |
image = gr.Image(type="pil")
|
34 |
predict_btn = gr.Button("Predict", variant="primary")
|
|
|
1 |
from transformers import AutoProcessor, AutoModelForCausalLM
|
2 |
+
from peft import PeftModel, PeftConfig
|
3 |
import gradio as gr
|
4 |
import torch
|
5 |
|
6 |
+
# Load the processor
|
7 |
processor = AutoProcessor.from_pretrained("microsoft/git-base")
|
8 |
+
|
9 |
+
# Load the base model (the pre-trained model you're adapting with LoRA)
|
10 |
+
base_model = AutoModelForCausalLM.from_pretrained("microsoft/git-base")
|
11 |
+
|
12 |
+
# Load the adapter configuration
|
13 |
+
adapter_config_path = "./" # Path to your adapter_config.json
|
14 |
+
adapter_model_path = "./" # Path to your adapter_model.safetensors
|
15 |
+
|
16 |
+
# Load the LoRA adapter using Peft
|
17 |
+
peft_config = PeftConfig.from_pretrained(adapter_config_path)
|
18 |
+
model = PeftModel.from_pretrained(base_model, adapter_model_path, config=peft_config)
|
19 |
|
20 |
def predict(image):
|
21 |
try:
|
|
|
39 |
print("Error during prediction:", str(e))
|
40 |
return "Error: " + str(e)
|
41 |
|
42 |
+
# Gradio Interface
|
43 |
with gr.Blocks() as demo:
|
44 |
image = gr.Image(type="pil")
|
45 |
predict_btn = gr.Button("Predict", variant="primary")
|