douwantech/gpt-sovits-train 🖼️📝 → ❓

▶️ 209 runs 📅 Jun 2024 ⚙️ Cog 0.16.9 🔗 GitHub ⚖️ License
audio-model-training gpt-sovits model-training text-to-speech tts-training voice-cloning voice-model-training
About

“Train and fine-tune GPT-SoVITS voice models from custom datasets for high-quality voice cloning and TTS.”
Example Output

Output

{"zip_url":"https://replicate.delivery/pbxt/ga1uOrFf6fikm0dCby3ywnEMOI5MSIWiiaBUOTC2U6XgWRATA/351b84a7-8622-4c51-a803-ae848d62d158.zip","audio_url":"https://replicate.delivery/pbxt/FXc7LfgsbdWpekCGC3gwAemvhvBMStSI1zbh61TUpG0ItiAmA/origin.mp3"}
Performance Metrics

224.42s Prediction Time
378.97s Total Time
Input Parameters
audio_or_video_url (required) Type: string: Train audio URL or video URL
aliyun_oss_configure Type: stringDefault:: If need upload to aliyun oss directly set this configure. { "access_key_id": "your_access_key_id", "access_key_secret": "your_access_key_secret", "bucket_name": "your_bucket_name", "endpoint": "your_endpoint", "domain": "your_domain" }
Output Schema
zip_url Type: stringFormat: uri: Zip Url
audio_url Type: stringFormat: uri: Audio Url
oss_zip_url Type: string: Oss Zip Url
Example Execution Logs
Copied file to input/351b84a7-8622-4c51-a803-ae848d62d158/origin.mp3
执行完毕，请检查输出文件
2024-06-19 22:59:23,923 - modelscope - INFO - PyTorch version 2.0.1+cu118 Found.
2024-06-19 22:59:23,924 - modelscope - INFO - Loading ast index from /root/.cache/modelscope/ast_indexer
2024-06-19 22:59:23,924 - modelscope - INFO - No valid ast index found from /root/.cache/modelscope/ast_indexer, generating ast index from prebuilt!
2024-06-19 22:59:23,996 - modelscope - INFO - Loading done! Current index file version is 1.10.0, with md5 b51edac0938f2bdbb763b5eb2eac94ee and a total number of 946 components indexed
2024-06-19 22:59:29,620 - modelscope - WARNING - Model revision not specified, use revision: v1.0.2
Downloading:   0%|          | 0.00/1.45k [00:00<?, ?B/s]
Downloading: 100%|██████████| 1.45k/1.45k [00:00<00:00, 144kB/s]
Downloading:   0%|          | 0.00/903 [00:00<?, ?B/s]
Downloading: 100%|██████████| 903/903 [00:00<00:00, 108kB/s]
Downloading:   0%|          | 0.00/177k [00:00<?, ?B/s]
Downloading: 100%|██████████| 177k/177k [00:00<00:00, 1.16MB/s]
Downloading: 100%|██████████| 177k/177k [00:00<00:00, 1.16MB/s]
Downloading:   0%|          | 0.00/88.2k [00:00<?, ?B/s]
Downloading: 100%|██████████| 88.2k/88.2k [00:00<00:00, 873kB/s]
Downloading: 100%|██████████| 88.2k/88.2k [00:00<00:00, 870kB/s]
Downloading:   0%|          | 0.00/55.3M [00:00<?, ?B/s]
Downloading:  29%|██▉       | 16.0M/55.3M [00:00<00:01, 29.8MB/s]
Downloading:  87%|████████▋ | 48.0M/55.3M [00:00<00:00, 84.0MB/s]
Downloading: 100%|██████████| 55.3M/55.3M [00:00<00:00, 78.4MB/s]
Downloading:   0%|          | 0.00/12.8k [00:00<?, ?B/s]
Downloading: 100%|██████████| 12.8k/12.8k [00:00<00:00, 8.86MB/s]
Downloading:   0%|          | 0.00/75.0k [00:00<?, ?B/s]
Downloading: 100%|██████████| 75.0k/75.0k [00:00<00:00, 738kB/s]
Downloading: 100%|██████████| 75.0k/75.0k [00:00<00:00, 736kB/s]
Downloading:   0%|          | 0.00/152k [00:00<?, ?B/s]
Downloading: 100%|██████████| 152k/152k [00:00<00:00, 991kB/s]
Downloading: 100%|██████████| 152k/152k [00:00<00:00, 989kB/s]
2024-06-19 22:59:40,492 - modelscope - INFO - initiate model from /root/.cache/modelscope/hub/damo/speech_frcrn_ans_cirm_16k
2024-06-19 22:59:40,492 - modelscope - INFO - initiate model from location /root/.cache/modelscope/hub/damo/speech_frcrn_ans_cirm_16k.
2024-06-19 22:59:40,494 - modelscope - INFO - initialize model from /root/.cache/modelscope/hub/damo/speech_frcrn_ans_cirm_16k
2024-06-19 22:59:41,013 - modelscope - WARNING - No preprocessor field found in cfg.
2024-06-19 22:59:41,013 - modelscope - WARNING - No val key and type key found in preprocessor domain of configuration.json file.
2024-06-19 22:59:41,013 - modelscope - WARNING - Cannot find available config to build preprocessor at mode inference, current config: {'model_dir': '/root/.cache/modelscope/hub/damo/speech_frcrn_ans_cirm_16k'}. trying to build by task and model information.
2024-06-19 22:59:41,013 - modelscope - WARNING - No preprocessor key ('speech_frcrn_ans_cirm_16k', 'acoustic-noise-suppression') found in PREPROCESSOR_MAP, skip building preprocessor.
0%|          | 0/1 [00:00<?, ?it/s]inputs:(1, 254096)
padding: 26096
inputs after padding:(1, 280192)
100%|██████████| 1/1 [00:01<00:00,  1.37s/it]
100%|██████████| 1/1 [00:01<00:00,  1.37s/it]
Please install rotary_embedding_torch by:
pip install -U rotary_embedding_torch
Please install rotary_embedding_torch by:
pip install -U rotary_embedding_torch
Please install rotary_embedding_torch by:
pip install -U rotary_embedding_torch
Please install rotary_embedding_torch by:
pip install -U rotary_embedding_torch
tables:
-----------    ** dataset_classes **    --------------
| class name   | class location                               |
| AudioDataset | funasr/datasets/audio_datasets/datasets.py:7 |
-----------    ** index_ds_classes **    --------------
| class name   | class location                               |
| IndexDSJsonl | funasr/datasets/audio_datasets/index_ds.py:9 |
-----------    ** batch_sampler_classes **    --------------
| class name   | class location                               |
| BatchSampler | funasr/datasets/audio_datasets/samplers.py:7 |
-----------    ** frontend_classes **    --------------
| class name        | class location                       |
| WavFrontend       | funasr/frontends/wav_frontend.py:78  |
| WavFrontendOnline | funasr/frontends/wav_frontend.py:216 |
-----------    ** encoder_classes **    --------------
| class name            | class location                                        |
| BranchformerEncoder   | funasr/models/branchformer/encoder.py:294             |
| ConformerChunkEncoder | funasr/models/bat/conformer_chunk_encoder.py:315      |
| ConformerEncoder      | funasr/models/conformer/encoder.py:286                |
| DFSMN                 | funasr/models/fsmn_vad_streaming/encoder.py:232       |
| EBranchformerEncoder  | funasr/models/e_branchformer/encoder.py:177           |
| FSMN                  | funasr/models/fsmn_vad_streaming/encoder.py:161       |
| SANMEncoder           | funasr/models/sanm/encoder.py:161                     |
| SANMEncoderChunkOpt   | funasr/models/scama/encoder.py:162                    |
| SANMVadEncoder        | funasr/models/ct_transformer_streaming/encoder.py:148 |
| TransformerEncoder    | funasr/models/transformer/encoder.py:139              |
-----------    ** predictor_classes **    --------------
| class name     | class location                                     |
| CifPredictor   | funasr/models/paraformer/cif_predictor.py:15       |
| CifPredictorV2 | funasr/models/paraformer/cif_predictor.py:141      |
| CifPredictorV3 | funasr/models/bicif_paraformer/cif_predictor.py:95 |
-----------    ** model_classes **    --------------
| class name             | class location                                     |
| BiCifParaformer        | funasr/models/bicif_paraformer/model.py:37         |
| Branchformer           | funasr/models/branchformer/model.py:6              |
| CAMPPlus               | funasr/models/campplus/model.py:30                 |
| CTTransformer          | funasr/models/ct_transformer/model.py:30           |
| CTTransformerStreaming | funasr/models/ct_transformer_streaming/model.py:27 |
| Conformer              | funasr/models/conformer/model.py:8                 |
| ContextualParaformer   | funasr/models/contextual_paraformer/model.py:43    |
| EBranchformer          | funasr/models/e_branchformer/model.py:6            |
| Emotion2vec            | funasr/models/emotion2vec/model.py:34              |
| FsmnVADStreaming       | funasr/models/fsmn_vad_streaming/model.py:267      |
| MonotonicAligner       | funasr/models/monotonic_aligner/model.py:24        |
| Paraformer             | funasr/models/paraformer/model.py:26               |
| ParaformerStreaming    | funasr/models/paraformer_streaming/model.py:37     |
| SANM                   | funasr/models/sanm/model.py:13                     |
| SCAMA                  | funasr/models/scama/model.py:38                    |
| SeacoParaformer        | funasr/models/seaco_paraformer/model.py:45         |
| Transformer            | funasr/models/transformer/model.py:20              |
| UniASR                 | funasr/models/uniasr/model.py:26                   |
-----------    ** decoder_classes **    --------------
| class name                                 | class location                                     |
| ContextualParaformerDecoder                | funasr/models/contextual_paraformer/decoder.py:103 |
| DynamicConvolution2DTransformerDecoder     | funasr/models/transformer/decoder.py:588           |
| DynamicConvolutionTransformerDecoder       | funasr/models/transformer/decoder.py:527           |
| FsmnDecoder                                | funasr/models/sanm/decoder.py:198                  |
| FsmnDecoderSCAMAOpt                        | funasr/models/scama/decoder.py:197                 |
| LightweightConvolution2DTransformerDecoder | funasr/models/transformer/decoder.py:465           |
| LightweightConvolutionTransformerDecoder   | funasr/models/transformer/decoder.py:404           |
| ParaformerSANDecoder                       | funasr/models/paraformer/decoder.py:529            |
| ParaformerSANMDecoder                      | funasr/models/paraformer/decoder.py:204            |
| TransformerDecoder                         | funasr/models/transformer/decoder.py:355           |
-----------    ** normalize_classes **    --------------
| class name   | class location                             |
| GlobalMVN    | funasr/models/normalize/global_mvn.py:11   |
| UtteranceMVN | funasr/models/normalize/utterance_mvn.py:8 |
-----------    ** specaug_classes **    --------------
| class name | class location                       |
| SpecAug    | funasr/models/specaug/specaug.py:14  |
| SpecAugLFR | funasr/models/specaug/specaug.py:104 |
-----------    ** tokenizer_classes **    --------------
| class name    | class location                        |
| CharTokenizer | funasr/tokenizer/char_tokenizer.py:10 |
2024-06-19 22:59:47,351 - modelscope - INFO - PyTorch version 2.0.1+cu118 Found.
2024-06-19 22:59:47,352 - modelscope - INFO - Loading ast index from /root/.cache/modelscope/ast_indexer
2024-06-19 22:59:47,398 - modelscope - INFO - Loading done! Current index file version is 1.10.0, with md5 b51edac0938f2bdbb763b5eb2eac94ee and a total number of 946 components indexed
2024-06-19 22:59:48,985 - modelscope - INFO - Use user-specified model revision: v2.0.4
Downloading:   0%|          | 0.00/10.9k [00:00<?, ?B/s]
Downloading: 100%|██████████| 10.9k/10.9k [00:00<00:00, 1.11MB/s]
Downloading:   0%|          | 0.00/173k [00:00<?, ?B/s]
Downloading: 100%|██████████| 173k/173k [00:00<00:00, 1.13MB/s]
Downloading: 100%|██████████| 173k/173k [00:00<00:00, 1.13MB/s]
Downloading:   0%|          | 0.00/2.45k [00:00<?, ?B/s]
Downloading: 100%|██████████| 2.45k/2.45k [00:00<00:00, 2.00MB/s]
Downloading:   0%|          | 0.00/472 [00:00<?, ?B/s]
Downloading: 100%|██████████| 472/472 [00:00<00:00, 254kB/s]
Downloading:   0%|          | 0.00/840M [00:00<?, ?B/s]
Downloading:  19%|█▉        | 160M/840M [00:03<00:16, 43.9MB/s]
Downloading:  38%|███▊      | 320M/840M [00:04<00:05, 96.7MB/s]
Downloading:  57%|█████▋    | 480M/840M [00:04<00:02, 137MB/s]
Downloading:  76%|███████▌  | 640M/840M [00:05<00:01, 194MB/s]
Downloading:  81%|████████  | 680M/840M [00:06<00:01, 100MB/s]
Downloading: 100%|█████████▉| 840M/840M [00:07<00:00, 153MB/s]
Downloading: 100%|█████████▉| 840M/840M [00:07<00:00, 123MB/s]
Downloading:   0%|          | 0.00/19.1k [00:00<?, ?B/s]
Downloading: 100%|██████████| 19.1k/19.1k [00:00<00:00, 381kB/s]
Downloading:   0%|          | 0.00/7.90M [00:00<?, ?B/s]
Downloading: 100%|██████████| 7.90M/7.90M [00:00<00:00, 17.2MB/s]
Downloading: 100%|██████████| 7.90M/7.90M [00:00<00:00, 17.0MB/s]
Downloading:   0%|          | 0.00/48.7k [00:00<?, ?B/s]
Downloading: 100%|██████████| 48.7k/48.7k [00:00<00:00, 481kB/s]
Downloading: 100%|██████████| 48.7k/48.7k [00:00<00:00, 479kB/s]
Downloading:   0%|          | 0.00/91.5k [00:00<?, ?B/s]
Downloading: 100%|██████████| 91.5k/91.5k [00:00<00:00, 903kB/s]
Downloading: 100%|██████████| 91.5k/91.5k [00:00<00:00, 901kB/s]
2024-06-19 23:00:08,435 - modelscope - INFO - Use user-specified model revision: v2.0.4
Downloading:   0%|          | 0.00/7.85k [00:00<?, ?B/s]
Downloading: 100%|██████████| 7.85k/7.85k [00:00<00:00, 6.33MB/s]
Downloading:   0%|          | 0.00/1.19k [00:00<?, ?B/s]
Downloading: 100%|██████████| 1.19k/1.19k [00:00<00:00, 978kB/s]
Downloading:   0%|          | 0.00/365 [00:00<?, ?B/s]
Downloading: 100%|██████████| 365/365 [00:00<00:00, 315kB/s]
Downloading:   0%|          | 0.00/1.64M [00:00<?, ?B/s]
Downloading: 100%|██████████| 1.64M/1.64M [00:00<00:00, 5.44MB/s]
Downloading: 100%|██████████| 1.64M/1.64M [00:00<00:00, 5.42MB/s]
Downloading:   0%|          | 0.00/8.45k [00:00<?, ?B/s]
Downloading: 100%|██████████| 8.45k/8.45k [00:00<00:00, 5.63MB/s]
Downloading:   0%|          | 0.00/27.3k [00:00<?, ?B/s]
Downloading: 100%|██████████| 27.3k/27.3k [00:00<00:00, 543kB/s]
Downloading:   0%|          | 0.00/2.16M [00:00<?, ?B/s]
Downloading: 100%|██████████| 2.16M/2.16M [00:00<00:00, 6.16MB/s]
Downloading: 100%|██████████| 2.16M/2.16M [00:00<00:00, 6.13MB/s]
2024-06-19 23:00:16,137 - modelscope - INFO - Use user-specified model revision: v2.0.4
Downloading:   0%|          | 0.00/6.00k [00:00<?, ?B/s]
Downloading: 100%|██████████| 6.00k/6.00k [00:00<00:00, 4.54MB/s]
Downloading:   0%|          | 0.00/810 [00:00<?, ?B/s]
Downloading: 100%|██████████| 810/810 [00:00<00:00, 657kB/s]
Downloading:   0%|          | 0.00/373 [00:00<?, ?B/s]
Downloading: 100%|██████████| 373/373 [00:00<00:00, 314kB/s]
Downloading:   0%|          | 0.00/278M [00:00<?, ?B/s]
Downloading:   6%|▌         | 16.0M/278M [00:00<00:09, 28.3MB/s]
Downloading:  11%|█▏        | 32.0M/278M [00:00<00:06, 41.0MB/s]
Downloading:  17%|█▋        | 48.0M/278M [00:01<00:04, 50.7MB/s]
Downloading:  23%|██▎       | 64.0M/278M [00:01<00:04, 53.7MB/s]
Downloading:  29%|██▊       | 80.0M/278M [00:01<00:03, 58.6MB/s]
Downloading:  34%|███▍      | 96.0M/278M [00:01<00:03, 58.6MB/s]
Downloading:  40%|████      | 112M/278M [00:02<00:02, 62.1MB/s]
Downloading:  46%|████▌     | 128M/278M [00:02<00:02, 61.0MB/s]
Downloading:  52%|█████▏    | 144M/278M [00:02<00:02, 63.7MB/s]
Downloading:  57%|█████▋    | 160M/278M [00:02<00:01, 62.2MB/s]
Downloading:  63%|██████▎   | 176M/278M [00:03<00:01, 64.4MB/s]
Downloading:  69%|██████▉   | 192M/278M [00:03<00:01, 62.8MB/s]
Downloading:  75%|███████▍  | 208M/278M [00:03<00:01, 64.8MB/s]
Downloading:  80%|████████  | 224M/278M [00:04<00:00, 63.1MB/s]
Downloading:  86%|████████▌ | 240M/278M [00:04<00:00, 65.1MB/s]
Downloading:  92%|█████████▏| 256M/278M [00:04<00:00, 63.7MB/s]
Downloading:  98%|█████████▊| 272M/278M [00:04<00:00, 65.2MB/s]
Downloading: 100%|██████████| 278M/278M [00:04<00:00, 65.5MB/s]
Downloading: 100%|██████████| 278M/278M [00:04<00:00, 59.9MB/s]
Downloading:   0%|          | 0.00/863 [00:00<?, ?B/s]
Downloading: 100%|██████████| 863/863 [00:00<00:00, 546kB/s]
Downloading:   0%|          | 0.00/11.2k [00:00<?, ?B/s]
Downloading: 100%|██████████| 11.2k/11.2k [00:00<00:00, 7.83MB/s]
Downloading:   0%|          | 0.00/151k [00:00<?, ?B/s]
Downloading: 100%|██████████| 151k/151k [00:00<00:00, 987kB/s]
Downloading: 100%|██████████| 151k/151k [00:00<00:00, 984kB/s]
Downloading:   0%|          | 0.00/4.01M [00:00<?, ?B/s]
Downloading: 100%|██████████| 4.01M/4.01M [00:00<00:00, 9.96MB/s]
Downloading: 100%|██████████| 4.01M/4.01M [00:00<00:00, 9.90MB/s]
0%|          | 0/1 [00:00<?, ?it/s]
0%|[34m          [0m| 0/2 [00:00<?, ?it/s][A
50%|[34m█████     [0m| 1/2 [00:00<00:00,  7.64it/s][A
{'load_data': '0.001', 'extract_feat': '0.018', 'forward': '0.131', 'batch_size': '1', 'rtf': '0.008'}, :  50%|[34m█████     [0m| 1/2 [00:00<00:00,  7.64it/s][A
rtf_avg: 0.008: 100%|[34m██████████[0m| 2/2 [00:00<00:00,  7.64it/s]                                                                                          [A
rtf_avg: 0.008: 100%|[34m██████████[0m| 2/2 [00:00<00:00, 15.13it/s]
time cost vad: 0.132
0%|[31m          [0m| 0/2 [00:00<?, ?it/s][A
0%|[34m          [0m| 0/4 [00:00<?, ?it/s][A[A
rtf_avg_per_sample: 0.014, time_speech_total_per_sample:  15.881, time_escape_total_per_sample: 0.218:  75%|[34m███████▌  [0m| 3/4 [00:00<00:00, 13.76it/s][A[A
0%|[34m          [0m| 0/2 [00:00<?, ?it/s][A[A[A
{'load_data': 0.0, 'extract_feat': 0.0, 'forward': '0.027', 'batch_size': '1', 'rtf': '-0.027'}, :  50%|[34m█████     [0m| 1/2 [00:00<00:00, 37.19it/s][A[A[A
rtf_avg: -0.027: 100%|[34m██████████[0m| 2/2 [00:00<00:00, 74.01it/s]                                                                                  [A[A[A
rtf_avg: -0.027: 100%|[34m██████████[0m| 2/2 [00:00<00:00, 73.69it/s]
50%|[31m█████     [0m| 1/2 [00:00<00:00,  4.05it/s][A
rtf_avg_all_samples: 0.016, time_speech_total_all_samples:  15.881, time_escape_total_all_samples: 0.247: 100%|[31m██████████[0m| 2/2 [00:00<00:00,  4.05it/s][A
rtf_avg_all_samples: 0.016, time_speech_total_all_samples:  15.881, time_escape_total_all_samples: 0.247: 100%|[31m██████████[0m| 2/2 [00:00<00:00,  8.10it/s]
rtf_avg_per_sample: 0.014, time_speech_total_per_sample:  15.881, time_escape_total_per_sample: 0.218:  75%|[34m███████▌  [0m| 3/4 [00:00<00:00, 12.20it/s]
100%|██████████| 1/1 [00:00<00:00,  2.63it/s]
100%|██████████| 1/1 [00:00<00:00,  2.63it/s]
ASR 任务完成->标注文件路径: /src/output/351b84a7-8622-4c51-a803-ae848d62d158/asr_opt/denoise_opt.list
"/root/.pyenv/versions/3.9.19/bin/python" GPT_SoVITS/prepare_datasets/1-get-text.py
"/root/.pyenv/versions/3.9.19/bin/python" GPT_SoVITS/prepare_datasets/1-get-text.py
('进度：1a-ing', {'__type__': 'update', 'visible': False}, {'__type__': 'update', 'visible': True})
IMPORTANT: You are using gradio version 3.38.0, however version 4.29.0 is available, please upgrade.
--------
Building prefix dict from the default dictionary ...
Dumping model to file cache /src/TEMP/jieba.cache
Loading model cost 1.245 seconds.
Prefix dict has been built succesfully.
('进度：1a-done', {'__type__': 'update', 'visible': False}, {'__type__': 'update', 'visible': True})
"/root/.pyenv/versions/3.9.19/bin/python" GPT_SoVITS/prepare_datasets/2-get-hubert-wav32k.py
"/root/.pyenv/versions/3.9.19/bin/python" GPT_SoVITS/prepare_datasets/2-get-hubert-wav32k.py
('进度：1a-done, 1b-ing', {'__type__': 'update', 'visible': False}, {'__type__': 'update', 'visible': True})
('进度：1a1b-done', {'__type__': 'update', 'visible': False}, {'__type__': 'update', 'visible': True})
"/root/.pyenv/versions/3.9.19/bin/python" GPT_SoVITS/prepare_datasets/3-get-semantic.py
"/root/.pyenv/versions/3.9.19/bin/python" GPT_SoVITS/prepare_datasets/3-get-semantic.py
('进度：1a1b-done, 1cing', {'__type__': 'update', 'visible': False}, {'__type__': 'update', 'visible': True})
<All keys matched successfully>
<All keys matched successfully>
('进度：all-done', {'__type__': 'update', 'visible': False}, {'__type__': 'update', 'visible': True})
('一键三连进程结束', {'__type__': 'update', 'visible': True}, {'__type__': 'update', 'visible': False})
('SoVITS训练开始："/root/.pyenv/versions/3.9.19/bin/python" GPT_SoVITS/s2_train.py --config "/src/TEMP/tmp_s2.json"', {'__type__': 'update', 'visible': False}, {'__type__': 'update', 'visible': True})
"/root/.pyenv/versions/3.9.19/bin/python" GPT_SoVITS/s2_train.py --config "/src/TEMP/tmp_s2.json"
IMPORTANT: You are using gradio version 3.38.0, however version 4.29.0 is available, please upgrade.
--------
INFO:351b84a7-8622-4c51-a803-ae848d62d158:{'train': {'log_interval': 100, 'eval_interval': 500, 'seed': 1234, 'epochs': 8, 'learning_rate': 0.0001, 'betas': [0.8, 0.99], 'eps': 1e-09, 'batch_size': 11, 'fp16_run': True, 'lr_decay': 0.999875, 'segment_size': 20480, 'init_lr_ratio': 1, 'warmup_epochs': 0, 'c_mel': 45, 'c_kl': 1.0, 'text_low_lr_rate': 0.4, 'pretrained_s2G': 'GPT_SoVITS/pretrained_models/s2G488k.pth', 'pretrained_s2D': 'GPT_SoVITS/pretrained_models/s2D488k.pth', 'if_save_latest': True, 'if_save_every_weights': True, 'save_every_epoch': 4, 'gpu_numbers': '0-1'}, 'data': {'max_wav_value': 32768.0, 'sampling_rate': 32000, 'filter_length': 2048, 'hop_length': 640, 'win_length': 2048, 'n_mel_channels': 128, 'mel_fmin': 0.0, 'mel_fmax': None, 'add_blank': True, 'n_speakers': 300, 'cleaned_text': True, 'exp_dir': 'logs/351b84a7-8622-4c51-a803-ae848d62d158'}, 'model': {'inter_channels': 192, 'hidden_channels': 192, 'filter_channels': 768, 'n_heads': 2, 'n_layers': 6, 'kernel_size': 3, 'p_dropout': 0.1, 'resblock': '1', 'resblock_kernel_sizes': [3, 7, 11], 'resblock_dilation_sizes': [[1, 3, 5], [1, 3, 5], [1, 3, 5]], 'upsample_rates': [10, 8, 2, 2, 2], 'upsample_initial_channel': 512, 'upsample_kernel_sizes': [16, 16, 8, 2, 2], 'n_layers_q': 3, 'use_spectral_norm': False, 'gin_channels': 512, 'semantic_frame_rate': '25hz', 'freeze_quantizer': True}, 's2_ckpt_dir': 'logs/351b84a7-8622-4c51-a803-ae848d62d158', 'content_module': 'cnhubert', 'save_weight_dir': 'SoVITS_weights', 'name': '351b84a7-8622-4c51-a803-ae848d62d158', 'pretrain': None, 'resume_step': None}
INFO:torch.distributed.distributed_c10d:Added key: store_based_barrier_key:1 to store for rank: 0
INFO:torch.distributed.distributed_c10d:Rank 0: Completed store-based barrier for key:store_based_barrier_key:1 with 1 nodes.
phoneme_data_len: 1
wav_data_len: 100
0%|          | 0/100 [00:00<?, ?it/s]
100%|██████████| 100/100 [00:00<00:00, 73921.47it/s]
skipped_phone:  0 , skipped_dur:  0
total left:  100
INFO:351b84a7-8622-4c51-a803-ae848d62d158:loaded pretrained GPT_SoVITS/pretrained_models/s2G488k.pth
<All keys matched successfully>
INFO:351b84a7-8622-4c51-a803-ae848d62d158:loaded pretrained GPT_SoVITS/pretrained_models/s2D488k.pth
<All keys matched successfully>
/root/.pyenv/versions/3.9.19/lib/python3.9/site-packages/torch/optim/lr_scheduler.py:139: UserWarning: Detected call of `lr_scheduler.step()` before `optimizer.step()`. In PyTorch 1.1.0 and later, you should call them in the opposite order: `optimizer.step()` before `lr_scheduler.step()`.  Failure to do this will result in PyTorch skipping the first value of the learning rate schedule. See more details at https://pytorch.org/docs/stable/optim.html#how-to-adjust-learning-rate
warnings.warn("Detected call of `lr_scheduler.step()` before `optimizer.step()`. "
0%|          | 0/10 [00:00<?, ?it/s]/root/.pyenv/versions/3.9.19/lib/python3.9/site-packages/torch/functional.py:641: UserWarning: stft with return_complex=False is deprecated. In a future pytorch release, stft will return complex tensors for all inputs, and return_complex=False will raise an error.
Note: you can still call torch.view_as_real on the complex output to recover the old return format. (Triggered internally at ../aten/src/ATen/native/SpectralOps.cpp:862.)
return _VF.stft(input, n_fft, hop_length, win_length, window,  # type: ignore[attr-defined]
/root/.pyenv/versions/3.9.19/lib/python3.9/site-packages/torch/functional.py:641: UserWarning: stft with return_complex=False is deprecated. In a future pytorch release, stft will return complex tensors for all inputs, and return_complex=False will raise an error.
Note: you can still call torch.view_as_real on the complex output to recover the old return format. (Triggered internally at ../aten/src/ATen/native/SpectralOps.cpp:862.)
return _VF.stft(input, n_fft, hop_length, win_length, window,  # type: ignore[attr-defined]
/root/.pyenv/versions/3.9.19/lib/python3.9/site-packages/torch/functional.py:641: UserWarning: stft with return_complex=False is deprecated. In a future pytorch release, stft will return complex tensors for all inputs, and return_complex=False will raise an error.
Note: you can still call torch.view_as_real on the complex output to recover the old return format. (Triggered internally at ../aten/src/ATen/native/SpectralOps.cpp:862.)
return _VF.stft(input, n_fft, hop_length, win_length, window,  # type: ignore[attr-defined]
/root/.pyenv/versions/3.9.19/lib/python3.9/site-packages/torch/functional.py:641: UserWarning: stft with return_complex=False is deprecated. In a future pytorch release, stft will return complex tensors for all inputs, and return_complex=False will raise an error.
Note: you can still call torch.view_as_real on the complex output to recover the old return format. (Triggered internally at ../aten/src/ATen/native/SpectralOps.cpp:862.)
return _VF.stft(input, n_fft, hop_length, win_length, window,  # type: ignore[attr-defined]
/root/.pyenv/versions/3.9.19/lib/python3.9/site-packages/torch/functional.py:641: UserWarning: stft with return_complex=False is deprecated. In a future pytorch release, stft will return complex tensors for all inputs, and return_complex=False will raise an error.
Note: you can still call torch.view_as_real on the complex output to recover the old return format. (Triggered internally at ../aten/src/ATen/native/SpectralOps.cpp:862.)
return _VF.stft(input, n_fft, hop_length, win_length, window,  # type: ignore[attr-defined]
/root/.pyenv/versions/3.9.19/lib/python3.9/site-packages/torch/functional.py:641: UserWarning: stft with return_complex=False is deprecated. In a future pytorch release, stft will return complex tensors for all inputs, and return_complex=False will raise an error.
Note: you can still call torch.view_as_real on the complex output to recover the old return format. (Triggered internally at ../aten/src/ATen/native/SpectralOps.cpp:862.)
return _VF.stft(input, n_fft, hop_length, win_length, window,  # type: ignore[attr-defined]
/root/.pyenv/versions/3.9.19/lib/python3.9/site-packages/torch/functional.py:641: UserWarning: stft with return_complex=False is deprecated. In a future pytorch release, stft will return complex tensors for all inputs, and return_complex=False will raise an error.
Note: you can still call torch.view_as_real on the complex output to recover the old return format. (Triggered internally at ../aten/src/ATen/native/SpectralOps.cpp:862.)
return _VF.stft(input, n_fft, hop_length, win_length, window,  # type: ignore[attr-defined]
/root/.pyenv/versions/3.9.19/lib/python3.9/site-packages/torch/functional.py:641: UserWarning: ComplexHalf support is experimental and many operators don't support it yet. (Triggered internally at ../aten/src/ATen/EmptyTensor.cpp:31.)
return _VF.stft(input, n_fft, hop_length, win_length, window,  # type: ignore[attr-defined]
[W reducer.cpp:1300] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters in the forward pass. This flag results in an extra traversal of the autograd graph every iteration,  which can adversely affect performance. If your model indeed never has any unused parameters in the forward pass, consider turning this flag off. Note that this warning may be a false positive if your model has flow control causing later iterations to have unused parameters. (function operator())
/root/.pyenv/versions/3.9.19/lib/python3.9/site-packages/torch/autograd/__init__.py:200: UserWarning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed.  This is not an error, but may impair performance.
grad.sizes() = [1, 9, 96], strides() = [152736, 96, 1]
bucket_view.sizes() = [1, 9, 96], strides() = [864, 96, 1] (Triggered internally at ../torch/csrc/distributed/c10d/reducer.cpp:323.)
Variable._execution_engine.run_backward(  # Calls into the C++ engine to run the backward pass
INFO:351b84a7-8622-4c51-a803-ae848d62d158:Train Epoch: 1 [0%]
INFO:351b84a7-8622-4c51-a803-ae848d62d158:[2.389455556869507, 2.1822030544281006, 5.724943161010742, 21.071016311645508, 0.0, 2.827393054962158, 0, 9.99875e-05]
10%|█         | 1/10 [00:09<01:27,  9.74s/it]
20%|██        | 2/10 [00:10<00:34,  4.25s/it]
30%|███       | 3/10 [00:10<00:17,  2.48s/it]
40%|████      | 4/10 [00:10<00:09,  1.65s/it]
50%|█████     | 5/10 [00:11<00:05,  1.20s/it]
60%|██████    | 6/10 [00:11<00:03,  1.09it/s]
70%|███████   | 7/10 [00:12<00:02,  1.35it/s]
80%|████████  | 8/10 [00:12<00:01,  1.55it/s]
90%|█████████ | 9/10 [00:12<00:00,  1.77it/s]
100%|██████████| 10/10 [00:13<00:00,  1.97it/s]
100%|██████████| 10/10 [00:13<00:00,  1.33s/it]
INFO:351b84a7-8622-4c51-a803-ae848d62d158:====> Epoch: 1
0%|          | 0/10 [00:00<?, ?it/s]
10%|█         | 1/10 [00:02<00:20,  2.29s/it]
20%|██        | 2/10 [00:02<00:09,  1.20s/it]
30%|███       | 3/10 [00:03<00:05,  1.18it/s]
40%|████      | 4/10 [00:03<00:04,  1.48it/s]
50%|█████     | 5/10 [00:03<00:02,  1.73it/s]
60%|██████    | 6/10 [00:04<00:02,  1.93it/s]
70%|███████   | 7/10 [00:04<00:01,  2.08it/s]
80%|████████  | 8/10 [00:05<00:00,  2.19it/s]
90%|█████████ | 9/10 [00:05<00:00,  2.27it/s]
100%|██████████| 10/10 [00:05<00:00,  2.33it/s]
100%|██████████| 10/10 [00:05<00:00,  1.67it/s]
INFO:351b84a7-8622-4c51-a803-ae848d62d158:====> Epoch: 2
0%|          | 0/10 [00:00<?, ?it/s]
10%|█         | 1/10 [00:02<00:21,  2.34s/it]
20%|██        | 2/10 [00:02<00:09,  1.23s/it]
30%|███       | 3/10 [00:03<00:06,  1.14it/s]
40%|████      | 4/10 [00:03<00:04,  1.45it/s]
50%|█████     | 5/10 [00:04<00:02,  1.71it/s]
60%|██████    | 6/10 [00:04<00:02,  1.90it/s]
70%|███████   | 7/10 [00:04<00:01,  2.05it/s]
80%|████████  | 8/10 [00:05<00:00,  2.17it/s]
90%|█████████ | 9/10 [00:05<00:00,  2.25it/s]
100%|██████████| 10/10 [00:06<00:00,  2.32it/s]
100%|██████████| 10/10 [00:06<00:00,  1.64it/s]
INFO:351b84a7-8622-4c51-a803-ae848d62d158:====> Epoch: 3
0%|          | 0/10 [00:00<?, ?it/s]
10%|█         | 1/10 [00:02<00:19,  2.21s/it]
20%|██        | 2/10 [00:02<00:09,  1.17s/it]
30%|███       | 3/10 [00:03<00:05,  1.18it/s]
40%|████      | 4/10 [00:03<00:04,  1.47it/s]
50%|█████     | 5/10 [00:03<00:02,  1.73it/s]
60%|██████    | 6/10 [00:04<00:02,  1.93it/s]
70%|███████   | 7/10 [00:04<00:01,  2.08it/s]
80%|████████  | 8/10 [00:05<00:00,  2.19it/s]
90%|█████████ | 9/10 [00:05<00:00,  2.27it/s]
100%|██████████| 10/10 [00:05<00:00,  2.34it/s]
100%|██████████| 10/10 [00:05<00:00,  1.68it/s]
INFO:351b84a7-8622-4c51-a803-ae848d62d158:Saving model and optimizer state at iteration 4 to logs/351b84a7-8622-4c51-a803-ae848d62d158/logs_s2/G_233333333333.pth
INFO:351b84a7-8622-4c51-a803-ae848d62d158:Saving model and optimizer state at iteration 4 to logs/351b84a7-8622-4c51-a803-ae848d62d158/logs_s2/D_233333333333.pth
INFO:351b84a7-8622-4c51-a803-ae848d62d158:saving ckpt 351b84a7-8622-4c51-a803-ae848d62d158_e4:Success.
INFO:351b84a7-8622-4c51-a803-ae848d62d158:====> Epoch: 4
0%|          | 0/10 [00:00<?, ?it/s]
10%|█         | 1/10 [00:02<00:20,  2.33s/it]
20%|██        | 2/10 [00:02<00:09,  1.22s/it]
30%|███       | 3/10 [00:03<00:06,  1.16it/s]
40%|████      | 4/10 [00:03<00:04,  1.46it/s]
50%|█████     | 5/10 [00:04<00:02,  1.71it/s]
60%|██████    | 6/10 [00:04<00:02,  1.77it/s]
70%|███████   | 7/10 [00:04<00:01,  1.95it/s]
80%|████████  | 8/10 [00:05<00:00,  2.09it/s]
90%|█████████ | 9/10 [00:05<00:00,  2.20it/s]
100%|██████████| 10/10 [00:06<00:00,  2.27it/s]
100%|██████████| 10/10 [00:06<00:00,  1.62it/s]
INFO:351b84a7-8622-4c51-a803-ae848d62d158:====> Epoch: 5
0%|          | 0/10 [00:00<?, ?it/s]
10%|█         | 1/10 [00:02<00:21,  2.39s/it]
20%|██        | 2/10 [00:02<00:09,  1.24s/it]
30%|███       | 3/10 [00:03<00:06,  1.15it/s]
40%|████      | 4/10 [00:03<00:04,  1.46it/s]
50%|█████     | 5/10 [00:04<00:02,  1.71it/s]
60%|██████    | 6/10 [00:04<00:02,  1.91it/s]
70%|███████   | 7/10 [00:04<00:01,  2.06it/s]
80%|████████  | 8/10 [00:05<00:00,  2.18it/s]
90%|█████████ | 9/10 [00:05<00:00,  2.26it/s]
100%|██████████| 10/10 [00:06<00:00,  2.32it/s]
100%|██████████| 10/10 [00:06<00:00,  1.64it/s]
INFO:351b84a7-8622-4c51-a803-ae848d62d158:====> Epoch: 6
0%|          | 0/10 [00:00<?, ?it/s]
10%|█         | 1/10 [00:02<00:20,  2.32s/it]
20%|██        | 2/10 [00:02<00:09,  1.20s/it]
30%|███       | 3/10 [00:03<00:06,  1.16it/s]
40%|████      | 4/10 [00:03<00:04,  1.46it/s]
50%|█████     | 5/10 [00:04<00:02,  1.71it/s]
60%|██████    | 6/10 [00:04<00:02,  1.91it/s]
70%|███████   | 7/10 [00:04<00:01,  2.06it/s]
80%|████████  | 8/10 [00:05<00:00,  2.18it/s]
90%|█████████ | 9/10 [00:05<00:00,  2.26it/s]
100%|██████████| 10/10 [00:06<00:00,  2.32it/s]
100%|██████████| 10/10 [00:06<00:00,  1.66it/s]
INFO:351b84a7-8622-4c51-a803-ae848d62d158:====> Epoch: 7
0%|          | 0/10 [00:00<?, ?it/s]
10%|█         | 1/10 [00:02<00:21,  2.34s/it]
20%|██        | 2/10 [00:02<00:09,  1.22s/it]
30%|███       | 3/10 [00:03<00:06,  1.15it/s]
40%|████      | 4/10 [00:03<00:04,  1.45it/s]
50%|█████     | 5/10 [00:04<00:02,  1.70it/s]
60%|██████    | 6/10 [00:04<00:02,  1.90it/s]
70%|███████   | 7/10 [00:04<00:01,  2.04it/s]
80%|████████  | 8/10 [00:05<00:00,  2.16it/s]
90%|█████████ | 9/10 [00:05<00:00,  2.24it/s]
100%|██████████| 10/10 [00:06<00:00,  2.31it/s]
100%|██████████| 10/10 [00:06<00:00,  1.64it/s]
INFO:351b84a7-8622-4c51-a803-ae848d62d158:Saving model and optimizer state at iteration 8 to logs/351b84a7-8622-4c51-a803-ae848d62d158/logs_s2/G_233333333333.pth
INFO:351b84a7-8622-4c51-a803-ae848d62d158:Saving model and optimizer state at iteration 8 to logs/351b84a7-8622-4c51-a803-ae848d62d158/logs_s2/D_233333333333.pth
INFO:351b84a7-8622-4c51-a803-ae848d62d158:saving ckpt 351b84a7-8622-4c51-a803-ae848d62d158_e8:Success.
INFO:351b84a7-8622-4c51-a803-ae848d62d158:====> Epoch: 8
('SoVITS训练完成', {'__type__': 'update', 'visible': True}, {'__type__': 'update', 'visible': False})
('GPT训练开始："/root/.pyenv/versions/3.9.19/bin/python" GPT_SoVITS/s1_train.py --config_file "/src/TEMP/tmp_s1.yaml" ', {'__type__': 'update', 'visible': False}, {'__type__': 'update', 'visible': True})
"/root/.pyenv/versions/3.9.19/bin/python" GPT_SoVITS/s1_train.py --config_file "/src/TEMP/tmp_s1.yaml"
IMPORTANT: You are using gradio version 3.38.0, however version 4.29.0 is available, please upgrade.
--------
Seed set to 1234
Using 16bit Automatic Mixed Precision (AMP)
GPU available: True (cuda), used: True
TPU available: False, using: 0 TPU cores
HPU available: False, using: 0 HPUs
<All keys matched successfully>
ckpt_path: None
[rank: 0] Seed set to 1234
Initializing distributed: GLOBAL_RANK: 0, MEMBER: 1/1
----------------------------------------------------------------------------------------------------
distributed_backend=nccl
All distributed processes registered. Starting with 1 processes
----------------------------------------------------------------------------------------------------
Missing logger folder: logs/351b84a7-8622-4c51-a803-ae848d62d158/logs_s1/logs_s1
semantic_data_len: 1
phoneme_data_len: 1
item_name                                     semantic_audio
0  origin.mp3_0000000000_0000508480.wav  520 721 1005 578 283 919 290 96 142 545 8 17 7...
dataset.__len__(): 100
LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0,1]
| Name  | Type                 | Params | Mode
-------------------------------------------------------
0 | model | Text2SemanticDecoder | 77.5 M | train
-------------------------------------------------------
77.5 M    Trainable params
0         Non-trainable params
77.5 M    Total params
309.975   Total estimated model params size (MB)
/root/.pyenv/versions/3.9.19/lib/python3.9/site-packages/pytorch_lightning/loops/fit_loop.py:298: The number of training batches (10) is smaller than the logging interval Trainer(log_every_n_steps=50). Set a lower value for log_every_n_steps if you want to see logs for the training epoch.
Training: |          | 0/? [00:00<?, ?it/s]
Training:   0%|          | 0/10 [00:00<?, ?it/s]
Epoch 0:   0%|          | 0/10 [00:00<?, ?it/s]
Epoch 0:  10%|█         | 1/10 [00:00<00:03,  2.34it/s]
Epoch 0:  10%|█         | 1/10 [00:00<00:03,  2.34it/s, v_num=0, total_loss_step=2.2e+4, lr_step=1e-5, top_3_acc_step=0.132]
Epoch 0:  20%|██        | 2/10 [00:00<00:02,  3.33it/s, v_num=0, total_loss_step=2.2e+4, lr_step=1e-5, top_3_acc_step=0.132]
Epoch 0:  20%|██        | 2/10 [00:00<00:02,  3.33it/s, v_num=0, total_loss_step=2.19e+4, lr_step=1e-5, top_3_acc_step=0.126]
Epoch 0:  30%|███       | 3/10 [00:00<00:01,  3.91it/s, v_num=0, total_loss_step=2.19e+4, lr_step=1e-5, top_3_acc_step=0.126]
Epoch 0:  30%|███       | 3/10 [00:00<00:01,  3.91it/s, v_num=0, total_loss_step=2.19e+4, lr_step=1e-5, top_3_acc_step=0.130]
Epoch 0:  40%|████      | 4/10 [00:00<00:01,  4.29it/s, v_num=0, total_loss_step=2.19e+4, lr_step=1e-5, top_3_acc_step=0.130]
Epoch 0:  40%|████      | 4/10 [00:00<00:01,  4.29it/s, v_num=0, total_loss_step=2.2e+4, lr_step=1e-5, top_3_acc_step=0.131]
Epoch 0:  50%|█████     | 5/10 [00:01<00:01,  4.52it/s, v_num=0, total_loss_step=2.2e+4, lr_step=1e-5, top_3_acc_step=0.131]
Epoch 0:  50%|█████     | 5/10 [00:01<00:01,  4.51it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.010, top_3_acc_step=0.128]
Epoch 0:  60%|██████    | 6/10 [00:01<00:00,  4.70it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.010, top_3_acc_step=0.128]
Epoch 0:  60%|██████    | 6/10 [00:01<00:00,  4.70it/s, v_num=0, total_loss_step=2.2e+4, lr_step=0.010, top_3_acc_step=0.127]
Epoch 0:  70%|███████   | 7/10 [00:01<00:00,  4.85it/s, v_num=0, total_loss_step=2.2e+4, lr_step=0.010, top_3_acc_step=0.127]
Epoch 0:  70%|███████   | 7/10 [00:01<00:00,  4.85it/s, v_num=0, total_loss_step=21934.0, lr_step=0.010, top_3_acc_step=0.130]
Epoch 0:  80%|████████  | 8/10 [00:01<00:00,  4.96it/s, v_num=0, total_loss_step=21934.0, lr_step=0.010, top_3_acc_step=0.130]
Epoch 0:  80%|████████  | 8/10 [00:01<00:00,  4.96it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.010, top_3_acc_step=0.129]
Epoch 0:  90%|█████████ | 9/10 [00:01<00:00,  5.04it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.010, top_3_acc_step=0.129]
Epoch 0:  90%|█████████ | 9/10 [00:01<00:00,  5.04it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.131]
Epoch 0: 100%|██████████| 10/10 [00:01<00:00,  5.39it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.131]
Epoch 0: 100%|██████████| 10/10 [00:01<00:00,  5.39it/s, v_num=0, total_loss_step=1.99e+3, lr_step=0.002, top_3_acc_step=0.144]
Epoch 0: 100%|██████████| 10/10 [00:01<00:00,  5.38it/s, v_num=0, total_loss_step=1.99e+3, lr_step=0.002, top_3_acc_step=0.144, total_loss_epoch=2.17e+4, lr_epoch=0.00464, top_3_acc_epoch=0.129]
Epoch 0:   0%|          | 0/10 [00:00<?, ?it/s, v_num=0, total_loss_step=1.99e+3, lr_step=0.002, top_3_acc_step=0.144, total_loss_epoch=2.17e+4, lr_epoch=0.00464, top_3_acc_epoch=0.129]
Epoch 1:   0%|          | 0/10 [00:00<?, ?it/s, v_num=0, total_loss_step=1.99e+3, lr_step=0.002, top_3_acc_step=0.144, total_loss_epoch=2.17e+4, lr_epoch=0.00464, top_3_acc_epoch=0.129]
Epoch 1:  10%|█         | 1/10 [00:00<00:01,  5.41it/s, v_num=0, total_loss_step=1.99e+3, lr_step=0.002, top_3_acc_step=0.144, total_loss_epoch=2.17e+4, lr_epoch=0.00464, top_3_acc_epoch=0.129]
Epoch 1:  10%|█         | 1/10 [00:00<00:01,  5.39it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.129, total_loss_epoch=2.17e+4, lr_epoch=0.00464, top_3_acc_epoch=0.129]
Epoch 1:  20%|██        | 2/10 [00:00<00:01,  5.61it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.129, total_loss_epoch=2.17e+4, lr_epoch=0.00464, top_3_acc_epoch=0.129]
Epoch 1:  20%|██        | 2/10 [00:00<00:01,  5.60it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.124, total_loss_epoch=2.17e+4, lr_epoch=0.00464, top_3_acc_epoch=0.129]
Epoch 1:  30%|███       | 3/10 [00:00<00:01,  5.73it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.124, total_loss_epoch=2.17e+4, lr_epoch=0.00464, top_3_acc_epoch=0.129]
Epoch 1:  30%|███       | 3/10 [00:00<00:01,  5.73it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.134, total_loss_epoch=2.17e+4, lr_epoch=0.00464, top_3_acc_epoch=0.129]
Epoch 1:  40%|████      | 4/10 [00:00<00:01,  5.80it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.134, total_loss_epoch=2.17e+4, lr_epoch=0.00464, top_3_acc_epoch=0.129]
Epoch 1:  40%|████      | 4/10 [00:00<00:01,  5.79it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.130, total_loss_epoch=2.17e+4, lr_epoch=0.00464, top_3_acc_epoch=0.129]
Epoch 1:  50%|█████     | 5/10 [00:00<00:00,  5.78it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.130, total_loss_epoch=2.17e+4, lr_epoch=0.00464, top_3_acc_epoch=0.129]
Epoch 1:  50%|█████     | 5/10 [00:00<00:00,  5.77it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.129, total_loss_epoch=2.17e+4, lr_epoch=0.00464, top_3_acc_epoch=0.129]
Epoch 1:  60%|██████    | 6/10 [00:01<00:00,  5.81it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.129, total_loss_epoch=2.17e+4, lr_epoch=0.00464, top_3_acc_epoch=0.129]
Epoch 1:  60%|██████    | 6/10 [00:01<00:00,  5.81it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.128, total_loss_epoch=2.17e+4, lr_epoch=0.00464, top_3_acc_epoch=0.129]
Epoch 1:  70%|███████   | 7/10 [00:01<00:00,  5.84it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.128, total_loss_epoch=2.17e+4, lr_epoch=0.00464, top_3_acc_epoch=0.129]
Epoch 1:  70%|███████   | 7/10 [00:01<00:00,  5.84it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.129, total_loss_epoch=2.17e+4, lr_epoch=0.00464, top_3_acc_epoch=0.129]
Epoch 1:  80%|████████  | 8/10 [00:01<00:00,  5.86it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.129, total_loss_epoch=2.17e+4, lr_epoch=0.00464, top_3_acc_epoch=0.129]
Epoch 1:  80%|████████  | 8/10 [00:01<00:00,  5.86it/s, v_num=0, total_loss_step=2.2e+4, lr_step=0.002, top_3_acc_step=0.126, total_loss_epoch=2.17e+4, lr_epoch=0.00464, top_3_acc_epoch=0.129]
Epoch 1:  90%|█████████ | 9/10 [00:01<00:00,  5.85it/s, v_num=0, total_loss_step=2.2e+4, lr_step=0.002, top_3_acc_step=0.126, total_loss_epoch=2.17e+4, lr_epoch=0.00464, top_3_acc_epoch=0.129]
Epoch 1:  90%|█████████ | 9/10 [00:01<00:00,  5.85it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.130, total_loss_epoch=2.17e+4, lr_epoch=0.00464, top_3_acc_epoch=0.129]
Epoch 1: 100%|██████████| 10/10 [00:01<00:00,  6.31it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.130, total_loss_epoch=2.17e+4, lr_epoch=0.00464, top_3_acc_epoch=0.129]
Epoch 1: 100%|██████████| 10/10 [00:01<00:00,  6.30it/s, v_num=0, total_loss_step=1.98e+3, lr_step=0.002, top_3_acc_step=0.142, total_loss_epoch=2.17e+4, lr_epoch=0.00464, top_3_acc_epoch=0.129]
Epoch 1: 100%|██████████| 10/10 [00:01<00:00,  6.30it/s, v_num=0, total_loss_step=1.98e+3, lr_step=0.002, top_3_acc_step=0.142, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 1:   0%|          | 0/10 [00:00<?, ?it/s, v_num=0, total_loss_step=1.98e+3, lr_step=0.002, top_3_acc_step=0.142, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 2:   0%|          | 0/10 [00:00<?, ?it/s, v_num=0, total_loss_step=1.98e+3, lr_step=0.002, top_3_acc_step=0.142, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 2:  10%|█         | 1/10 [00:00<00:01,  5.31it/s, v_num=0, total_loss_step=1.98e+3, lr_step=0.002, top_3_acc_step=0.142, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 2:  10%|█         | 1/10 [00:00<00:01,  5.30it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.126, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 2:  20%|██        | 2/10 [00:00<00:01,  5.58it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.126, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 2:  20%|██        | 2/10 [00:00<00:01,  5.57it/s, v_num=0, total_loss_step=2.2e+4, lr_step=0.002, top_3_acc_step=0.124, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 2:  30%|███       | 3/10 [00:00<00:01,  5.70it/s, v_num=0, total_loss_step=2.2e+4, lr_step=0.002, top_3_acc_step=0.124, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 2:  30%|███       | 3/10 [00:00<00:01,  5.69it/s, v_num=0, total_loss_step=2.2e+4, lr_step=0.002, top_3_acc_step=0.129, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 2:  40%|████      | 4/10 [00:00<00:01,  5.77it/s, v_num=0, total_loss_step=2.2e+4, lr_step=0.002, top_3_acc_step=0.129, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 2:  40%|████      | 4/10 [00:00<00:01,  5.76it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.125, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 2:  50%|█████     | 5/10 [00:00<00:00,  5.77it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.125, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 2:  50%|█████     | 5/10 [00:00<00:00,  5.77it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.130, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 2:  60%|██████    | 6/10 [00:01<00:00,  5.78it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.130, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 2:  60%|██████    | 6/10 [00:01<00:00,  5.78it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.135, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 2:  70%|███████   | 7/10 [00:01<00:00,  5.81it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.135, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 2:  70%|███████   | 7/10 [00:01<00:00,  5.81it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.133, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 2:  80%|████████  | 8/10 [00:01<00:00,  5.83it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.133, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 2:  80%|████████  | 8/10 [00:01<00:00,  5.83it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.129, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 2:  90%|█████████ | 9/10 [00:01<00:00,  5.83it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.129, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 2:  90%|█████████ | 9/10 [00:01<00:00,  5.83it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.131, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 2: 100%|██████████| 10/10 [00:01<00:00,  6.28it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.131, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 2: 100%|██████████| 10/10 [00:01<00:00,  6.27it/s, v_num=0, total_loss_step=1.99e+3, lr_step=0.002, top_3_acc_step=0.139, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 2: 100%|██████████| 10/10 [00:01<00:00,  6.27it/s, v_num=0, total_loss_step=1.99e+3, lr_step=0.002, top_3_acc_step=0.139, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 2:   0%|          | 0/10 [00:00<?, ?it/s, v_num=0, total_loss_step=1.99e+3, lr_step=0.002, top_3_acc_step=0.139, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 3:   0%|          | 0/10 [00:00<?, ?it/s, v_num=0, total_loss_step=1.99e+3, lr_step=0.002, top_3_acc_step=0.139, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 3:  10%|█         | 1/10 [00:00<00:01,  5.41it/s, v_num=0, total_loss_step=1.99e+3, lr_step=0.002, top_3_acc_step=0.139, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 3:  10%|█         | 1/10 [00:00<00:01,  5.39it/s, v_num=0, total_loss_step=2.2e+4, lr_step=0.002, top_3_acc_step=0.128, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 3:  20%|██        | 2/10 [00:00<00:01,  5.57it/s, v_num=0, total_loss_step=2.2e+4, lr_step=0.002, top_3_acc_step=0.128, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 3:  20%|██        | 2/10 [00:00<00:01,  5.56it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.125, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 3:  30%|███       | 3/10 [00:00<00:01,  5.67it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.125, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 3:  30%|███       | 3/10 [00:00<00:01,  5.66it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.129, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 3:  40%|████      | 4/10 [00:00<00:01,  5.73it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.129, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 3:  40%|████      | 4/10 [00:00<00:01,  5.72it/s, v_num=0, total_loss_step=2.2e+4, lr_step=0.002, top_3_acc_step=0.128, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 3:  50%|█████     | 5/10 [00:00<00:00,  5.74it/s, v_num=0, total_loss_step=2.2e+4, lr_step=0.002, top_3_acc_step=0.128, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 3:  50%|█████     | 5/10 [00:00<00:00,  5.73it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.132, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 3:  60%|██████    | 6/10 [00:01<00:00,  5.75it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.132, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 3:  60%|██████    | 6/10 [00:01<00:00,  5.74it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.128, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 3:  70%|███████   | 7/10 [00:01<00:00,  5.78it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.128, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 3:  70%|███████   | 7/10 [00:01<00:00,  5.77it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.132, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 3:  80%|████████  | 8/10 [00:01<00:00,  5.80it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.132, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 3:  80%|████████  | 8/10 [00:01<00:00,  5.80it/s, v_num=0, total_loss_step=2.2e+4, lr_step=0.002, top_3_acc_step=0.127, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 3:  90%|█████████ | 9/10 [00:01<00:00,  5.80it/s, v_num=0, total_loss_step=2.2e+4, lr_step=0.002, top_3_acc_step=0.127, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 3:  90%|█████████ | 9/10 [00:01<00:00,  5.80it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.130, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 3: 100%|██████████| 10/10 [00:01<00:00,  6.25it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.130, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 3: 100%|██████████| 10/10 [00:01<00:00,  6.25it/s, v_num=0, total_loss_step=2e+3, lr_step=0.002, top_3_acc_step=0.122, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 3: 100%|██████████| 10/10 [00:01<00:00,  6.24it/s, v_num=0, total_loss_step=2e+3, lr_step=0.002, top_3_acc_step=0.122, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 3:   0%|          | 0/10 [00:00<?, ?it/s, v_num=0, total_loss_step=2e+3, lr_step=0.002, top_3_acc_step=0.122, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 4:   0%|          | 0/10 [00:00<?, ?it/s, v_num=0, total_loss_step=2e+3, lr_step=0.002, top_3_acc_step=0.122, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 4:  10%|█         | 1/10 [00:00<00:01,  5.23it/s, v_num=0, total_loss_step=2e+3, lr_step=0.002, top_3_acc_step=0.122, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 4:  10%|█         | 1/10 [00:00<00:01,  5.22it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.129, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 4:  20%|██        | 2/10 [00:00<00:01,  5.58it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.129, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 4:  20%|██        | 2/10 [00:00<00:01,  5.58it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.138, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 4:  30%|███       | 3/10 [00:00<00:01,  5.71it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.138, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 4:  30%|███       | 3/10 [00:00<00:01,  5.71it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.132, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 4:  40%|████      | 4/10 [00:00<00:01,  5.78it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.132, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 4:  40%|████      | 4/10 [00:00<00:01,  5.78it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.133, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 4:  50%|█████     | 5/10 [00:00<00:00,  5.74it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.133, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 4:  50%|█████     | 5/10 [00:00<00:00,  5.73it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.131, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 4:  60%|██████    | 6/10 [00:01<00:00,  5.78it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.131, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 4:  60%|██████    | 6/10 [00:01<00:00,  5.78it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.126, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 4:  70%|███████   | 7/10 [00:01<00:00,  5.82it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.126, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 4:  70%|███████   | 7/10 [00:01<00:00,  5.81it/s, v_num=0, total_loss_step=2.2e+4, lr_step=0.002, top_3_acc_step=0.130, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 4:  80%|████████  | 8/10 [00:01<00:00,  5.84it/s, v_num=0, total_loss_step=2.2e+4, lr_step=0.002, top_3_acc_step=0.130, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 4:  80%|████████  | 8/10 [00:01<00:00,  5.83it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.127, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 4:  90%|█████████ | 9/10 [00:01<00:00,  5.84it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.127, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 4:  90%|█████████ | 9/10 [00:01<00:00,  5.84it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.131, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 4: 100%|██████████| 10/10 [00:01<00:00,  6.29it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.131, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 4: 100%|██████████| 10/10 [00:01<00:00,  6.29it/s, v_num=0, total_loss_step=2e+3, lr_step=0.002, top_3_acc_step=0.124, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.129]
Epoch 4: 100%|██████████| 10/10 [00:01<00:00,  6.28it/s, v_num=0, total_loss_step=2e+3, lr_step=0.002, top_3_acc_step=0.124, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.131]
Epoch 4:   0%|          | 0/10 [00:00<?, ?it/s, v_num=0, total_loss_step=2e+3, lr_step=0.002, top_3_acc_step=0.124, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.131]
Epoch 5:   0%|          | 0/10 [00:00<?, ?it/s, v_num=0, total_loss_step=2e+3, lr_step=0.002, top_3_acc_step=0.124, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.131]
Epoch 5:  10%|█         | 1/10 [00:00<00:01,  5.30it/s, v_num=0, total_loss_step=2e+3, lr_step=0.002, top_3_acc_step=0.124, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.131]
Epoch 5:  10%|█         | 1/10 [00:00<00:01,  5.28it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.131, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.131]
Epoch 5:  20%|██        | 2/10 [00:00<00:01,  5.60it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.131, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.131]
Epoch 5:  20%|██        | 2/10 [00:00<00:01,  5.59it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.127, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.131]
Epoch 5:  30%|███       | 3/10 [00:00<00:01,  5.70it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.127, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.131]
Epoch 5:  30%|███       | 3/10 [00:00<00:01,  5.70it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.128, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.131]
Epoch 5:  40%|████      | 4/10 [00:00<00:01,  5.77it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.128, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.131]
Epoch 5:  40%|████      | 4/10 [00:00<00:01,  5.77it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.134, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.131]
Epoch 5:  50%|█████     | 5/10 [00:00<00:00,  5.50it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.134, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.131]
Epoch 5:  50%|█████     | 5/10 [00:00<00:00,  5.49it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.130, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.131]
Epoch 5:  60%|██████    | 6/10 [00:01<00:00,  5.57it/s, v_num=0, total_loss_step=2.19e+4, lr_step=0.002, top_3_acc_step=0.130, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.131]
Epoch 5:  60%|██████    | 6/10 [00:01<00:00,  5.57it/s, v_num=0, total_loss_step=2.17e+4, lr_step=0.002, top_3_acc_step=0.131, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.131]
Epoch 5:  70%|███████   | 7/10 [00:01<00:00,  5.60it/s, v_num=0, total_loss_step=2.17e+4, lr_step=0.002, top_3_acc_step=0.131, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.131]
Epoch 5:  70%|███████   | 7/10 [00:01<00:00,  5.60it/s, v_num=0, total_loss_step=2.17e+4, lr_step=0.002, top_3_acc_step=0.135, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.131]
Epoch 5:  80%|████████  | 8/10 [00:01<00:00,  5.65it/s, v_num=0, total_loss_step=2.17e+4, lr_step=0.002, top_3_acc_step=0.135, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.131]
Epoch 5:  80%|████████  | 8/10 [00:01<00:00,  5.65it/s, v_num=0, total_loss_step=2.17e+4, lr_step=0.002, top_3_acc_step=0.133, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.131]
Epoch 5:  90%|█████████ | 9/10 [00:01<00:00,  5.60it/s, v_num=0, total_loss_step=2.17e+4, lr_step=0.002, top_3_acc_step=0.133, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.131]
Epoch 5:  90%|█████████ | 9/10 [00:01<00:00,  5.60it/s, v_num=0, total_loss_step=2.17e+4, lr_step=0.002, top_3_acc_step=0.134, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.131]
Epoch 5: 100%|██████████| 10/10 [00:01<00:00,  6.03it/s, v_num=0, total_loss_step=2.17e+4, lr_step=0.002, top_3_acc_step=0.134, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.131]
Epoch 5: 100%|██████████| 10/10 [00:01<00:00,  6.03it/s, v_num=0, total_loss_step=1.9e+3, lr_step=0.002, top_3_acc_step=0.157, total_loss_epoch=2.17e+4, lr_epoch=0.002, top_3_acc_epoch=0.131]
Epoch 5: 100%|██████████| 10/10 [00:01<00:00,  6.02it/s, v_num=0, total_loss_step=1.9e+3, lr_step=0.002, top_3_acc_step=0.157, total_loss_epoch=2.16e+4, lr_epoch=0.002, top_3_acc_epoch=0.132]
Epoch 5:   0%|          | 0/10 [00:00<?, ?it/s, v_num=0, total_loss_step=1.9e+3, lr_step=0.002, top_3_acc_step=0.157, total_loss_epoch=2.16e+4, lr_epoch=0.002, top_3_acc_epoch=0.132]
Epoch 6:   0%|          | 0/10 [00:00<?, ?it/s, v_num=0, total_loss_step=1.9e+3, lr_step=0.002, top_3_acc_step=0.157, total_loss_epoch=2.16e+4, lr_epoch=0.002, top_3_acc_epoch=0.132]
Epoch 6:  10%|█         | 1/10 [00:00<00:01,  5.23it/s, v_num=0, total_loss_step=1.9e+3, lr_step=0.002, top_3_acc_step=0.157, total_loss_epoch=2.16e+4, lr_epoch=0.002, top_3_acc_epoch=0.132]
Epoch 6:  10%|█         | 1/10 [00:00<00:01,  5.22it/s, v_num=0, total_loss_step=2.08e+4, lr_step=0.002, top_3_acc_step=0.159, total_loss_epoch=2.16e+4, lr_epoch=0.002, top_3_acc_epoch=0.132]
Epoch 6:  20%|██        | 2/10 [00:00<00:01,  5.57it/s, v_num=0, total_loss_step=2.08e+4, lr_step=0.002, top_3_acc_step=0.159, total_loss_epoch=2.16e+4, lr_epoch=0.002, top_3_acc_epoch=0.132]
Epoch 6:  20%|██        | 2/10 [00:00<00:01,  5.56it/s, v_num=0, total_loss_step=2.09e+4, lr_step=0.002, top_3_acc_step=0.155, total_loss_epoch=2.16e+4, lr_epoch=0.002, top_3_acc_epoch=0.132]
Epoch 6:  30%|███       | 3/10 [00:00<00:01,  5.70it/s, v_num=0, total_loss_step=2.09e+4, lr_step=0.002, top_3_acc_step=0.155, total_loss_epoch=2.16e+4, lr_epoch=0.002, top_3_acc_epoch=0.132]
Epoch 6:  30%|███       | 3/10 [00:00<00:01,  5.69it/s, v_num=0, total_loss_step=2.09e+4, lr_step=0.002, top_3_acc_step=0.158, total_loss_epoch=2.16e+4, lr_epoch=0.002, top_3_acc_epoch=0.132]
Epoch 6:  40%|████      | 4/10 [00:00<00:01,  5.76it/s, v_num=0, total_loss_step=2.09e+4, lr_step=0.002, top_3_acc_step=0.158, total_loss_epoch=2.16e+4, lr_epoch=0.002, top_3_acc_epoch=0.132]
Epoch 6:  40%|████      | 4/10 [00:00<00:01,  5.76it/s, v_num=0, total_loss_step=2.09e+4, lr_step=0.002, top_3_acc_step=0.160, total_loss_epoch=2.16e+4, lr_epoch=0.002, top_3_acc_epoch=0.132]
Epoch 6:  50%|█████     | 5/10 [00:00<00:00,  5.63it/s, v_num=0, total_loss_step=2.09e+4, lr_step=0.002, top_3_acc_step=0.160, total_loss_epoch=2.16e+4, lr_epoch=0.002, top_3_acc_epoch=0.132]
Epoch 6:  50%|█████     | 5/10 [00:00<00:00,  5.63it/s, v_num=0, total_loss_step=2.09e+4, lr_step=0.002, top_3_acc_step=0.160, total_loss_epoch=2.16e+4, lr_epoch=0.002, top_3_acc_epoch=0.132]
Epoch 6:  60%|██████    | 6/10 [00:01<00:00,  5.69it/s, v_num=0, total_loss_step=2.09e+4, lr_step=0.002, top_3_acc_step=0.160, total_loss_epoch=2.16e+4, lr_epoch=0.002, top_3_acc_epoch=0.132]
Epoch 6:  60%|██████    | 6/10 [00:01<00:00,  5.68it/s, v_num=0, total_loss_step=2.01e+4, lr_step=0.002, top_3_acc_step=0.189, total_loss_epoch=2.16e+4, lr_epoch=0.002, top_3_acc_epoch=0.132]
Epoch 6:  70%|███████   | 7/10 [00:01<00:00,  5.72it/s, v_num=0, total_loss_step=2.01e+4, lr_step=0.002, top_3_acc_step=0.189, total_loss_epoch=2.16e+4, lr_epoch=0.002, top_3_acc_epoch=0.132]
Epoch 6:  70%|███████   | 7/10 [00:01<00:00,  5.72it/s, v_num=0, total_loss_step=2.01e+4, lr_step=0.002, top_3_acc_step=0.186, total_loss_epoch=2.16e+4, lr_epoch=0.002, top_3_acc_epoch=0.132]
Epoch 6:  80%|████████  | 8/10 [00:01<00:00,  5.76it/s, v_num=0, total_loss_step=2.01e+4, lr_step=0.002, top_3_acc_step=0.186, total_loss_epoch=2.16e+4, lr_epoch=0.002, top_3_acc_epoch=0.132]
Epoch 6:  80%|████████  | 8/10 [00:01<00:00,  5.75it/s, v_num=0, total_loss_step=2.01e+4, lr_step=0.002, top_3_acc_step=0.179, total_loss_epoch=2.16e+4, lr_epoch=0.002, top_3_acc_epoch=0.132]
Epoch 6:  90%|█████████ | 9/10 [00:01<00:00,  5.67it/s, v_num=0, total_loss_step=2.01e+4, lr_step=0.002, top_3_acc_step=0.179, total_loss_epoch=2.16e+4, lr_epoch=0.002, top_3_acc_epoch=0.132]
Epoch 6:  90%|█████████ | 9/10 [00:01<00:00,  5.67it/s, v_num=0, total_loss_step=2.01e+4, lr_step=0.002, top_3_acc_step=0.185, total_loss_epoch=2.16e+4, lr_epoch=0.002, top_3_acc_epoch=0.132]
Epoch 6: 100%|██████████| 10/10 [00:01<00:00,  6.10it/s, v_num=0, total_loss_step=2.01e+4, lr_step=0.002, top_3_acc_step=0.185, total_loss_epoch=2.16e+4, lr_epoch=0.002, top_3_acc_epoch=0.132]
Epoch 6: 100%|██████████| 10/10 [00:01<00:00,  6.10it/s, v_num=0, total_loss_step=1.75e+3, lr_step=0.002, top_3_acc_step=0.195, total_loss_epoch=2.16e+4, lr_epoch=0.002, top_3_acc_epoch=0.132]
Epoch 6: 100%|██████████| 10/10 [00:01<00:00,  6.09it/s, v_num=0, total_loss_step=1.75e+3, lr_step=0.002, top_3_acc_step=0.195, total_loss_epoch=2.03e+4, lr_epoch=0.002, top_3_acc_epoch=0.170]
Epoch 6:   0%|          | 0/10 [00:00<?, ?it/s, v_num=0, total_loss_step=1.75e+3, lr_step=0.002, top_3_acc_step=0.195, total_loss_epoch=2.03e+4, lr_epoch=0.002, top_3_acc_epoch=0.170]
Epoch 7:   0%|          | 0/10 [00:00<?, ?it/s, v_num=0, total_loss_step=1.75e+3, lr_step=0.002, top_3_acc_step=0.195, total_loss_epoch=2.03e+4, lr_epoch=0.002, top_3_acc_epoch=0.170]
Epoch 7:  10%|█         | 1/10 [00:00<00:01,  5.27it/s, v_num=0, total_loss_step=1.75e+3, lr_step=0.002, top_3_acc_step=0.195, total_loss_epoch=2.03e+4, lr_epoch=0.002, top_3_acc_epoch=0.170]
Epoch 7:  10%|█         | 1/10 [00:00<00:01,  5.26it/s, v_num=0, total_loss_step=1.92e+4, lr_step=0.002, top_3_acc_step=0.216, total_loss_epoch=2.03e+4, lr_epoch=0.002, top_3_acc_epoch=0.170]
Epoch 7:  20%|██        | 2/10 [00:00<00:01,  5.60it/s, v_num=0, total_loss_step=1.92e+4, lr_step=0.002, top_3_acc_step=0.216, total_loss_epoch=2.03e+4, lr_epoch=0.002, top_3_acc_epoch=0.170]
Epoch 7:  20%|██        | 2/10 [00:00<00:01,  5.59it/s, v_num=0, total_loss_step=1.92e+4, lr_step=0.002, top_3_acc_step=0.221, total_loss_epoch=2.03e+4, lr_epoch=0.002, top_3_acc_epoch=0.170]
Epoch 7:  30%|███       | 3/10 [00:00<00:01,  5.72it/s, v_num=0, total_loss_step=1.92e+4, lr_step=0.002, top_3_acc_step=0.221, total_loss_epoch=2.03e+4, lr_epoch=0.002, top_3_acc_epoch=0.170]
Epoch 7:  30%|███       | 3/10 [00:00<00:01,  5.71it/s, v_num=0, total_loss_step=1.91e+4, lr_step=0.002, top_3_acc_step=0.221, total_loss_epoch=2.03e+4, lr_epoch=0.002, top_3_acc_epoch=0.170]
Epoch 7:  40%|████      | 4/10 [00:00<00:01,  5.78it/s, v_num=0, total_loss_step=1.91e+4, lr_step=0.002, top_3_acc_step=0.221, total_loss_epoch=2.03e+4, lr_epoch=0.002, top_3_acc_epoch=0.170]
Epoch 7:  40%|████      | 4/10 [00:00<00:01,  5.77it/s, v_num=0, total_loss_step=1.92e+4, lr_step=0.002, top_3_acc_step=0.217, total_loss_epoch=2.03e+4, lr_epoch=0.002, top_3_acc_epoch=0.170]
Epoch 7:  50%|█████     | 5/10 [00:00<00:00,  5.63it/s, v_num=0, total_loss_step=1.92e+4, lr_step=0.002, top_3_acc_step=0.217, total_loss_epoch=2.03e+4, lr_epoch=0.002, top_3_acc_epoch=0.170]
Epoch 7:  50%|█████     | 5/10 [00:00<00:00,  5.62it/s, v_num=0, total_loss_step=1.92e+4, lr_step=0.002, top_3_acc_step=0.225, total_loss_epoch=2.03e+4, lr_epoch=0.002, top_3_acc_epoch=0.170]
Epoch 7:  60%|██████    | 6/10 [00:01<00:00,  5.69it/s, v_num=0, total_loss_step=1.92e+4, lr_step=0.002, top_3_acc_step=0.225, total_loss_epoch=2.03e+4, lr_epoch=0.002, top_3_acc_epoch=0.170]
Epoch 7:  60%|██████    | 6/10 [00:01<00:00,  5.69it/s, v_num=0, total_loss_step=1.82e+4, lr_step=0.002, top_3_acc_step=0.260, total_loss_epoch=2.03e+4, lr_epoch=0.002, top_3_acc_epoch=0.170]
Epoch 7:  70%|███████   | 7/10 [00:01<00:00,  5.73it/s, v_num=0, total_loss_step=1.82e+4, lr_step=0.002, top_3_acc_step=0.260, total_loss_epoch=2.03e+4, lr_epoch=0.002, top_3_acc_epoch=0.170]
Epoch 7:  70%|███████   | 7/10 [00:01<00:00,  5.73it/s, v_num=0, total_loss_step=1.82e+4, lr_step=0.002, top_3_acc_step=0.263, total_loss_epoch=2.03e+4, lr_epoch=0.002, top_3_acc_epoch=0.170]
Epoch 7:  80%|████████  | 8/10 [00:01<00:00,  5.76it/s, v_num=0, total_loss_step=1.82e+4, lr_step=0.002, top_3_acc_step=0.263, total_loss_epoch=2.03e+4, lr_epoch=0.002, top_3_acc_epoch=0.170]
Epoch 7:  80%|████████  | 8/10 [00:01<00:00,  5.76it/s, v_num=0, total_loss_step=1.82e+4, lr_step=0.002, top_3_acc_step=0.255, total_loss_epoch=2.03e+4, lr_epoch=0.002, top_3_acc_epoch=0.170]
Epoch 7:  90%|█████████ | 9/10 [00:01<00:00,  5.70it/s, v_num=0, total_loss_step=1.82e+4, lr_step=0.002, top_3_acc_step=0.255, total_loss_epoch=2.03e+4, lr_epoch=0.002, top_3_acc_epoch=0.170]
Epoch 7:  90%|█████████ | 9/10 [00:01<00:00,  5.70it/s, v_num=0, total_loss_step=1.82e+4, lr_step=0.002, top_3_acc_step=0.256, total_loss_epoch=2.03e+4, lr_epoch=0.002, top_3_acc_epoch=0.170]
Epoch 7: 100%|██████████| 10/10 [00:01<00:00,  6.14it/s, v_num=0, total_loss_step=1.82e+4, lr_step=0.002, top_3_acc_step=0.256, total_loss_epoch=2.03e+4, lr_epoch=0.002, top_3_acc_epoch=0.170]
Epoch 7: 100%|██████████| 10/10 [00:01<00:00,  6.13it/s, v_num=0, total_loss_step=1.56e+3, lr_step=0.002, top_3_acc_step=0.324, total_loss_epoch=2.03e+4, lr_epoch=0.002, top_3_acc_epoch=0.170]
Epoch 7: 100%|██████████| 10/10 [00:01<00:00,  6.13it/s, v_num=0, total_loss_step=1.56e+3, lr_step=0.002, top_3_acc_step=0.324, total_loss_epoch=1.86e+4, lr_epoch=0.002, top_3_acc_epoch=0.238]
Epoch 7:   0%|          | 0/10 [00:00<?, ?it/s, v_num=0, total_loss_step=1.56e+3, lr_step=0.002, top_3_acc_step=0.324, total_loss_epoch=1.86e+4, lr_epoch=0.002, top_3_acc_epoch=0.238]
Epoch 8:   0%|          | 0/10 [00:00<?, ?it/s, v_num=0, total_loss_step=1.56e+3, lr_step=0.002, top_3_acc_step=0.324, total_loss_epoch=1.86e+4, lr_epoch=0.002, top_3_acc_epoch=0.238]
Epoch 8:  10%|█         | 1/10 [00:00<00:01,  5.31it/s, v_num=0, total_loss_step=1.56e+3, lr_step=0.002, top_3_acc_step=0.324, total_loss_epoch=1.86e+4, lr_epoch=0.002, top_3_acc_epoch=0.238]
Epoch 8:  10%|█         | 1/10 [00:00<00:01,  5.29it/s, v_num=0, total_loss_step=1.71e+4, lr_step=0.002, top_3_acc_step=0.303, total_loss_epoch=1.86e+4, lr_epoch=0.002, top_3_acc_epoch=0.238]
Epoch 8:  20%|██        | 2/10 [00:00<00:01,  5.59it/s, v_num=0, total_loss_step=1.71e+4, lr_step=0.002, top_3_acc_step=0.303, total_loss_epoch=1.86e+4, lr_epoch=0.002, top_3_acc_epoch=0.238]
Epoch 8:  20%|██        | 2/10 [00:00<00:01,  5.59it/s, v_num=0, total_loss_step=1.71e+4, lr_step=0.002, top_3_acc_step=0.317, total_loss_epoch=1.86e+4, lr_epoch=0.002, top_3_acc_epoch=0.238]
Epoch 8:  30%|███       | 3/10 [00:00<00:01,  5.72it/s, v_num=0, total_loss_step=1.71e+4, lr_step=0.002, top_3_acc_step=0.317, total_loss_epoch=1.86e+4, lr_epoch=0.002, top_3_acc_epoch=0.238]
Epoch 8:  30%|███       | 3/10 [00:00<00:01,  5.72it/s, v_num=0, total_loss_step=1.71e+4, lr_step=0.002, top_3_acc_step=0.314, total_loss_epoch=1.86e+4, lr_epoch=0.002, top_3_acc_epoch=0.238]
Epoch 8:  40%|████      | 4/10 [00:00<00:01,  5.79it/s, v_num=0, total_loss_step=1.71e+4, lr_step=0.002, top_3_acc_step=0.314, total_loss_epoch=1.86e+4, lr_epoch=0.002, top_3_acc_epoch=0.238]
Epoch 8:  40%|████      | 4/10 [00:00<00:01,  5.78it/s, v_num=0, total_loss_step=1.71e+4, lr_step=0.002, top_3_acc_step=0.310, total_loss_epoch=1.86e+4, lr_epoch=0.002, top_3_acc_epoch=0.238]
Epoch 8:  50%|█████     | 5/10 [00:00<00:00,  5.64it/s, v_num=0, total_loss_step=1.71e+4, lr_step=0.002, top_3_acc_step=0.310, total_loss_epoch=1.86e+4, lr_epoch=0.002, top_3_acc_epoch=0.238]
Epoch 8:  50%|█████     | 5/10 [00:00<00:00,  5.64it/s, v_num=0, total_loss_step=1.71e+4, lr_step=0.002, top_3_acc_step=0.313, total_loss_epoch=1.86e+4, lr_epoch=0.002, top_3_acc_epoch=0.238]
Epoch 8:  60%|██████    | 6/10 [00:01<00:00,  5.69it/s, v_num=0, total_loss_step=1.71e+4, lr_step=0.002, top_3_acc_step=0.313, total_loss_epoch=1.86e+4, lr_epoch=0.002, top_3_acc_epoch=0.238]
Epoch 8:  60%|██████    | 6/10 [00:01<00:00,  5.69it/s, v_num=0, total_loss_step=1.59e+4, lr_step=0.002, top_3_acc_step=0.376, total_loss_epoch=1.86e+4, lr_epoch=0.002, top_3_acc_epoch=0.238]
Epoch 8:  70%|███████   | 7/10 [00:01<00:00,  5.74it/s, v_num=0, total_loss_step=1.59e+4, lr_step=0.002, top_3_acc_step=0.376, total_loss_epoch=1.86e+4, lr_epoch=0.002, top_3_acc_epoch=0.238]
Epoch 8:  70%|███████   | 7/10 [00:01<00:00,  5.73it/s, v_num=0, total_loss_step=1.59e+4, lr_step=0.002, top_3_acc_step=0.374, total_loss_epoch=1.86e+4, lr_epoch=0.002, top_3_acc_epoch=0.238]
Epoch 8:  80%|████████  | 8/10 [00:01<00:00,  5.77it/s, v_num=0, total_loss_step=1.59e+4, lr_step=0.002, top_3_acc_step=0.374, total_loss_epoch=1.86e+4, lr_epoch=0.002, top_3_acc_epoch=0.238]
Epoch 8:  80%|████████  | 8/10 [00:01<00:00,  5.77it/s, v_num=0, total_loss_step=1.59e+4, lr_step=0.002, top_3_acc_step=0.383, total_loss_epoch=1.86e+4, lr_epoch=0.002, top_3_acc_epoch=0.238]
Epoch 8:  90%|█████████ | 9/10 [00:01<00:00,  5.69it/s, v_num=0, total_loss_step=1.59e+4, lr_step=0.002, top_3_acc_step=0.383, total_loss_epoch=1.86e+4, lr_epoch=0.002, top_3_acc_epoch=0.238]
Epoch 8:  90%|█████████ | 9/10 [00:01<00:00,  5.68it/s, v_num=0, total_loss_step=1.59e+4, lr_step=0.002, top_3_acc_step=0.373, total_loss_epoch=1.86e+4, lr_epoch=0.002, top_3_acc_epoch=0.238]
Epoch 8: 100%|██████████| 10/10 [00:01<00:00,  6.12it/s, v_num=0, total_loss_step=1.59e+4, lr_step=0.002, top_3_acc_step=0.373, total_loss_epoch=1.86e+4, lr_epoch=0.002, top_3_acc_epoch=0.238]
Epoch 8: 100%|██████████| 10/10 [00:01<00:00,  6.12it/s, v_num=0, total_loss_step=1.34e+3, lr_step=0.002, top_3_acc_step=0.461, total_loss_epoch=1.86e+4, lr_epoch=0.002, top_3_acc_epoch=0.238]
Epoch 8: 100%|██████████| 10/10 [00:01<00:00,  6.11it/s, v_num=0, total_loss_step=1.34e+3, lr_step=0.002, top_3_acc_step=0.461, total_loss_epoch=1.64e+4, lr_epoch=0.002, top_3_acc_epoch=0.341]
Epoch 8:   0%|          | 0/10 [00:00<?, ?it/s, v_num=0, total_loss_step=1.34e+3, lr_step=0.002, top_3_acc_step=0.461, total_loss_epoch=1.64e+4, lr_epoch=0.002, top_3_acc_epoch=0.341]
Epoch 9:   0%|          | 0/10 [00:00<?, ?it/s, v_num=0, total_loss_step=1.34e+3, lr_step=0.002, top_3_acc_step=0.461, total_loss_epoch=1.64e+4, lr_epoch=0.002, top_3_acc_epoch=0.341]
Epoch 9:  10%|█         | 1/10 [00:00<00:01,  5.24it/s, v_num=0, total_loss_step=1.34e+3, lr_step=0.002, top_3_acc_step=0.461, total_loss_epoch=1.64e+4, lr_epoch=0.002, top_3_acc_epoch=0.341]
Epoch 9:  10%|█         | 1/10 [00:00<00:01,  5.22it/s, v_num=0, total_loss_step=1.46e+4, lr_step=0.002, top_3_acc_step=0.444, total_loss_epoch=1.64e+4, lr_epoch=0.002, top_3_acc_epoch=0.341]
Epoch 9:  20%|██        | 2/10 [00:00<00:01,  5.58it/s, v_num=0, total_loss_step=1.46e+4, lr_step=0.002, top_3_acc_step=0.444, total_loss_epoch=1.64e+4, lr_epoch=0.002, top_3_acc_epoch=0.341]
Epoch 9:  20%|██        | 2/10 [00:00<00:01,  5.57it/s, v_num=0, total_loss_step=1.46e+4, lr_step=0.002, top_3_acc_step=0.448, total_loss_epoch=1.64e+4, lr_epoch=0.002, top_3_acc_epoch=0.341]
Epoch 9:  30%|███       | 3/10 [00:00<00:01,  5.70it/s, v_num=0, total_loss_step=1.46e+4, lr_step=0.002, top_3_acc_step=0.448, total_loss_epoch=1.64e+4, lr_epoch=0.002, top_3_acc_epoch=0.341]
Epoch 9:  30%|███       | 3/10 [00:00<00:01,  5.69it/s, v_num=0, total_loss_step=1.46e+4, lr_step=0.002, top_3_acc_step=0.442, total_loss_epoch=1.64e+4, lr_epoch=0.002, top_3_acc_epoch=0.341]
Epoch 9:  40%|████      | 4/10 [00:00<00:01,  5.76it/s, v_num=0, total_loss_step=1.46e+4, lr_step=0.002, top_3_acc_step=0.442, total_loss_epoch=1.64e+4, lr_epoch=0.002, top_3_acc_epoch=0.341]
Epoch 9:  40%|████      | 4/10 [00:00<00:01,  5.76it/s, v_num=0, total_loss_step=1.46e+4, lr_step=0.002, top_3_acc_step=0.442, total_loss_epoch=1.64e+4, lr_epoch=0.002, top_3_acc_epoch=0.341]
Epoch 9:  50%|█████     | 5/10 [00:00<00:00,  5.57it/s, v_num=0, total_loss_step=1.46e+4, lr_step=0.002, top_3_acc_step=0.442, total_loss_epoch=1.64e+4, lr_epoch=0.002, top_3_acc_epoch=0.341]
Epoch 9:  50%|█████     | 5/10 [00:00<00:00,  5.57it/s, v_num=0, total_loss_step=1.46e+4, lr_step=0.002, top_3_acc_step=0.448, total_loss_epoch=1.64e+4, lr_epoch=0.002, top_3_acc_epoch=0.341]
Epoch 9:  60%|██████    | 6/10 [00:01<00:00,  5.65it/s, v_num=0, total_loss_step=1.46e+4, lr_step=0.002, top_3_acc_step=0.448, total_loss_epoch=1.64e+4, lr_epoch=0.002, top_3_acc_epoch=0.341]
Epoch 9:  60%|██████    | 6/10 [00:01<00:00,  5.64it/s, v_num=0, total_loss_step=1.32e+4, lr_step=0.002, top_3_acc_step=0.510, total_loss_epoch=1.64e+4, lr_epoch=0.002, top_3_acc_epoch=0.341]
Epoch 9:  70%|███████   | 7/10 [00:01<00:00,  5.70it/s, v_num=0, total_loss_step=1.32e+4, lr_step=0.002, top_3_acc_step=0.510, total_loss_epoch=1.64e+4, lr_epoch=0.002, top_3_acc_epoch=0.341]
Epoch 9:  70%|███████   | 7/10 [00:01<00:00,  5.69it/s, v_num=0, total_loss_step=1.31e+4, lr_step=0.002, top_3_acc_step=0.519, total_loss_epoch=1.64e+4, lr_epoch=0.002, top_3_acc_epoch=0.341]
Epoch 9:  80%|████████  | 8/10 [00:01<00:00,  5.73it/s, v_num=0, total_loss_step=1.31e+4, lr_step=0.002, top_3_acc_step=0.519, total_loss_epoch=1.64e+4, lr_epoch=0.002, top_3_acc_epoch=0.341]
Epoch 9:  80%|████████  | 8/10 [00:01<00:00,  5.73it/s, v_num=0, total_loss_step=1.31e+4, lr_step=0.002, top_3_acc_step=0.516, total_loss_epoch=1.64e+4, lr_epoch=0.002, top_3_acc_epoch=0.341]
Epoch 9:  90%|█████████ | 9/10 [00:01<00:00,  5.74it/s, v_num=0, total_loss_step=1.31e+4, lr_step=0.002, top_3_acc_step=0.516, total_loss_epoch=1.64e+4, lr_epoch=0.002, top_3_acc_epoch=0.341]
Epoch 9:  90%|█████████ | 9/10 [00:01<00:00,  5.74it/s, v_num=0, total_loss_step=1.32e+4, lr_step=0.002, top_3_acc_step=0.514, total_loss_epoch=1.64e+4, lr_epoch=0.002, top_3_acc_epoch=0.341]
Epoch 9: 100%|██████████| 10/10 [00:01<00:00,  6.19it/s, v_num=0, total_loss_step=1.32e+4, lr_step=0.002, top_3_acc_step=0.514, total_loss_epoch=1.64e+4, lr_epoch=0.002, top_3_acc_epoch=0.341]
Epoch 9: 100%|██████████| 10/10 [00:01<00:00,  6.19it/s, v_num=0, total_loss_step=1.21e+3, lr_step=0.002, top_3_acc_step=0.501, total_loss_epoch=1.64e+4, lr_epoch=0.002, top_3_acc_epoch=0.341]
Epoch 9: 100%|██████████| 10/10 [00:01<00:00,  6.18it/s, v_num=0, total_loss_step=1.21e+3, lr_step=0.002, top_3_acc_step=0.501, total_loss_epoch=1.38e+4, lr_epoch=0.002, top_3_acc_epoch=0.476]
Epoch 9:   0%|          | 0/10 [00:00<?, ?it/s, v_num=0, total_loss_step=1.21e+3, lr_step=0.002, top_3_acc_step=0.501, total_loss_epoch=1.38e+4, lr_epoch=0.002, top_3_acc_epoch=0.476]
Epoch 10:   0%|          | 0/10 [00:00<?, ?it/s, v_num=0, total_loss_step=1.21e+3, lr_step=0.002, top_3_acc_step=0.501, total_loss_epoch=1.38e+4, lr_epoch=0.002, top_3_acc_epoch=0.476]
Epoch 10:  10%|█         | 1/10 [00:00<00:01,  5.27it/s, v_num=0, total_loss_step=1.21e+3, lr_step=0.002, top_3_acc_step=0.501, total_loss_epoch=1.38e+4, lr_epoch=0.002, top_3_acc_epoch=0.476]
Epoch 10:  10%|█         | 1/10 [00:00<00:01,  5.25it/s, v_num=0, total_loss_step=1.32e+4, lr_step=0.002, top_3_acc_step=0.509, total_loss_epoch=1.38e+4, lr_epoch=0.002, top_3_acc_epoch=0.476]
Epoch 10:  20%|██        | 2/10 [00:00<00:01,  5.57it/s, v_num=0, total_loss_step=1.32e+4, lr_step=0.002, top_3_acc_step=0.509, total_loss_epoch=1.38e+4, lr_epoch=0.002, top_3_acc_epoch=0.476]
Epoch 10:  20%|██        | 2/10 [00:00<00:01,  5.56it/s, v_num=0, total_loss_step=1.32e+4, lr_step=0.002, top_3_acc_step=0.512, total_loss_epoch=1.38e+4, lr_epoch=0.002, top_3_acc_epoch=0.476]
Epoch 10:  30%|███       | 3/10 [00:00<00:01,  5.70it/s, v_num=0, total_loss_step=1.32e+4, lr_step=0.002, top_3_acc_step=0.512, total_loss_epoch=1.38e+4, lr_epoch=0.002, top_3_acc_epoch=0.476]
Epoch 10:  30%|███       | 3/10 [00:00<00:01,  5.70it/s, v_num=0, total_loss_step=1.32e+4, lr_step=0.002, top_3_acc_step=0.511, total_loss_epoch=1.38e+4, lr_epoch=0.002, top_3_acc_epoch=0.476]
Epoch 10:  40%|████      | 4/10 [00:00<00:01,  5.76it/s, v_num=0, total_loss_step=1.32e+4, lr_step=0.002, top_3_acc_step=0.511, total_loss_epoch=1.38e+4, lr_epoch=0.002, top_3_acc_epoch=0.476]
Epoch 10:  40%|████      | 4/10 [00:00<00:01,  5.76it/s, v_num=0, total_loss_step=1.32e+4, lr_step=0.002, top_3_acc_step=0.506, total_loss_epoch=1.38e+4, lr_epoch=0.002, top_3_acc_epoch=0.476]
Epoch 10:  50%|█████     | 5/10 [00:00<00:00,  5.72it/s, v_num=0, total_loss_step=1.32e+4, lr_step=0.002, top_3_acc_step=0.506, total_loss_epoch=1.38e+4, lr_epoch=0.002, top_3_acc_epoch=0.476]
Epoch 10:  50%|█████     | 5/10 [00:00<00:00,  5.72it/s, v_num=0, total_loss_step=1.31e+4, lr_step=0.002, top_3_acc_step=0.513, total_loss_epoch=1.38e+4, lr_epoch=0.002, top_3_acc_epoch=0.476]
Epoch 10:  60%|██████    | 6/10 [00:01<00:00,  5.76it/s, v_num=0, total_loss_step=1.31e+4, lr_step=0.002, top_3_acc_step=0.513, total_loss_epoch=1.38e+4, lr_epoch=0.002, top_3_acc_epoch=0.476]
Epoch 10:  60%|██████    | 6/10 [00:01<00:00,  5.76it/s, v_num=0, total_loss_step=1.32e+4, lr_step=0.002, top_3_acc_step=0.508, total_loss_epoch=1.38e+4, lr_epoch=0.002, top_3_acc_epoch=0.476]
Epoch 10:  70%|███████   | 7/10 [00:01<00:00,  5.79it/s, v_num=0, total_loss_step=1.32e+4, lr_step=0.002, top_3_acc_step=0.508, total_loss_epoch=1.38e+4, lr_epoch=0.002, top_3_acc_epoch=0.476]
Epoch 10:  70%|███████   | 7/10 [00:01<00:00,  5.79it/s, v_num=0, total_loss_step=1.32e+4, lr_step=0.002, top_3_acc_step=0.505, total_loss_epoch=1.38e+4, lr_epoch=0.002, top_3_acc_epoch=0.476]
Epoch 10:  80%|████████  | 8/10 [00:01<00:00,  5.81it/s, v_num=0, total_loss_step=1.32e+4, lr_step=0.002, top_3_acc_step=0.505, total_loss_epoch=1.38e+4, lr_epoch=0.002, top_3_acc_epoch=0.476]
Epoch 10:  80%|████████  | 8/10 [00:01<00:00,  5.81it/s, v_num=0, total_loss_step=1.31e+4, lr_step=0.002, top_3_acc_step=0.513, total_loss_epoch=1.38e+4, lr_epoch=0.002, top_3_acc_epoch=0.476]
Epoch 10:  90%|█████████ | 9/10 [00:01<00:00,  5.74it/s, v_num=0, total_loss_step=1.31e+4, lr_step=0.002, top_3_acc_step=0.513, total_loss_epoch=1.38e+4, lr_epoch=0.002, top_3_acc_epoch=0.476]
Epoch 10:  90%|█████████ | 9/10 [00:01<00:00,  5.74it/s, v_num=0, total_loss_step=1.31e+4, lr_step=0.002, top_3_acc_step=0.520, total_loss_epoch=1.38e+4, lr_epoch=0.002, top_3_acc_epoch=0.476]
Epoch 10: 100%|██████████| 10/10 [00:01<00:00,  6.19it/s, v_num=0, total_loss_step=1.31e+4, lr_step=0.002, top_3_acc_step=0.520, total_loss_epoch=1.38e+4, lr_epoch=0.002, top_3_acc_epoch=0.476]
Epoch 10: 100%|██████████| 10/10 [00:01<00:00,  6.18it/s, v_num=0, total_loss_step=1.08e+3, lr_step=0.002, top_3_acc_step=0.567, total_loss_epoch=1.38e+4, lr_epoch=0.002, top_3_acc_epoch=0.476]
Epoch 10: 100%|██████████| 10/10 [00:01<00:00,  6.18it/s, v_num=0, total_loss_step=1.08e+3, lr_step=0.002, top_3_acc_step=0.567, total_loss_epoch=1.31e+4, lr_epoch=0.002, top_3_acc_epoch=0.511]
Epoch 10:   0%|          | 0/10 [00:00<?, ?it/s, v_num=0, total_loss_step=1.08e+3, lr_step=0.002, top_3_acc_step=0.567, total_loss_epoch=1.31e+4, lr_epoch=0.002, top_3_acc_epoch=0.511]
Epoch 11:   0%|          | 0/10 [00:00<?, ?it/s, v_num=0, total_loss_step=1.08e+3, lr_step=0.002, top_3_acc_step=0.567, total_loss_epoch=1.31e+4, lr_epoch=0.002, top_3_acc_epoch=0.511]
Epoch 11:  10%|█         | 1/10 [00:00<00:01,  5.43it/s, v_num=0, total_loss_step=1.08e+3, lr_step=0.002, top_3_acc_step=0.567, total_loss_epoch=1.31e+4, lr_epoch=0.002, top_3_acc_epoch=0.511]
Epoch 11:  10%|█         | 1/10 [00:00<00:01,  5.42it/s, v_num=0, total_loss_step=1.17e+4, lr_step=0.002, top_3_acc_step=0.587, total_loss_epoch=1.31e+4, lr_epoch=0.002, top_3_acc_epoch=0.511]
Epoch 11:  20%|██        | 2/10 [00:00<00:01,  5.62it/s, v_num=0, total_loss_step=1.17e+4, lr_step=0.002, top_3_acc_step=0.587, total_loss_epoch=1.31e+4, lr_epoch=0.002, top_3_acc_epoch=0.511]
Epoch 11:  20%|██        | 2/10 [00:00<00:01,  5.61it/s, v_num=0, total_loss_step=1.17e+4, lr_step=0.002, top_3_acc_step=0.592, total_loss_epoch=1.31e+4, lr_epoch=0.002, top_3_acc_epoch=0.511]
Epoch 11:  30%|███       | 3/10 [00:00<00:01,  5.73it/s, v_num=0, total_loss_step=1.17e+4, lr_step=0.002, top_3_acc_step=0.592, total_loss_epoch=1.31e+4, lr_epoch=0.002, top_3_acc_epoch=0.511]
Epoch 11:  30%|███       | 3/10 [00:00<00:01,  5.73it/s, v_num=0, total_loss_step=1.17e+4, lr_step=0.002, top_3_acc_step=0.578, total_loss_epoch=1.31e+4, lr_epoch=0.002, top_3_acc_epoch=0.511]
Epoch 11:  40%|████      | 4/10 [00:00<00:01,  5.80it/s, v_num=0, total_loss_step=1.17e+4, lr_step=0.002, top_3_acc_step=0.578, total_loss_epoch=1.31e+4, lr_epoch=0.002, top_3_acc_epoch=0.511]
Epoch 11:  40%|████      | 4/10 [00:00<00:01,  5.79it/s, v_num=0, total_loss_step=1.17e+4, lr_step=0.002, top_3_acc_step=0.578, total_loss_epoch=1.31e+4, lr_epoch=0.002, top_3_acc_epoch=0.511]
Epoch 11:  50%|█████     | 5/10 [00:00<00:00,  5.65it/s, v_num=0, total_loss_step=1.17e+4, lr_step=0.002, top_3_acc_step=0.578, total_loss_epoch=1.31e+4, lr_epoch=0.002, top_3_acc_epoch=0.511]
Epoch 11:  50%|█████     | 5/10 [00:00<00:00,  5.65it/s, v_num=0, total_loss_step=1.17e+4, lr_step=0.002, top_3_acc_step=0.589, total_loss_epoch=1.31e+4, lr_epoch=0.002, top_3_acc_epoch=0.511]
Epoch 11:  60%|██████    | 6/10 [00:01<00:00,  5.71it/s, v_num=0, total_loss_step=1.17e+4, lr_step=0.002, top_3_acc_step=0.589, total_loss_epoch=1.31e+4, lr_epoch=0.002, top_3_acc_epoch=0.511]
Epoch 11:  60%|██████    | 6/10 [00:01<00:00,  5.70it/s, v_num=0, total_loss_step=1.02e+4, lr_step=0.002, top_3_acc_step=0.661, total_loss_epoch=1.31e+4, lr_epoch=0.002, top_3_acc_epoch=0.511]
Epoch 11:  70%|███████   | 7/10 [00:01<00:00,  5.74it/s, v_num=0, total_loss_step=1.02e+4, lr_step=0.002, top_3_acc_step=0.661, total_loss_epoch=1.31e+4, lr_epoch=0.002, top_3_acc_epoch=0.511]
Epoch 11:  70%|███████   | 7/10 [00:01<00:00,  5.74it/s, v_num=0, total_loss_step=1.02e+4, lr_step=0.002, top_3_acc_step=0.658, total_loss_epoch=1.31e+4, lr_epoch=0.002, top_3_acc_epoch=0.511]
Epoch 11:  80%|████████  | 8/10 [00:01<00:00,  5.77it/s, v_num=0, total_loss_step=1.02e+4, lr_step=0.002, top_3_acc_step=0.658, total_loss_epoch=1.31e+4, lr_epoch=0.002, top_3_acc_epoch=0.511]
Epoch 11:  80%|████████  | 8/10 [00:01<00:00,  5.77it/s, v_num=0, total_loss_step=1.02e+4, lr_step=0.002, top_3_acc_step=0.668, total_loss_epoch=1.31e+4, lr_epoch=0.002, top_3_acc_epoch=0.511]
Epoch 11:  90%|█████████ | 9/10 [00:01<00:00,  5.68it/s, v_num=0, total_loss_step=1.02e+4, lr_step=0.002, top_3_acc_step=0.668, total_loss_epoch=1.31e+4, lr_epoch=0.002, top_3_acc_epoch=0.511]
Epoch 11:  90%|█████████ | 9/10 [00:01<00:00,  5.68it/s, v_num=0, total_loss_step=1.02e+4, lr_step=0.002, top_3_acc_step=0.665, total_loss_epoch=1.31e+4, lr_epoch=0.002, top_3_acc_epoch=0.511]
Epoch 11: 100%|██████████| 10/10 [00:01<00:00,  6.12it/s, v_num=0, total_loss_step=1.02e+4, lr_step=0.002, top_3_acc_step=0.665, total_loss_epoch=1.31e+4, lr_epoch=0.002, top_3_acc_epoch=0.511]
Epoch 11: 100%|██████████| 10/10 [00:01<00:00,  6.12it/s, v_num=0, total_loss_step=783.0, lr_step=0.002, top_3_acc_step=0.752, total_loss_epoch=1.31e+4, lr_epoch=0.002, top_3_acc_epoch=0.511]
Epoch 11: 100%|██████████| 10/10 [00:01<00:00,  6.12it/s, v_num=0, total_loss_step=783.0, lr_step=0.002, top_3_acc_step=0.752, total_loss_epoch=1.09e+4, lr_epoch=0.002, top_3_acc_epoch=0.621]
Epoch 11:   0%|          | 0/10 [00:00<?, ?it/s, v_num=0, total_loss_step=783.0, lr_step=0.002, top_3_acc_step=0.752, total_loss_epoch=1.09e+4, lr_epoch=0.002, top_3_acc_epoch=0.621]
Epoch 12:   0%|          | 0/10 [00:00<?, ?it/s, v_num=0, total_loss_step=783.0, lr_step=0.002, top_3_acc_step=0.752, total_loss_epoch=1.09e+4, lr_epoch=0.002, top_3_acc_epoch=0.621]
Epoch 12:  10%|█         | 1/10 [00:00<00:01,  5.47it/s, v_num=0, total_loss_step=783.0, lr_step=0.002, top_3_acc_step=0.752, total_loss_epoch=1.09e+4, lr_epoch=0.002, top_3_acc_epoch=0.621]
Epoch 12:  10%|█         | 1/10 [00:00<00:01,  5.45it/s, v_num=0, total_loss_step=8.83e+3, lr_step=0.002, top_3_acc_step=0.730, total_loss_epoch=1.09e+4, lr_epoch=0.002, top_3_acc_epoch=0.621]
Epoch 12:  20%|██        | 2/10 [00:00<00:01,  5.69it/s, v_num=0, total_loss_step=8.83e+3, lr_step=0.002, top_3_acc_step=0.730, total_loss_epoch=1.09e+4, lr_epoch=0.002, top_3_acc_epoch=0.621]
Epoch 12:  20%|██        | 2/10 [00:00<00:01,  5.68it/s, v_num=0, total_loss_step=8.75e+3, lr_step=0.002, top_3_acc_step=0.739, total_loss_epoch=1.09e+4, lr_epoch=0.002, top_3_acc_epoch=0.621]
Epoch 12:  30%|███       | 3/10 [00:00<00:01,  5.75it/s, v_num=0, total_loss_step=8.75e+3, lr_step=0.002, top_3_acc_step=0.739, total_loss_epoch=1.09e+4, lr_epoch=0.002, top_3_acc_epoch=0.621]
Epoch 12:  30%|███       | 3/10 [00:00<00:01,  5.75it/s, v_num=0, total_loss_step=8.78e+3, lr_step=0.002, top_3_acc_step=0.732, total_loss_epoch=1.09e+4, lr_epoch=0.002, top_3_acc_epoch=0.621]
Epoch 12:  40%|████      | 4/10 [00:00<00:01,  5.81it/s, v_num=0, total_loss_step=8.78e+3, lr_step=0.002, top_3_acc_step=0.732, total_loss_epoch=1.09e+4, lr_epoch=0.002, top_3_acc_epoch=0.621]
Epoch 12:  40%|████      | 4/10 [00:00<00:01,  5.80it/s, v_num=0, total_loss_step=8.71e+3, lr_step=0.002, top_3_acc_step=0.734, total_loss_epoch=1.09e+4, lr_epoch=0.002, top_3_acc_epoch=0.621]
Epoch 12:  50%|█████     | 5/10 [00:00<00:00,  5.68it/s, v_num=0, total_loss_step=8.71e+3, lr_step=0.002, top_3_acc_step=0.734, total_loss_epoch=1.09e+4, lr_epoch=0.002, top_3_acc_epoch=0.621]
Epoch 12:  50%|█████     | 5/10 [00:00<00:00,  5.68it/s, v_num=0, total_loss_step=8.76e+3, lr_step=0.002, top_3_acc_step=0.736, total_loss_epoch=1.09e+4, lr_epoch=0.002, top_3_acc_epoch=0.621]
Epoch 12:  60%|██████    | 6/10 [00:01<00:00,  5.74it/s, v_num=0, total_loss_step=8.76e+3, lr_step=0.002, top_3_acc_step=0.736, total_loss_epoch=1.09e+4, lr_epoch=0.002, top_3_acc_epoch=0.621]
Epoch 12:  60%|██████    | 6/10 [00:01<00:00,  5.74it/s, v_num=0, total_loss_step=7.45e+3, lr_step=0.002, top_3_acc_step=0.797, total_loss_epoch=1.09e+4, lr_epoch=0.002, top_3_acc_epoch=0.621]
Epoch 12:  70%|███████   | 7/10 [00:01<00:00,  5.76it/s, v_num=0, total_loss_step=7.45e+3, lr_step=0.002, top_3_acc_step=0.797, total_loss_epoch=1.09e+4, lr_epoch=0.002, top_3_acc_epoch=0.621]
Epoch 12:  70%|███████   | 7/10 [00:01<00:00,  5.76it/s, v_num=0, total_loss_step=7.4e+3, lr_step=0.002, top_3_acc_step=0.794, total_loss_epoch=1.09e+4, lr_epoch=0.002, top_3_acc_epoch=0.621]
Epoch 12:  80%|████████  | 8/10 [00:01<00:00,  5.78it/s, v_num=0, total_loss_step=7.4e+3, lr_step=0.002, top_3_acc_step=0.794, total_loss_epoch=1.09e+4, lr_epoch=0.002, top_3_acc_epoch=0.621]
Epoch 12:  80%|████████  | 8/10 [00:01<00:00,  5.78it/s, v_num=0, total_loss_step=7.47e+3, lr_step=0.002, top_3_acc_step=0.792, total_loss_epoch=1.09e+4, lr_epoch=0.002, top_3_acc_epoch=0.621]
Epoch 12:  90%|█████████ | 9/10 [00:01<00:00,  5.72it/s, v_num=0, total_loss_step=7.47e+3, lr_step=0.002, top_3_acc_step=0.792, total_loss_epoch=1.09e+4, lr_epoch=0.002, top_3_acc_epoch=0.621]
Epoch 12:  90%|█████████ | 9/10 [00:01<00:00,  5.71it/s, v_num=0, total_loss_step=7.4e+3, lr_step=0.002, top_3_acc_step=0.798, total_loss_epoch=1.09e+4, lr_epoch=0.002, top_3_acc_epoch=0.621]
Epoch 12: 100%|██████████| 10/10 [00:01<00:00,  6.15it/s, v_num=0, total_loss_step=7.4e+3, lr_step=0.002, top_3_acc_step=0.798, total_loss_epoch=1.09e+4, lr_epoch=0.002, top_3_acc_epoch=0.621]
Epoch 12: 100%|██████████| 10/10 [00:01<00:00,  6.15it/s, v_num=0, total_loss_step=570.0, lr_step=0.002, top_3_acc_step=0.843, total_loss_epoch=1.09e+4, lr_epoch=0.002, top_3_acc_epoch=0.621]
Epoch 12: 100%|██████████| 10/10 [00:01<00:00,  6.14it/s, v_num=0, total_loss_step=570.0, lr_step=0.002, top_3_acc_step=0.843, total_loss_epoch=8.1e+3, lr_epoch=0.002, top_3_acc_epoch=0.762]
Epoch 12:   0%|          | 0/10 [00:00<?, ?it/s, v_num=0, total_loss_step=570.0, lr_step=0.002, top_3_acc_step=0.843, total_loss_epoch=8.1e+3, lr_epoch=0.002, top_3_acc_epoch=0.762]
Epoch 13:   0%|          | 0/10 [00:00<?, ?it/s, v_num=0, total_loss_step=570.0, lr_step=0.002, top_3_acc_step=0.843, total_loss_epoch=8.1e+3, lr_epoch=0.002, top_3_acc_epoch=0.762]
Epoch 13:  10%|█         | 1/10 [00:00<00:01,  5.31it/s, v_num=0, total_loss_step=570.0, lr_step=0.002, top_3_acc_step=0.843, total_loss_epoch=8.1e+3, lr_epoch=0.002, top_3_acc_epoch=0.762]
Epoch 13:  10%|█         | 1/10 [00:00<00:01,  5.29it/s, v_num=0, total_loss_step=6.31e+3, lr_step=0.002, top_3_acc_step=0.851, total_loss_epoch=8.1e+3, lr_epoch=0.002, top_3_acc_epoch=0.762]
Epoch 13:  20%|██        | 2/10 [00:00<00:01,  5.62it/s, v_num=0, total_loss_step=6.31e+3, lr_step=0.002, top_3_acc_step=0.851, total_loss_epoch=8.1e+3, lr_epoch=0.002, top_3_acc_epoch=0.762]
Epoch 13:  20%|██        | 2/10 [00:00<00:01,  5.61it/s, v_num=0, total_loss_step=6.27e+3, lr_step=0.002, top_3_acc_step=0.857, total_loss_epoch=8.1e+3, lr_epoch=0.002, top_3_acc_epoch=0.762]
Epoch 13:  30%|███       | 3/10 [00:00<00:01,  5.70it/s, v_num=0, total_loss_step=6.27e+3, lr_step=0.002, top_3_acc_step=0.857, total_loss_epoch=8.1e+3, lr_epoch=0.002, top_3_acc_epoch=0.762]
Epoch 13:  30%|███       | 3/10 [00:00<00:01,  5.69it/s, v_num=0, total_loss_step=6.28e+3, lr_step=0.002, top_3_acc_step=0.856, total_loss_epoch=8.1e+3, lr_epoch=0.002, top_3_acc_epoch=0.762]
Epoch 13:  40%|████      | 4/10 [00:00<00:01,  5.76it/s, v_num=0, total_loss_step=6.28e+3, lr_step=0.002, top_3_acc_step=0.856, total_loss_epoch=8.1e+3, lr_epoch=0.002, top_3_acc_epoch=0.762]
Epoch 13:  40%|████      | 4/10 [00:00<00:01,  5.76it/s, v_num=0, total_loss_step=6.24e+3, lr_step=0.002, top_3_acc_step=0.856, total_loss_epoch=8.1e+3, lr_epoch=0.002, top_3_acc_epoch=0.762]
Epoch 13:  50%|█████     | 5/10 [00:00<00:00,  5.64it/s, v_num=0, total_loss_step=6.24e+3, lr_step=0.002, top_3_acc_step=0.856, total_loss_epoch=8.1e+3, lr_epoch=0.002, top_3_acc_epoch=0.762]
Epoch 13:  50%|█████     | 5/10 [00:00<00:00,  5.64it/s, v_num=0, total_loss_step=6.29e+3, lr_step=0.002, top_3_acc_step=0.854, total_loss_epoch=8.1e+3, lr_epoch=0.002, top_3_acc_epoch=0.762]
Epoch 13:  60%|██████    | 6/10 [00:01<00:00,  5.70it/s, v_num=0, total_loss_step=6.29e+3, lr_step=0.002, top_3_acc_step=0.854, total_loss_epoch=8.1e+3, lr_epoch=0.002, top_3_acc_epoch=0.762]
Epoch 13:  60%|██████    | 6/10 [00:01<00:00,  5.70it/s, v_num=0, total_loss_step=5.24e+3, lr_step=0.002, top_3_acc_step=0.901, total_loss_epoch=8.1e+3, lr_epoch=0.002, top_3_acc_epoch=0.762]
Epoch 13:  70%|███████   | 7/10 [00:01<00:00,  5.72it/s, v_num=0, total_loss_step=5.24e+3, lr_step=0.002, top_3_acc_step=0.901, total_loss_epoch=8.1e+3, lr_epoch=0.002, top_3_acc_epoch=0.762]
Epoch 13:  70%|███████   | 7/10 [00:01<00:00,  5.72it/s, v_num=0, total_loss_step=5.18e+3, lr_step=0.002, top_3_acc_step=0.910, total_loss_epoch=8.1e+3, lr_epoch=0.002, top_3_acc_epoch=0.762]
Epoch 13:  80%|████████  | 8/10 [00:01<00:00,  5.74it/s, v_num=0, total_loss_step=5.18e+3, lr_step=0.002, top_3_acc_step=0.910, total_loss_epoch=8.1e+3, lr_epoch=0.002, top_3_acc_epoch=0.762]
Epoch 13:  80%|████████  | 8/10 [00:01<00:00,  5.74it/s, v_num=0, total_loss_step=5.23e+3, lr_step=0.002, top_3_acc_step=0.904, total_loss_epoch=8.1e+3, lr_epoch=0.002, top_3_acc_epoch=0.762]
Epoch 13:  90%|█████████ | 9/10 [00:01<00:00,  5.66it/s, v_num=0, total_loss_step=5.23e+3, lr_step=0.002, top_3_acc_step=0.904, total_loss_epoch=8.1e+3, lr_epoch=0.002, top_3_acc_epoch=0.762]
Epoch 13:  90%|█████████ | 9/10 [00:01<00:00,  5.66it/s, v_num=0, total_loss_step=5.26e+3, lr_step=0.002, top_3_acc_step=0.899, total_loss_epoch=8.1e+3, lr_epoch=0.002, top_3_acc_epoch=0.762]
Epoch 13: 100%|██████████| 10/10 [00:01<00:00,  6.09it/s, v_num=0, total_loss_step=5.26e+3, lr_step=0.002, top_3_acc_step=0.899, total_loss_epoch=8.1e+3, lr_epoch=0.002, top_3_acc_epoch=0.762]
Epoch 13: 100%|██████████| 10/10 [00:01<00:00,  6.09it/s, v_num=0, total_loss_step=395.0, lr_step=0.002, top_3_acc_step=0.929, total_loss_epoch=8.1e+3, lr_epoch=0.002, top_3_acc_epoch=0.762]
Epoch 13: 100%|██████████| 10/10 [00:01<00:00,  6.09it/s, v_num=0, total_loss_step=395.0, lr_step=0.002, top_3_acc_step=0.929, total_loss_epoch=5.76e+3, lr_epoch=0.002, top_3_acc_epoch=0.877]
Epoch 13:   0%|          | 0/10 [00:00<?, ?it/s, v_num=0, total_loss_step=395.0, lr_step=0.002, top_3_acc_step=0.929, total_loss_epoch=5.76e+3, lr_epoch=0.002, top_3_acc_epoch=0.877]
Epoch 14:   0%|          | 0/10 [00:00<?, ?it/s, v_num=0, total_loss_step=395.0, lr_step=0.002, top_3_acc_step=0.929, total_loss_epoch=5.76e+3, lr_epoch=0.002, top_3_acc_epoch=0.877]
Epoch 14:  10%|█         | 1/10 [00:00<00:01,  5.43it/s, v_num=0, total_loss_step=395.0, lr_step=0.002, top_3_acc_step=0.929, total_loss_epoch=5.76e+3, lr_epoch=0.002, top_3_acc_epoch=0.877]
Epoch 14:  10%|█         | 1/10 [00:00<00:01,  5.41it/s, v_num=0, total_loss_step=4.31e+3, lr_step=0.002, top_3_acc_step=0.943, total_loss_epoch=5.76e+3, lr_epoch=0.002, top_3_acc_epoch=0.877]
Epoch 14:  20%|██        | 2/10 [00:00<00:01,  5.65it/s, v_num=0, total_loss_step=4.31e+3, lr_step=0.002, top_3_acc_step=0.943, total_loss_epoch=5.76e+3, lr_epoch=0.002, top_3_acc_epoch=0.877]
Epoch 14:  20%|██        | 2/10 [00:00<00:01,  5.64it/s, v_num=0, total_loss_step=4.3e+3, lr_step=0.002, top_3_acc_step=0.943, total_loss_epoch=5.76e+3, lr_epoch=0.002, top_3_acc_epoch=0.877]
Epoch 14:  30%|███       | 3/10 [00:00<00:01,  5.73it/s, v_num=0, total_loss_step=4.3e+3, lr_step=0.002, top_3_acc_step=0.943, total_loss_epoch=5.76e+3, lr_epoch=0.002, top_3_acc_epoch=0.877]
Epoch 14:  30%|███       | 3/10 [00:00<00:01,  5.72it/s, v_num=0, total_loss_step=4.28e+3, lr_step=0.002, top_3_acc_step=0.945, total_loss_epoch=5.76e+3, lr_epoch=0.002, top_3_acc_epoch=0.877]
Epoch 14:  40%|████      | 4/10 [00:00<00:01,  5.77it/s, v_num=0, total_loss_step=4.28e+3, lr_step=0.002, top_3_acc_step=0.945, total_loss_epoch=5.76e+3, lr_epoch=0.002, top_3_acc_epoch=0.877]
Epoch 14:  40%|████      | 4/10 [00:00<00:01,  5.76it/s, v_num=0, total_loss_step=4.32e+3, lr_step=0.002, top_3_acc_step=0.938, total_loss_epoch=5.76e+3, lr_epoch=0.002, top_3_acc_epoch=0.877]
Epoch 14:  50%|█████     | 5/10 [00:00<00:00,  5.62it/s, v_num=0, total_loss_step=4.32e+3, lr_step=0.002, top_3_acc_step=0.938, total_loss_epoch=5.76e+3, lr_epoch=0.002, top_3_acc_epoch=0.877]
Epoch 14:  50%|█████     | 5/10 [00:00<00:00,  5.61it/s, v_num=0, total_loss_step=4.31e+3, lr_step=0.002, top_3_acc_step=0.944, total_loss_epoch=5.76e+3, lr_epoch=0.002, top_3_acc_epoch=0.877]
Epoch 14:  60%|██████    | 6/10 [00:01<00:00,  5.68it/s, v_num=0, total_loss_step=4.31e+3, lr_step=0.002, top_3_acc_step=0.944, total_loss_epoch=5.76e+3, lr_epoch=0.002, top_3_acc_epoch=0.877]
Epoch 14:  60%|██████    | 6/10 [00:01<00:00,  5.68it/s, v_num=0, total_loss_step=3.49e+3, lr_step=0.002, top_3_acc_step=0.965, total_loss_epoch=5.76e+3, lr_epoch=0.002, top_3_acc_epoch=0.877]
Epoch 14:  70%|███████   | 7/10 [00:01<00:00,  5.72it/s, v_num=0, total_loss_step=3.49e+3, lr_step=0.002, top_3_acc_step=0.965, total_loss_epoch=5.76e+3, lr_epoch=0.002, top_3_acc_epoch=0.877]
Epoch 14:  70%|███████   | 7/10 [00:01<00:00,  5.72it/s, v_num=0, total_loss_step=3.44e+3, lr_step=0.002, top_3_acc_step=0.967, total_loss_epoch=5.76e+3, lr_epoch=0.002, top_3_acc_epoch=0.877]
Epoch 14:  80%|████████  | 8/10 [00:01<00:00,  5.76it/s, v_num=0, total_loss_step=3.44e+3, lr_step=0.002, top_3_acc_step=0.967, total_loss_epoch=5.76e+3, lr_epoch=0.002, top_3_acc_epoch=0.877]
Epoch 14:  80%|████████  | 8/10 [00:01<00:00,  5.75it/s, v_num=0, total_loss_step=3.41e+3, lr_step=0.002, top_3_acc_step=0.971, total_loss_epoch=5.76e+3, lr_epoch=0.002, top_3_acc_epoch=0.877]
Epoch 14:  90%|█████████ | 9/10 [00:01<00:00,  5.70it/s, v_num=0, total_loss_step=3.41e+3, lr_step=0.002, top_3_acc_step=0.971, total_loss_epoch=5.76e+3, lr_epoch=0.002, top_3_acc_epoch=0.877]
Epoch 14:  90%|█████████ | 9/10 [00:01<00:00,  5.69it/s, v_num=0, total_loss_step=3.44e+3, lr_step=0.002, top_3_acc_step=0.971, total_loss_epoch=5.76e+3, lr_epoch=0.002, top_3_acc_epoch=0.877]
Epoch 14: 100%|██████████| 10/10 [00:01<00:00,  6.13it/s, v_num=0, total_loss_step=3.44e+3, lr_step=0.002, top_3_acc_step=0.971, total_loss_epoch=5.76e+3, lr_epoch=0.002, top_3_acc_epoch=0.877]
Epoch 14: 100%|██████████| 10/10 [00:01<00:00,  6.13it/s, v_num=0, total_loss_step=261.0, lr_step=0.002, top_3_acc_step=0.980, total_loss_epoch=5.76e+3, lr_epoch=0.002, top_3_acc_epoch=0.877]
Epoch 14: 100%|██████████| 10/10 [00:01<00:00,  6.12it/s, v_num=0, total_loss_step=261.0, lr_step=0.002, top_3_acc_step=0.980, total_loss_epoch=3.89e+3, lr_epoch=0.002, top_3_acc_epoch=0.954]`Trainer.fit` stopped: `max_epochs=15` reached.
Epoch 14: 100%|██████████| 10/10 [00:03<00:00,  2.80it/s, v_num=0, total_loss_step=261.0, lr_step=0.002, top_3_acc_step=0.980, total_loss_epoch=3.89e+3, lr_epoch=0.002, top_3_acc_epoch=0.954]
('GPT训练完成', {'__type__': 'update', 'visible': True}, {'__type__': 'update', 'visible': False})
Files in the zip archive:
input.wav
log.txt
asr_opt/denoise_opt.list
denoise_opt/origin.mp3_0000000000_0000508480.wav
351b84a7-8622-4c51-a803-ae848d62d158_e8_s80.pth
351b84a7-8622-4c51-a803-ae848d62d158-e15.ckpt
Created zip file: 351b84a7-8622-4c51-a803-ae848d62d158.zip
Version Details
Version ID: d501141112f7a2fc5223942c6803dcfb8395559a01a507c682f74c775772d1f4
Version Created: December 22, 2025
Run on Replicate →