update

app.py

@@ -1,10 +1,10 @@
 from pathlib import Path
 from typing import List, Dict, Tuple
 import matplotlib.colors as mpl_colors
-import os
 import pandas as pd
 import seaborn as sns
 import shinyswatch
+import run
 
 from shiny import App, Inputs, Outputs, Session, reactive, render, req, ui
 from transformers import SamModel, SamConfig, SamProcessor
@@ -46,36 +46,8 @@ def server(input: Inputs, output: Outputs, session: Session):
         if input.image_input():
             src = input.image_input()[0]['datapath']
             img = {"src": src, "width": "500px"} 
-
-            # Specify the cache directory
-            transformers_cache_dir = "/usr/local/lib/python3.9/site-packages/transformers"
-
-            # Set the TRANSFORMERS_CACHE environment variable
-            os.environ["TRANSFORMERS_CACHE"] = transformers_cache_dir
-
-            # Load the model configuration
-            model_config = SamConfig.from_pretrained("facebook/sam-vit-base")
-            processor = SamProcessor.from_pretrained("facebook/sam-vit-base")
-
-            # Create an instance of the model architecture with the loaded configuration
-            my_sam_model = SamModel(config=model_config)
-            #Update the model by loading the weights from saved file.
-            my_sam_model.load_state_dict(torch.load("sam_model.pth", map_location=torch.device('cpu')))
-
-            new_image = np.array(Image.open(src))
-            inputs = processor(new_image, return_tensors="pt")
-            inputs = {k: v.to(device) for k, v in inputs.items()}
-            # my_sam_model.eval()
-            # # forward pass
-            # with torch.no_grad():
-            #     outputs = my_sam_model(**inputs, multimask_output=False)
-
-            # # apply sigmoid
-            # single_patch_prob = torch.sigmoid(outputs.pred_masks.squeeze(1))
-            # # convert soft mask to hard mask
-            # single_patch_prob = single_patch_prob.cpu().numpy().squeeze()
-            # single_patch_prediction = (single_patch_prob > 0.5).astype(np.uint8)
-
+            x = run.pred(src)
+            print(x)
             return img
         return None
 

run.py

@@ -1,26 +1,31 @@
-# from transformers import SamModel, SamConfig, SamProcessor
-# import torch
-# import numpy as np
-# import matplotlib.pyplot as plt
-# import app
+from transformers import SamModel, SamConfig, SamProcessor
+import torch
+import numpy as np
+import matplotlib.pyplot as plt
+import app
 
-# # Load the model configuration
-# model_config = SamConfig.from_pretrained("facebook/sam-vit-base")
-# processor = SamProcessor.from_pretrained("facebook/sam-vit-base")
+def pred(src):
+    # Load the model configuration
+    model_config = SamConfig.from_pretrained("facebook/sam-vit-base")
+    processor = SamProcessor.from_pretrained("facebook/sam-vit-base")
 
-# # Create an instance of the model architecture with the loaded configuration
-# my_sam_model = SamModel(config=model_config)
-# #Update the model by loading the weights from saved file.
-# my_sam_model.load_state_dict(torch.load("sam_model.pth", map_location=torch.device('cpu')))
+    # Create an instance of the model architecture with the loaded configuration
+    my_sam_model = SamModel(config=model_config)
+    #Update the model by loading the weights from saved file.
+    my_sam_model.load_state_dict(torch.load("sam_model.pth", map_location=torch.device('cpu')))
 
+    new_image = np.array(Image.open(src))
+    inputs = processor(new_image, return_tensors="pt")
+    inputs = {k: v.to(device) for k, v in inputs.items()}
+    x = 1
+    # my_sam_model.eval()
+    # # forward pass
+    # with torch.no_grad():
+    #     outputs = my_sam_model(**inputs, multimask_output=False)
 
-# load image
-# test_image = {"src": src, "width": "500px"} 
-
-import os
-
-# Specify the cache directory
-transformers_cache_dir = "/usr/local/lib/python3.9/site-packages/transformers"
-
-# Set the TRANSFORMERS_CACHE environment variable
-os.environ["TRANSFORMERS_CACHE"] = transformers_cache_dir
+    # # apply sigmoid
+    # single_patch_prob = torch.sigmoid(outputs.pred_masks.squeeze(1))
+    # # convert soft mask to hard mask
+    # single_patch_prob = single_patch_prob.cpu().numpy().squeeze()
+    # single_patch_prediction = (single_patch_prob > 0.5).astype(np.uint8)
+    return x