reorganized and hopefully fixed torch.cat

app.py

@@ -28,11 +28,18 @@ with st.spinner("Loading Model"):
     valid_subs.insert(0, None)
 
 random_image = get_rand_img(sample_images)
+rand_idx = 0
 
 st.sidebar.title("Select a sample image")
+
+if st.sidebar.button("Random Sample Image"):
+    rand_idx, random_image = get_rand_img(sample_images)
+    sample_image = None
+    
 sample_image = st.sidebar.selectbox(
     "",
-    sample_images
+    sample_images,
+    index=rand_idx
 )
 
 st.sidebar.title("Select a Subreddit")
@@ -46,10 +53,6 @@ cap_prompt = st.sidebar.text_input(
     "Leave this blank for an unbiased caption", 
     value=""
 )
-
-if st.sidebar.button("Random Sample Image"):
-    random_image = get_rand_img(sample_images)
-    sample_image = None
     
     
 uploaded_image = None

model.py

@@ -64,7 +64,7 @@ class VirTexModel():
             
         if prompt is not "":
             cap_tokens = self.tokenizer.encode(prompt) 
-            subreddit_tokens = torch.cat((subreddit_tokens, cap_tokens))
+            subreddit_tokens = torch.cat([subreddit_tokens, cap_tokens])
             
         predictions: List[Dict[str, Any]] = []
         
@@ -105,5 +105,6 @@ def get_samples():
     return glob.glob(SAMPLES_PATH)
 
 def get_rand_img(samples):
-    return samples[random.randint(0,len(samples)-1)]
+    i = random.randint(0,len(samples)-1)
+    return i, samples[i]