ss

Dockerfile

@@ -16,12 +16,6 @@ RUN pip3 install -r requirements.txt
 
 RUN chmod -R 755 /app
 
-RUN mkdir -p /app/src/downloads && chmod -R 755 /app/src
-
-RUN mkdir -p /app/src/downloads/extracted && chmod -R 755 /app/src
-
-RUN mkdir -p /app/src/imagenet_resized && chmod -R 755 /app/src
-
 EXPOSE 8501
 
 HEALTHCHECK CMD curl --fail http://localhost:8501/_stcore/health

requirements.txt

@@ -7,4 +7,5 @@ scikit-learn
 seaborn
 huggingface_hub
 pandas
-python-dotenv
+python-dotenv
+datasets

src/streamlit_app.py

@@ -7,6 +7,7 @@ import tensorflow.keras as keras
 from tensorflow.keras.applications import VGG16
 from tensorflow.keras.layers import Dense, Flatten
 from tensorflow.keras.models import Model, load_model
+from datasets import load_dataset
 import matplotlib.pyplot as plt
 from sklearn.metrics import confusion_matrix
 import seaborn as sns
@@ -14,7 +15,7 @@ from huggingface_hub import HfApi
 import os
 
 # 📌 Percorso della cartella dove sarà salvato il dataset
-DATA_DIR = "/tmp"
+DATA_DIR = "/app" #"/tmp"
 
 # 📌 Autenticazione Hugging Face dal Secret nello Space
 HF_TOKEN = os.getenv("HF_TOKEN")
@@ -32,7 +33,12 @@ else:
 
 # 📌 Carica solo 300 immagini da `imagenet_resized/64x64`
 st.write("🔄 Caricamento di 300 immagini da `imagenet_resized/64x64`...")
-imagenet = tfds.load("imagenet_resized/64x64", split="train", as_supervised=True, download=True, data_dir=DATA_DIR)
+
+# 📌 Carica il dataset direttamente da Hugging Face
+dataset = load_dataset("sradc/imagenet_resized_64x64", split="train")
+
+#Carica il dataset esterno da imagenet
+#imagenet = tfds.load("imagenet_resized/64x64", split="train", as_supervised=True, download=True, data_dir=DATA_DIR)
 
 image_list = []
 label_list = []