update

examples/sample_filter/bad_case_find.py

@@ -14,12 +14,13 @@ def get_args():
         "--data_dir",
         # default=r"E:\Users\tianx\HuggingDatasets\vm_sound_classification\data\data",
         # default=r"E:\Users\tianx\HuggingDatasets\vm_sound_classification\data\us-3",
-        default=r"E:\Users\tianx\HuggingDatasets\vm_sound_classification\data\transfer",
+        # default=r"E:\Users\tianx\HuggingDatasets\vm_sound_classification\data\transfer",
+        default=r"E:\Users\tianx\HuggingDatasets\vm_sound_classification\data\id",
         type=str
     )
     parser.add_argument(
         "--keep_dir",
-        default=r"E:\Users\tianx\HuggingDatasets\vm_sound_classification\data\keep",
+        default=r"E:\Users\tianx\HuggingDatasets\vm_sound_classification\data\keep2",
         type=str
     )
     parser.add_argument(
@@ -42,7 +43,7 @@ def main():
 
     client = Client("http://127.0.0.1:7864/")
 
-    for idx, filename in tqdm(enumerate(data_dir.glob("*.wav"))):
+    for idx, filename in tqdm(enumerate(data_dir.glob("**/*.wav"))):
         # if idx < 400:
         #     continue
         filename = filename.as_posix()
@@ -65,9 +66,9 @@ def main():
         )
         prob2 = float(prob2)
 
-        if label1 == "voicemail" and label2 in ("voicemail", "bell") and prob1 > 1.0:
+        if label1 == "voicemail" and label2 in ("voicemail", "bell") and prob1 > 0.6:
             pass
-        elif label1 == "non_voicemail" and label2 not in ("voicemail", "bell") and prob1 > 0.6:
+        elif label1 == "non_voicemail" and label2 not in ("voicemail", "bell") and prob1 > 0.0:
             pass
         else:
             print(f"label1: {label1}, prob1: {prob1}, label2: {label2}, prob2: {prob2}")

examples/sample_filter/wav_find_by_task_excel.py

@@ -13,40 +13,28 @@ from project_settings import project_path
 
 
 """
-default="task_DC_Death_Cases_20250220.xlsx",
-default="task_DC_Death_Cases_20250221.xlsx",
-default="task_DC_Death_Cases_EC_UC_20250220.xlsx",
-default="task_DC_Death_Cases_EC_UC_20250221.xlsx",
-default="task_DcTask_1_ID_LIVE_20250220_20250220-1.xlsx",
-default="task_DcTask_1_ID_LIVE_20250221_20250221-1.xlsx",
-default="task_DcTask_3_ID_LIVE_20250220_20250220-1.xlsx",
-default="task_DcTask_3_ID_LIVE_20250221_20250221-1.xlsx",
-default="task_DcTask_5_ID_LIVE_20250220_20250220-1.xlsx",
-default="task_DcTask_5_ID_LIVE_20250221_20250221-1.xlsx",
-default="task_DcTask_9_ID_LIVE_20250220_20250220-1.xlsx",
-default="task_DcTask_PTP_ID_LIVE_20250220_20250220-1.xlsx",
-default="task_DcTask_PTP_ID_LIVE_20250220_20250220-1 (1).xlsx",
-default="task_DcTask_PTP_ID_LIVE_20250221_20250221-1.xlsx",
+
+task_DcTask_5_ID_LIVE_20250224_20250224-1.xlsx
+task_DcTask_9_ID_LIVE_20250223_20250223-1.xlsx
+task_DcTask_9_ID_LIVE_20250224_20250224-1.xlsx
+task_DcTask_PTP_ID_LIVE_20250222_20250222-1.xlsx
+task_DcTask_PTP_ID_LIVE_20250224_20250224-1.xlsx
 """
 
+
 def get_args():
     parser = argparse.ArgumentParser()
     parser.add_argument(
         "--task_file",
-        # default="task_DC_Death_Cases_20250220.xlsx",
-        # default="task_DC_Death_Cases_20250221.xlsx",
-        # default="task_DC_Death_Cases_EC_UC_20250220.xlsx",
-        # default="task_DC_Death_Cases_EC_UC_20250221.xlsx",
-        # default="task_DcTask_1_ID_LIVE_20250220_20250220-1.xlsx",
-        # default="task_DcTask_1_ID_LIVE_20250221_20250221-1.xlsx",
-        # default="task_DcTask_3_ID_LIVE_20250220_20250220-1.xlsx",
-        # default="task_DcTask_3_ID_LIVE_20250221_20250221-1.xlsx",
-        # default="task_DcTask_5_ID_LIVE_20250220_20250220-1.xlsx",
-        # default="task_DcTask_5_ID_LIVE_20250221_20250221-1.xlsx",
-        # default="task_DcTask_9_ID_LIVE_20250220_20250220-1.xlsx",
-        # default="task_DcTask_PTP_ID_LIVE_20250220_20250220-1.xlsx",
-        # default="task_DcTask_PTP_ID_LIVE_20250220_20250220-1 (1).xlsx",
-        default="task_DcTask_PTP_ID_LIVE_20250221_20250221-1.xlsx",
+        # default="task_DcTask_1_ID_LIVE_20250223_20250223-1.xlsx",
+        # default="task_DcTask_1_ID_LIVE_20250224_20250224-1.xlsx",
+        # default="task_DcTask_3_ID_LIVE_20250224_20250224-1.xlsx",
+        # default="task_DcTask_5_ID_LIVE_20250223_20250223-1.xlsx",
+        # default="task_DcTask_5_ID_LIVE_20250224_20250224-1.xlsx",
+        # default="task_DcTask_9_ID_LIVE_20250223_20250223-1.xlsx",
+        # default="task_DcTask_9_ID_LIVE_20250224_20250224-1.xlsx",
+        # default="task_DcTask_PTP_ID_LIVE_20250222_20250222-1.xlsx",
+        default="task_DcTask_PTP_ID_LIVE_20250224_20250224-1.xlsx",
         type=str
     )
     parser.add_argument(

examples/vm_sound_classification/run_batch.sh

@@ -21,10 +21,10 @@
 #--config_file "yaml/conv2d-classifier-4-ch4.yaml"
 
 
-sh run.sh --stage 0 --stop_stage 6 --system_version centos --file_folder_name file_dir --final_model_name sound-8-ch4 \
---filename_patterns "/data/tianxing/PycharmProjects/datasets/voicemail/*/wav_finished/*/*.wav" \
---label_plan 8 \
---config_file "yaml/conv2d-classifier-8-ch4.yaml"
+#sh run.sh --stage 0 --stop_stage 6 --system_version centos --file_folder_name file_dir --final_model_name sound-8-ch4 \
+#--filename_patterns "/data/tianxing/PycharmProjects/datasets/voicemail/*/wav_finished/*/*.wav" \
+#--label_plan 8 \
+#--config_file "yaml/conv2d-classifier-8-ch4.yaml"
 
 
 # sound ch8
@@ -99,23 +99,23 @@ sh run.sh --stage 0 --stop_stage 6 --system_version centos --file_folder_name fi
 #--config_file "yaml/conv2d-classifier-4-ch32.yaml"
 
 
-sh run.sh --stage 0 --stop_stage 6 --system_version centos --file_folder_name file_dir --final_model_name sound-8-ch32 \
---filename_patterns "/data/tianxing/PycharmProjects/datasets/voicemail/*/wav_finished/*/*.wav" \
---label_plan 8 \
---config_file "yaml/conv2d-classifier-8-ch32.yaml"
+#sh run.sh --stage 0 --stop_stage 6 --system_version centos --file_folder_name file_dir --final_model_name sound-8-ch32 \
+#--filename_patterns "/data/tianxing/PycharmProjects/datasets/voicemail/*/wav_finished/*/*.wav" \
+#--label_plan 8 \
+#--config_file "yaml/conv2d-classifier-8-ch32.yaml"
 
 
 # pretrained voicemail
 
-sh run.sh --stage 0 --stop_stage 6 --system_version centos --file_folder_name file_dir --final_model_name voicemail-2-ch4 \
---filename_patterns "/data/tianxing/PycharmProjects/datasets/voicemail/*/wav_finished/*/*.wav" \
---label_plan 2-voicemail \
---config_file "yaml/conv2d-classifier-2-ch4.yaml"
-
-sh run.sh --stage 0 --stop_stage 6 --system_version centos --file_folder_name file_dir --final_model_name voicemail-2-ch32 \
---filename_patterns "/data/tianxing/PycharmProjects/datasets/voicemail/*/wav_finished/*/*.wav" \
---label_plan 2-voicemail \
---config_file "yaml/conv2d-classifier-2-ch32.yaml"
+#sh run.sh --stage 0 --stop_stage 6 --system_version centos --file_folder_name file_dir --final_model_name voicemail-2-ch4 \
+#--filename_patterns "/data/tianxing/PycharmProjects/datasets/voicemail/*/wav_finished/*/*.wav" \
+#--label_plan 2-voicemail \
+#--config_file "yaml/conv2d-classifier-2-ch4.yaml"
+#
+#sh run.sh --stage 0 --stop_stage 6 --system_version centos --file_folder_name file_dir --final_model_name voicemail-2-ch32 \
+#--filename_patterns "/data/tianxing/PycharmProjects/datasets/voicemail/*/wav_finished/*/*.wav" \
+#--label_plan 2-voicemail \
+#--config_file "yaml/conv2d-classifier-2-ch32.yaml"
 
 
 # voicemail ch4
@@ -126,11 +126,11 @@ sh run.sh --stage 0 --stop_stage 6 --system_version centos --file_folder_name fi
 #--config_file "yaml/conv2d-classifier-2-ch4.yaml" \
 #--pretrained_model "/data/tianxing/PycharmProjects/vm_sound_classification/trained_models/voicemail-2-ch4.zip"
 
-sh run.sh --stage 0 --stop_stage 6 --system_version centos --file_folder_name file_dir --final_model_name voicemail-en-us-2-ch4 \
---filename_patterns "/data/tianxing/PycharmProjects/datasets/voicemail/en-US/wav_finished/*/*.wav" \
---label_plan 2-voicemail \
---config_file "yaml/conv2d-classifier-2-ch4.yaml" \
---pretrained_model "/data/tianxing/PycharmProjects/vm_sound_classification/trained_models/voicemail-2-ch4.zip"
+#sh run.sh --stage 0 --stop_stage 6 --system_version centos --file_folder_name file_dir --final_model_name voicemail-en-us-2-ch4 \
+#--filename_patterns "/data/tianxing/PycharmProjects/datasets/voicemail/en-US/wav_finished/*/*.wav" \
+#--label_plan 2-voicemail \
+#--config_file "yaml/conv2d-classifier-2-ch4.yaml" \
+#--pretrained_model "/data/tianxing/PycharmProjects/vm_sound_classification/trained_models/voicemail-2-ch4.zip"
 
 #sh run.sh --stage 0 --stop_stage 6 --system_version centos --file_folder_name file_dir --final_model_name voicemail-es-mx-2-ch4 \
 #--filename_patterns "/data/tianxing/PycharmProjects/datasets/voicemail/es-MX/wav_finished/*/*.wav" \
@@ -195,11 +195,11 @@ sh run.sh --stage 0 --stop_stage 6 --system_version centos --file_folder_name fi
 #--config_file "yaml/conv2d-classifier-2-ch32.yaml" \
 #--pretrained_model "/data/tianxing/PycharmProjects/vm_sound_classification/trained_models/voicemail-2-ch32.zip"
 
-sh run.sh --stage 0 --stop_stage 6 --system_version centos --file_folder_name file_dir --final_model_name voicemail-en-us-2-ch32 \
---filename_patterns "/data/tianxing/PycharmProjects/datasets/voicemail/en-US/wav_finished/*/*.wav" \
---label_plan 2-voicemail \
---config_file "yaml/conv2d-classifier-2-ch32.yaml" \
---pretrained_model "/data/tianxing/PycharmProjects/vm_sound_classification/trained_models/voicemail-2-ch32.zip"
+#sh run.sh --stage 0 --stop_stage 6 --system_version centos --file_folder_name file_dir --final_model_name voicemail-en-us-2-ch32 \
+#--filename_patterns "/data/tianxing/PycharmProjects/datasets/voicemail/en-US/wav_finished/*/*.wav" \
+#--label_plan 2-voicemail \
+#--config_file "yaml/conv2d-classifier-2-ch32.yaml" \
+#--pretrained_model "/data/tianxing/PycharmProjects/vm_sound_classification/trained_models/voicemail-2-ch32.zip"
 
 #sh run.sh --stage 0 --stop_stage 6 --system_version centos --file_folder_name file_dir --final_model_name voicemail-es-mx-2-ch32 \
 #--filename_patterns "/data/tianxing/PycharmProjects/datasets/voicemail/es-MX/wav_finished/*/*.wav" \