another trial

config.json

@@ -70,7 +70,7 @@
   "num_conv_pos_embeddings": 128,
   "num_feat_extract_layers": 7,
   "num_hidden_layers": 24,
-  "pad_token_id": 259,
+  "pad_token_id": 45,
   "transformers_version": "4.6.0.dev0",
-  "vocab_size": 260
+  "vocab_size": 46
 }

optimizer.pt

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:5bfc6508f4174c9f6359349b9687c28ddab0e09aa92126ab0fa9bcf89e6bb4fb
-size 2492208775

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:81a8acfe830487f7ddb00a2c4c6db079bdc50bf9d4ba7f6fd18f42934341b452
-size 1262999831
+oid sha256:a63fff73f92863b73d9fa36f173b126d733cdff623e49de63158bfb66b1dc8d4
+size 1262122455

scheduler.pt

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:057999ec38e996dec146de5e28f3aaf30dfbc86f0c54eee12d611f5e3b7423e8
-size 623

special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "[UNK]", "pad_token": "[PAD]"}

tokenizer_config.json

@@ -0,0 +1 @@
+{"unk_token": "[UNK]", "bos_token": "<s>", "eos_token": "</s>", "pad_token": "[PAD]", "do_lower_case": false, "word_delimiter_token": "|"}

trainer_state.json

@@ -1,310 +0,0 @@
-{
-  "best_metric": 4.255084991455078,
-  "best_model_checkpoint": "./xlsr-demo/checkpoint-80",
-  "epoch": 28.0,
-  "global_step": 84,
-  "is_hyper_param_search": false,
-  "is_local_process_zero": true,
-  "is_world_process_zero": true,
-  "log_history": [
-    {
-      "epoch": 1.33,
-      "learning_rate": 5.9999999999999995e-05,
-      "loss": 45.1762,
-      "step": 4
-    },
-    {
-      "epoch": 1.33,
-      "eval_loss": 32.95661544799805,
-      "eval_runtime": 0.6944,
-      "eval_samples_per_second": 14.402,
-      "eval_wer": 1.0,
-      "step": 4
-    },
-    {
-      "epoch": 2.67,
-      "learning_rate": 0.00023999999999999998,
-      "loss": 48.4094,
-      "step": 8
-    },
-    {
-      "epoch": 2.67,
-      "eval_loss": 32.567665100097656,
-      "eval_runtime": 0.5168,
-      "eval_samples_per_second": 19.351,
-      "eval_wer": 1.0,
-      "step": 8
-    },
-    {
-      "epoch": 4.0,
-      "learning_rate": 0.00028941176470588233,
-      "loss": 41.452,
-      "step": 12
-    },
-    {
-      "epoch": 4.0,
-      "eval_loss": 31.335033416748047,
-      "eval_runtime": 0.5187,
-      "eval_samples_per_second": 19.279,
-      "eval_wer": 1.0,
-      "step": 12
-    },
-    {
-      "epoch": 5.33,
-      "learning_rate": 0.0002788235294117647,
-      "loss": 43.7158,
-      "step": 16
-    },
-    {
-      "epoch": 5.33,
-      "eval_loss": 29.559152603149414,
-      "eval_runtime": 0.5095,
-      "eval_samples_per_second": 19.627,
-      "eval_wer": 1.0,
-      "step": 16
-    },
-    {
-      "epoch": 6.67,
-      "learning_rate": 0.00026470588235294115,
-      "loss": 30.5266,
-      "step": 20
-    },
-    {
-      "epoch": 6.67,
-      "eval_loss": 18.15114974975586,
-      "eval_runtime": 0.5292,
-      "eval_samples_per_second": 18.896,
-      "eval_wer": 1.0,
-      "step": 20
-    },
-    {
-      "epoch": 8.0,
-      "learning_rate": 0.0002505882352941176,
-      "loss": 27.451,
-      "step": 24
-    },
-    {
-      "epoch": 8.0,
-      "eval_loss": 13.7584810256958,
-      "eval_runtime": 0.5073,
-      "eval_samples_per_second": 19.713,
-      "eval_wer": 1.0,
-      "step": 24
-    },
-    {
-      "epoch": 9.33,
-      "learning_rate": 0.0002364705882352941,
-      "loss": 19.4205,
-      "step": 28
-    },
-    {
-      "epoch": 9.33,
-      "eval_loss": 10.164986610412598,
-      "eval_runtime": 0.5127,
-      "eval_samples_per_second": 19.503,
-      "eval_wer": 1.0,
-      "step": 28
-    },
-    {
-      "epoch": 10.67,
-      "learning_rate": 0.00022235294117647057,
-      "loss": 13.2132,
-      "step": 32
-    },
-    {
-      "epoch": 10.67,
-      "eval_loss": 7.898101806640625,
-      "eval_runtime": 0.5031,
-      "eval_samples_per_second": 19.877,
-      "eval_wer": 1.0,
-      "step": 32
-    },
-    {
-      "epoch": 12.0,
-      "learning_rate": 0.00020823529411764704,
-      "loss": 9.3548,
-      "step": 36
-    },
-    {
-      "epoch": 12.0,
-      "eval_loss": 6.538975715637207,
-      "eval_runtime": 0.5312,
-      "eval_samples_per_second": 18.825,
-      "eval_wer": 1.0,
-      "step": 36
-    },
-    {
-      "epoch": 13.33,
-      "learning_rate": 0.0001941176470588235,
-      "loss": 8.7253,
-      "step": 40
-    },
-    {
-      "epoch": 13.33,
-      "eval_loss": 5.714229583740234,
-      "eval_runtime": 0.4983,
-      "eval_samples_per_second": 20.07,
-      "eval_wer": 1.0,
-      "step": 40
-    },
-    {
-      "epoch": 14.67,
-      "learning_rate": 0.00017999999999999998,
-      "loss": 6.8295,
-      "step": 44
-    },
-    {
-      "epoch": 14.67,
-      "eval_loss": 5.214608192443848,
-      "eval_runtime": 0.5112,
-      "eval_samples_per_second": 19.561,
-      "eval_wer": 1.0,
-      "step": 44
-    },
-    {
-      "epoch": 16.0,
-      "learning_rate": 0.00016588235294117646,
-      "loss": 5.9458,
-      "step": 48
-    },
-    {
-      "epoch": 16.0,
-      "eval_loss": 4.887683868408203,
-      "eval_runtime": 0.5156,
-      "eval_samples_per_second": 19.394,
-      "eval_wer": 1.0,
-      "step": 48
-    },
-    {
-      "epoch": 17.33,
-      "learning_rate": 0.00015176470588235293,
-      "loss": 5.4214,
-      "step": 52
-    },
-    {
-      "epoch": 17.33,
-      "eval_loss": 4.679065704345703,
-      "eval_runtime": 1.0634,
-      "eval_samples_per_second": 9.403,
-      "eval_wer": 1.0,
-      "step": 52
-    },
-    {
-      "epoch": 18.67,
-      "learning_rate": 0.0001376470588235294,
-      "loss": 5.8716,
-      "step": 56
-    },
-    {
-      "epoch": 18.67,
-      "eval_loss": 4.548101425170898,
-      "eval_runtime": 0.4987,
-      "eval_samples_per_second": 20.051,
-      "eval_wer": 1.0,
-      "step": 56
-    },
-    {
-      "epoch": 20.0,
-      "learning_rate": 0.00012352941176470587,
-      "loss": 4.69,
-      "step": 60
-    },
-    {
-      "epoch": 20.0,
-      "eval_loss": 4.435084342956543,
-      "eval_runtime": 0.6371,
-      "eval_samples_per_second": 15.695,
-      "eval_wer": 1.0,
-      "step": 60
-    },
-    {
-      "epoch": 21.33,
-      "learning_rate": 0.00010941176470588234,
-      "loss": 4.6092,
-      "step": 64
-    },
-    {
-      "epoch": 21.33,
-      "eval_loss": 4.350079536437988,
-      "eval_runtime": 0.5165,
-      "eval_samples_per_second": 19.361,
-      "eval_wer": 1.0,
-      "step": 64
-    },
-    {
-      "epoch": 22.67,
-      "learning_rate": 9.529411764705882e-05,
-      "loss": 4.9462,
-      "step": 68
-    },
-    {
-      "epoch": 22.67,
-      "eval_loss": 4.301529884338379,
-      "eval_runtime": 0.4931,
-      "eval_samples_per_second": 20.278,
-      "eval_wer": 1.0,
-      "step": 68
-    },
-    {
-      "epoch": 24.0,
-      "learning_rate": 8.117647058823529e-05,
-      "loss": 4.7406,
-      "step": 72
-    },
-    {
-      "epoch": 24.0,
-      "eval_loss": 4.267996311187744,
-      "eval_runtime": 0.5187,
-      "eval_samples_per_second": 19.279,
-      "eval_wer": 1.0,
-      "step": 72
-    },
-    {
-      "epoch": 25.33,
-      "learning_rate": 6.705882352941176e-05,
-      "loss": 4.5484,
-      "step": 76
-    },
-    {
-      "epoch": 25.33,
-      "eval_loss": 4.258701801300049,
-      "eval_runtime": 0.5059,
-      "eval_samples_per_second": 19.768,
-      "eval_wer": 1.0,
-      "step": 76
-    },
-    {
-      "epoch": 26.67,
-      "learning_rate": 5.294117647058824e-05,
-      "loss": 4.2841,
-      "step": 80
-    },
-    {
-      "epoch": 26.67,
-      "eval_loss": 4.255084991455078,
-      "eval_runtime": 0.493,
-      "eval_samples_per_second": 20.286,
-      "eval_wer": 1.0,
-      "step": 80
-    },
-    {
-      "epoch": 28.0,
-      "learning_rate": 3.882352941176471e-05,
-      "loss": 4.4733,
-      "step": 84
-    },
-    {
-      "epoch": 28.0,
-      "eval_loss": 4.255660533905029,
-      "eval_runtime": 0.5041,
-      "eval_samples_per_second": 19.839,
-      "eval_wer": 1.0,
-      "step": 84
-    }
-  ],
-  "max_steps": 90,
-  "num_train_epochs": 30,
-  "total_flos": 3.564185651712e+17,
-  "trial_name": null,
-  "trial_params": null
-}

training_args.bin

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:2c2220a89d8cda2c9e9ddfe7fd922e39268d2ceebcd96397807b4516c8d2e156
-size 2351

vocab.json

@@ -0,0 +1 @@
+{"ㅝ": 0, "ㄱ": 1, "ㅂ": 2, "ㅑ": 3, "ㅜ": 4, ",": 5, "ㅄ": 6, "ㄴ": 7, "ㅕ": 8, "ㅟ": 9, "ㅐ": 10, "ㅆ": 11, "1": 12, "ㅊ": 13, "ㅖ": 14, "ㅈ": 15, "ㅉ": 16, "ㅙ": 17, "ㅛ": 18, ".": 19, "ㅌ": 20, "ㅇ": 21, "ㅚ": 23, "ㄶ": 24, "ㅣ": 25, "ㄸ": 26, "ㅍ": 27, "ㄹ": 28, "ㅔ": 29, "ㅢ": 30, "ㅒ": 31, "ㅏ": 32, "ㄷ": 33, "ㅅ": 34, "ㅠ": 35, "ㅓ": 36, "ㅘ": 37, "ㅋ": 38, "ㅁ": 39, "ㅎ": 40, "ㄲ": 41, "ㅗ": 42, "ㅡ": 43, "|": 22, "[UNK]": 44, "[PAD]": 45}