Mozilla DeepSpeech 中的“时代测试 [数字]”是什么意思？

Question

Mozilla DeepSpeech 中的“Test of Epoch [number]”是什么意思？

在下面的例子中，它说Test of Epoch 77263，尽管我的理解应该只有1个纪元，因为我给了--display_step 1 --limit_train 1 --limit_dev 1 --limit_test 1 --early_stop False --epoch 1作为参数：

dernoncourt@ilcomp:~/asr/DeepSpeech$ ./DeepSpeech.py --train_files data/common-voice-v1/cv-valid-train.csv,data/common-voice-v1/cv-other-train.csv --dev_files data/common-voice-v1/cv-valid-dev.csv --test_files data/common-voice-v1/cv-valid-test.csv --decoder_library_path /asr/DeepSpeech/libctc_decoder_with_kenlm.so --fulltrace True --display_step 1  --limit_train 1  --limit_dev 1  --limit_test 1 --early_stop False --epoch 1
W Parameter --validation_step needs to be >0 for early stopping to work
I Test of Epoch 77263 - WER: 1.000000, loss: 60.50202560424805, mean edit distance: 0.894737
I --------------------------------------------------------------------------------
I WER: 1.000000, loss: 58.900837, mean edit distance: 0.894737
I  - src: "how do you like her"
I  - res: "i "
I --------------------------------------------------------------------------------
I WER: 1.000000, loss: 60.517113, mean edit distance: 0.894737
I  - src: "how do you like her"
I  - res: "i "
I --------------------------------------------------------------------------------
I WER: 1.000000, loss: 60.668221, mean edit distance: 0.894737
I  - src: "how do you like her"
I  - res: "i "
I --------------------------------------------------------------------------------
I WER: 1.000000, loss: 61.921925, mean edit distance: 0.894737
I  - src: "how do you like her"
I  - res: "i "
I --------------------------------------------------------------------------------

Answer 1

Explanation by Tilman Kamp:

这实际上不是错误，因为当前纪元是根据基础计算的 您当前的参数和快照持久的全局 步数。仔细看看这段摘录：

# Number of GPUs per worker - fixed for now by local reality or cluster setup
gpus_per_worker = len(available_devices)

# Number of batches processed per job per worker
batches_per_job  = gpus_per_worker * max(1, FLAGS.iters_per_worker)

# Number of batches per global step
batches_per_step = gpus_per_worker * max(1, FLAGS.replicas_to_agg)

# Number of global steps per epoch - to be at least 1
steps_per_epoch = max(1, model_feeder.train.total_batches // batches_per_step)

# The start epoch of our training
# Number of GPUs per worker - fixed for now by local reality or cluster setup
gpus_per_worker = len(available_devices)

# Number of batches processed per job per worker
batches_per_job  = gpus_per_worker * max(1, FLAGS.iters_per_worker)

# Number of batches per global step
batches_per_step = gpus_per_worker * max(1, FLAGS.replicas_to_agg)

# Number of global steps per epoch - to be at least 1
steps_per_epoch = max(1, model_feeder.train.total_batches // batches_per_step)

# The start epoch of our training
self._epoch = step // steps_per_epoch

所以发生的情况是您在训练期间的 set-size 与 您当前的设置大小。因此奇怪的纪元数。

简化示例（不混淆批量大小）：如果您曾经受过训练 1000 个样本训练集的 5 个 epoch，你得到了 5000 个“全局步骤” （在您的快照中保留为数字）。本次培训后你 将命令行参数更改为一组大小 1（您的 --limit_* 参数）。 “突然”你会显示 5000 纪元，因为 5000 全局步骤意味着应用大小为 1 5000 次的数据集。

带走：使用--checkpoint_dir 参数来避免此类问题。