moe 8 exp no freeze

script.py

@@ -51,38 +51,38 @@ print('Define Model')
 # model_path = './checkpoints/RAWNET_ASVSPOOF_FOR_INTHEWILD_PURDUE.pth'
 # model.load_state_dict(torch.load(model_path, map_location=device))
 
-# LCNN MODEL
-model = LCNN(return_emb=False).to(device)
-# model_path = './checkpoints/LCNN_ASVSPOOF_FOR_INTHEWILD_PURDUE.pth'
-# model_path = './checkpoints/LCNN_ALL_DATA.pth'
-# model_path = './checkpoints/LCNN_ALL_DATA_AUG.pth'
-# model_path = './checkpoints/LCNN_ALL_DATA_TTS_AUG.pth'
-model_path = './checkpoints/LCNN_ALL_DATA_TTS_MOD.pth'
-model.load_state_dict(torch.load(model_path, map_location=device))
+# # LCNN MODEL
+# model = LCNN(return_emb=False).to(device)
+# # model_path = './checkpoints/LCNN_ASVSPOOF_FOR_INTHEWILD_PURDUE.pth'
+# # model_path = './checkpoints/LCNN_ALL_DATA.pth'
+# # model_path = './checkpoints/LCNN_ALL_DATA_AUG.pth'
+# # model_path = './checkpoints/LCNN_ALL_DATA_TTS_AUG.pth'
+# model_path = './checkpoints/LCNN_ALL_DATA_TTS_MOD.pth'
+# model.load_state_dict(torch.load(model_path, map_location=device))
 
-# # MOE MODEL
-# expert_1 = LCNN(return_emb=True).to(device)
-# expert_2 = LCNN(return_emb=True).to(device)
-# expert_3 = LCNN(return_emb=True).to(device)
-# expert_4 = LCNN(return_emb=True).to(device)
-# expert_5 = LCNN(return_emb=True).to(device)
-# expert_6 = LCNN(return_emb=True).to(device)
-#
-# # # model = UltimateMOE(experts=[expert_1, expert_2, expert_3, expert_4])
-# # # model_path = './checkpoints/MOE_ULTIMATE.pth'
-#
-# # model = MOE_attention(experts=[expert_1, expert_2, expert_3, expert_4, expert_5, expert_6], device=device)
-# # # model_path = './checkpoints/MOE_ATTENTION.pth'
-# # model_path = './checkpoints/MOE_TRANSF.pth'
-#
-# expert_7 = LCNN(return_emb=True).to(device)
-# expert_8 = LCNN(return_emb=True).to(device)
-# model = MOE_attention(experts=[expert_1, expert_2, expert_3, expert_4, expert_5, expert_6, expert_7, expert_8], device=device, freezing=True)
-# # model_path = './checkpoints/MOE_TRANSF_7EXP.pth'
-# # model_path = './checkpoints/MOE_TRANSF_7EXP_AUG.pth'
-# # model_path = './checkpoints/MOE_TRANSF_7EXP_AUG_NO_FREEZE.pth'
+# MOE MODEL
+expert_1 = LCNN(return_emb=True).to(device)
+expert_2 = LCNN(return_emb=True).to(device)
+expert_3 = LCNN(return_emb=True).to(device)
+expert_4 = LCNN(return_emb=True).to(device)
+expert_5 = LCNN(return_emb=True).to(device)
+expert_6 = LCNN(return_emb=True).to(device)
+
+# # model = UltimateMOE(experts=[expert_1, expert_2, expert_3, expert_4])
+# # model_path = './checkpoints/MOE_ULTIMATE.pth'
+
+# model = MOE_attention(experts=[expert_1, expert_2, expert_3, expert_4, expert_5, expert_6], device=device)
+# # model_path = './checkpoints/MOE_ATTENTION.pth'
+# model_path = './checkpoints/MOE_TRANSF.pth'
+
+expert_7 = LCNN(return_emb=True).to(device)
+expert_8 = LCNN(return_emb=True).to(device)
+model = MOE_attention(experts=[expert_1, expert_2, expert_3, expert_4, expert_5, expert_6, expert_7, expert_8], device=device, freezing=True)
+# model_path = './checkpoints/MOE_TRANSF_7EXP.pth'
+# model_path = './checkpoints/MOE_TRANSF_7EXP_AUG.pth'
+# model_path = './checkpoints/MOE_TRANSF_7EXP_AUG_NO_FREEZE.pth'
 # model_path = './checkpoints/MOE_TRANSF_8EXP_AUG.pth'
-# # model_path = './checkpoints/MOE_TRANSF_8EXP_AUG_NO_FREEZE.pth'
+model_path = './checkpoints/MOE_TRANSF_8EXP_AUG_NO_FREEZE.pth'
 
 
 model = (model).to(device)
@@ -96,18 +96,18 @@ print('Loaded Weights')
 # del model
 # model = Model().to(device)
 
-SAMPLING_RATE_CODES = {
-    8000: 2,
-    16000: 3,
-    22050: 5,
-    24000: 7,
-    32000: 11,
-    44100: 13,
-    48000: 17,
-    "other": 19
-}
-
-seen_frequencies = set()
+# SAMPLING_RATE_CODES = {
+#     8000: 2,
+#     16000: 3,
+#     22050: 5,
+#     24000: 7,
+#     32000: 11,
+#     44100: 13,
+#     48000: 17,
+#     "other": 19
+# }
+#
+# seen_frequencies = set()
 
 # iterate over the dataset
 out = []
@@ -151,17 +151,17 @@ for el in tqdm.tqdm(dataset_remote):
     # "id" and "pred" are required. "score" will not be used in scoring but we encourage you to include it. We'll use it for analysis of the results
 
     # RUNNING ON HUGGINGFACE
-    # total_time = time.time() - start_time
-
-    freq = sr if sr in SAMPLING_RATE_CODES else "other"
-
-    # Assegna total_time: codice se è la prima occorrenza, 0 altrimenti
-    if freq not in seen_frequencies:
-        total_time = SAMPLING_RATE_CODES[freq]
-        seen_frequencies.add(freq)
-    # else:
-    #     total_time = 0
-    total_time = 1
+    total_time = time.time() - start_time
+
+    # freq = sr if sr in SAMPLING_RATE_CODES else "other"
+    #
+    # # Assegna total_time: codice se è la prima occorrenza, 0 altrimenti
+    # if freq not in seen_frequencies:
+    #     total_time = SAMPLING_RATE_CODES[freq]
+    #     seen_frequencies.add(freq)
+    # # else:
+    # #     total_time = 0
+    # total_time = 1
 
     out.append(dict(id=el["id"], pred=pred, score=score, time=total_time))
     # # RUNNING LOCALLY