dfdfd

app.py

@@ -29,48 +29,32 @@ def predict_image(name_image):
 	categories = ['Normal', 'Tuberculose']
 	url_image = './'
 
-#	'./' + name_image
-
-	# from huggingface_hub import HfFileSystem
-	# model_repo = 'valencar/modelo_raios_x'
-	#fs = HfFileSystem()
-	#file_hf = "hf://valencar/modelo_raios_x/dados_teste/" + name_image
-
-	# from datasets import Dataset, Image
-	# dataset = Dataset.from_dict({"image": ["path/to/image_1", "path/to/image_2", ..., "path/to/image_n"]}).cast_column("image", Image())
-	# image_hf = dataset[0]["image"]
-# <PIL.PngImagePlugin.PngImageFile image mode=RGBA size=1200x215 at 0x15E6D7160>]
-
-
-	
-	# image_hf = "valencar/modelo_raios_x/dados_teste/" + name_image 
 
 #	image_hf = "valencar/modelo_raios_x/dados_teste/" + "CHNCXR_0001_0.png"
 
-
 	from PIL import Image
 	import requests
 
-
-
 	# url = "https://huggingface.co/valencar/modelo_raios_x/blob/main/dados_teste/CHNCXR_0001_0.png"
 
 	# req = requests.get(url, stream=True).raw
 	# #image_hf = Image.open(req)
 	# image_hf = req
 
-	imagem = Image.open(name_image)
-	imagem.save("image_raiox","png")
-	nome_imagem_salva = 'image_raiox.png'
-	image_hf = nome_imagem_salva
+	hf_home = '/root/.cache/huggingface/'
+
+	# imagem = Image.open(hf_home + name_image)
+	# imagem.save("image_raiox","png")
+	# nome_imagem_salva = 'image_raiox.png'
+	# image_hf = nome_imagem_salva
 	
-	img = img_orig = image.load_img(image_hf, target_size = (IMAGE_HEIGHT, IMAGE_WIDTH))
+	img = img_orig = image.load_img(hf_home + name_image, target_size = (IMAGE_HEIGHT, IMAGE_WIDTH))
 	img = image.img_to_array(img)
 	img = np.expand_dims(img, axis = 0)
 	img = img/255.0
+  
 
-
-	col1.image(image_hf, width=450)
+	col1.image(hf_home + name_image, width=450)
 
 	pred = st.session_state.model.predict(img)
 	classe = np.argmax(pred)