sync

checkpoint-20/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fc9a37bf342d6b4931c7270220724a92600d8294837cdbcd124a0ec5e19ec080
+oid sha256:0f436b11b0e72d1e94060965296a9e9e846ca0f9771daa24e1a9345c5f9a7d16
 size 2490337361

checkpoint-20/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3628cb625fb06fb9b5feb8070ed125a0961b552a86df1fcdcbb45d1703f7757f
+oid sha256:9d76176b1de3e024882e5dc97f4c62f8c353ad6bd2bb044b8780d02f878d5a8b
 size 1262063089

checkpoint-20/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:73e968b5fb75f3d028c5b87fc29d361b46586d839eb1fc406be3e77c69542778
+oid sha256:59257a32aba32d8c5fcf991ca9a70b329ffb647f6a2161088a777522fce19900
 size 14567

checkpoint-20/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a8d1040cd70f91f849688103143e8c9d631ff10c4acee0b2f4d6ff9e27727776
+oid sha256:de50320fd62b63ba68a5d5884bc2615e3b1236c94802e7d58610a99abb679720
 size 559

checkpoint-20/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dd820f300afb9f82707d847f7f02b33e058c7d1676a355fccacc0f7f2f251fcd
+oid sha256:dff68c58af865abbc7152291f4ccc6e0dc5bda2584ed11e74751073d18e44ac9
 size 623

checkpoint-20/trainer_state.json

@@ -1,7 +1,7 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.26490066225165565,
+  "epoch": 0.25559105431309903,
   "global_step": 20,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
@@ -9,29 +9,29 @@
   "log_history": [
     {
       "epoch": 0.13,
-      "learning_rate": 0.00045,
-      "loss": 10.7126,
+      "learning_rate": 1.8749999999999998e-06,
+      "loss": 13.2907,
       "step": 10
     },
     {
       "epoch": 0.26,
-      "learning_rate": 0.0007469387755102041,
-      "loss": 5.3922,
+      "learning_rate": 3.7499999999999997e-06,
+      "loss": 17.4007,
       "step": 20
     },
     {
       "epoch": 0.26,
-      "eval_loss": 3.5064597129821777,
-      "eval_runtime": 4.8648,
-      "eval_samples_per_second": 20.145,
-      "eval_steps_per_second": 0.822,
+      "eval_loss": 21.748823165893555,
+      "eval_runtime": 3.3192,
+      "eval_samples_per_second": 30.127,
+      "eval_steps_per_second": 1.205,
       "eval_wer": 1.0,
       "step": 20
     }
   ],
-  "max_steps": 750,
+  "max_steps": 780,
   "num_train_epochs": 10,
-  "total_flos": 4.155165485184e+17,
+  "total_flos": 3.123017532990259e+17,
   "trial_name": null,
   "trial_params": null
 }

checkpoint-20/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ad06bcea7129bc90fba0c606327162a14f89fd4622d165a2ff161b26e6fbb59e
+oid sha256:5806ee760f93397337bb9a8c9234fc7d900c26bb857abd80187ccde2f46dde70
 size 3055

checkpoint-300/config.json

@@ -1,115 +0,0 @@
-{
-  "_name_or_path": "KBLab/wav2vec2-large-voxrex",
-  "activation_dropout": 0.1,
-  "adapter_kernel_size": 3,
-  "adapter_stride": 2,
-  "add_adapter": false,
-  "apply_spec_augment": true,
-  "architectures": [
-    "Wav2Vec2ForCTC"
-  ],
-  "attention_dropout": 0.0,
-  "bos_token_id": 1,
-  "classifier_proj_size": 256,
-  "codevector_dim": 768,
-  "contrastive_logits_temperature": 0.1,
-  "conv_bias": true,
-  "conv_dim": [
-    512,
-    512,
-    512,
-    512,
-    512,
-    512,
-    512
-  ],
-  "conv_kernel": [
-    10,
-    3,
-    3,
-    3,
-    3,
-    2,
-    2
-  ],
-  "conv_stride": [
-    5,
-    2,
-    2,
-    2,
-    2,
-    2,
-    2
-  ],
-  "ctc_loss_reduction": "mean",
-  "ctc_zero_infinity": false,
-  "diversity_loss_weight": 0.1,
-  "do_stable_layer_norm": true,
-  "eos_token_id": 2,
-  "feat_extract_activation": "gelu",
-  "feat_extract_dropout": 0.0,
-  "feat_extract_norm": "layer",
-  "feat_proj_dropout": 0.0,
-  "feat_quantizer_dropout": 0.0,
-  "final_dropout": 0.0,
-  "hidden_act": "gelu",
-  "hidden_dropout": 0.0,
-  "hidden_size": 1024,
-  "initializer_range": 0.02,
-  "intermediate_size": 4096,
-  "layer_norm_eps": 1e-05,
-  "layerdrop": 0.0,
-  "mask_channel_length": 10,
-  "mask_channel_min_space": 1,
-  "mask_channel_other": 0.0,
-  "mask_channel_prob": 0.0,
-  "mask_channel_selection": "static",
-  "mask_feature_length": 64,
-  "mask_feature_min_masks": 0,
-  "mask_feature_prob": 0.25,
-  "mask_time_length": 10,
-  "mask_time_min_masks": 2,
-  "mask_time_min_space": 1,
-  "mask_time_other": 0.0,
-  "mask_time_prob": 0.75,
-  "mask_time_selection": "static",
-  "model_type": "wav2vec2",
-  "num_adapter_layers": 3,
-  "num_attention_heads": 16,
-  "num_codevector_groups": 2,
-  "num_codevectors_per_group": 320,
-  "num_conv_pos_embedding_groups": 16,
-  "num_conv_pos_embeddings": 128,
-  "num_feat_extract_layers": 7,
-  "num_hidden_layers": 24,
-  "num_negatives": 100,
-  "output_hidden_size": 1024,
-  "pad_token_id": 31,
-  "proj_codevector_dim": 768,
-  "tdnn_dilation": [
-    1,
-    2,
-    3,
-    1,
-    1
-  ],
-  "tdnn_dim": [
-    512,
-    512,
-    512,
-    512,
-    1500
-  ],
-  "tdnn_kernel": [
-    5,
-    3,
-    3,
-    1,
-    1
-  ],
-  "torch_dtype": "float32",
-  "transformers_version": "4.17.0.dev0",
-  "use_weighted_layer_sum": false,
-  "vocab_size": 34,
-  "xvector_output_dim": 512
-}

checkpoint-300/optimizer.pt

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:7ae9e41884d3305d9a0797fd14fd35fc8e619e12951eedb6e1396e3ead551313
-size 2490337809

checkpoint-300/preprocessor_config.json

@@ -1,9 +0,0 @@
-{
-  "do_normalize": true,
-  "feature_extractor_type": "Wav2Vec2FeatureExtractor",
-  "feature_size": 1,
-  "padding_side": "right",
-  "padding_value": 0,
-  "return_attention_mask": true,
-  "sampling_rate": 16000
-}

checkpoint-300/pytorch_model.bin

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:4ba896f27a39cdc0a9c70e56aeff43d05c53ba85e3a65d2c02f13edcce3b7ac8
-size 1262063089

checkpoint-300/rng_state.pth

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:7dad8817bedcf4b45f9a2d36af4404126640a18d28d8189d505ec2ab56d4b6a1
-size 14503

checkpoint-300/scaler.pt

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:0967b9f865f16344c55f5ccc3cf7d6e8e97ca61dda304e931ca6bad130f48dd1
-size 559

checkpoint-300/scheduler.pt

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:303fe9412b2bbace437d42323a895571355b0ae20c611a474d9492be12287f9c
-size 623

checkpoint-300/trainer_state.json

@@ -1,160 +0,0 @@
-{
-  "best_metric": null,
-  "best_model_checkpoint": null,
-  "epoch": 3.8434504792332267,
-  "global_step": 300,
-  "is_hyper_param_search": false,
-  "is_local_process_zero": true,
-  "is_world_process_zero": true,
-  "log_history": [
-    {
-      "epoch": 0.26,
-      "learning_rate": 0.0007307692307692308,
-      "loss": 4.2559,
-      "step": 20
-    },
-    {
-      "epoch": 0.51,
-      "learning_rate": 0.0007115384615384615,
-      "loss": 3.068,
-      "step": 40
-    },
-    {
-      "epoch": 0.64,
-      "eval_loss": 3.058549165725708,
-      "eval_runtime": 3.3353,
-      "eval_samples_per_second": 29.982,
-      "eval_steps_per_second": 1.199,
-      "eval_wer": 1.0,
-      "step": 50
-    },
-    {
-      "epoch": 0.77,
-      "learning_rate": 0.0006923076923076924,
-      "loss": 3.0637,
-      "step": 60
-    },
-    {
-      "epoch": 1.03,
-      "learning_rate": 0.0006730769230769232,
-      "loss": 3.0752,
-      "step": 80
-    },
-    {
-      "epoch": 1.28,
-      "learning_rate": 0.0006538461538461538,
-      "loss": 3.0083,
-      "step": 100
-    },
-    {
-      "epoch": 1.28,
-      "eval_loss": 3.1246652603149414,
-      "eval_runtime": 3.2285,
-      "eval_samples_per_second": 30.974,
-      "eval_steps_per_second": 1.239,
-      "eval_wer": 1.0,
-      "step": 100
-    },
-    {
-      "epoch": 1.54,
-      "learning_rate": 0.0006346153846153846,
-      "loss": 3.0159,
-      "step": 120
-    },
-    {
-      "epoch": 1.79,
-      "learning_rate": 0.0006153846153846154,
-      "loss": 2.977,
-      "step": 140
-    },
-    {
-      "epoch": 1.92,
-      "eval_loss": 2.973355770111084,
-      "eval_runtime": 3.1716,
-      "eval_samples_per_second": 31.53,
-      "eval_steps_per_second": 1.261,
-      "eval_wer": 1.0,
-      "step": 150
-    },
-    {
-      "epoch": 2.05,
-      "learning_rate": 0.0005961538461538461,
-      "loss": 3.0056,
-      "step": 160
-    },
-    {
-      "epoch": 2.31,
-      "learning_rate": 0.000576923076923077,
-      "loss": 2.9591,
-      "step": 180
-    },
-    {
-      "epoch": 2.56,
-      "learning_rate": 0.0005576923076923078,
-      "loss": 2.9398,
-      "step": 200
-    },
-    {
-      "epoch": 2.56,
-      "eval_loss": 2.940675735473633,
-      "eval_runtime": 3.1691,
-      "eval_samples_per_second": 31.554,
-      "eval_steps_per_second": 1.262,
-      "eval_wer": 1.0,
-      "step": 200
-    },
-    {
-      "epoch": 2.82,
-      "learning_rate": 0.0005384615384615384,
-      "loss": 2.9295,
-      "step": 220
-    },
-    {
-      "epoch": 3.08,
-      "learning_rate": 0.0005192307692307692,
-      "loss": 2.9465,
-      "step": 240
-    },
-    {
-      "epoch": 3.2,
-      "eval_loss": 2.835913896560669,
-      "eval_runtime": 3.1824,
-      "eval_samples_per_second": 31.423,
-      "eval_steps_per_second": 1.257,
-      "eval_wer": 1.0,
-      "step": 250
-    },
-    {
-      "epoch": 3.33,
-      "learning_rate": 0.0005,
-      "loss": 2.9053,
-      "step": 260
-    },
-    {
-      "epoch": 3.59,
-      "learning_rate": 0.0004807692307692308,
-      "loss": 2.899,
-      "step": 280
-    },
-    {
-      "epoch": 3.84,
-      "learning_rate": 0.0004615384615384616,
-      "loss": 2.8861,
-      "step": 300
-    },
-    {
-      "epoch": 3.84,
-      "eval_loss": 2.8373122215270996,
-      "eval_runtime": 3.1455,
-      "eval_samples_per_second": 31.791,
-      "eval_steps_per_second": 1.272,
-      "eval_wer": 1.0,
-      "step": 300
-    }
-  ],
-  "max_steps": 780,
-  "num_train_epochs": 10,
-  "total_flos": 4.5237874175174973e+18,
-  "trial_name": null,
-  "trial_params": null
-}

checkpoint-300/training_args.bin

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:31becb5e425437ea470512000be6c31bb502b71fe15942ff8cf0cc01eb8267a0
-size 3055

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3628cb625fb06fb9b5feb8070ed125a0961b552a86df1fcdcbb45d1703f7757f
+oid sha256:9d76176b1de3e024882e5dc97f4c62f8c353ad6bd2bb044b8780d02f878d5a8b
 size 1262063089

run.sh

@@ -6,13 +6,11 @@ python run_speech_recognition_ctc.py \
 	--eval_split_name="test" \
 	--output_dir="./" \
 	--overwrite_output_dir \
-	--num_train_epochs="10" \
+	--num_train_epochs="3" \
 	--per_device_train_batch_size="32" \
 	--per_device_eval_batch_size="32" \
 	--gradient_accumulation_steps="4" \
 	--learning_rate="7.5e-4" \
-	--max_train_samples="10000" \
-    --max_eval_samples="100" \
 	--warmup_ratio="0.02" \
 	--length_column_name="input_length" \
 	--evaluation_strategy="steps" \
@@ -40,27 +38,26 @@ python run_speech_recognition_ctc.py \
 
 python run_speech_recognition_ctc.py \
 	--dataset_name="mozilla-foundation/common_voice_8_0" \
-	--model_name_or_path="KBLab/wav2vec2-large-voxrex" \
+	--model_name_or_path="marinone94/xls-r-300m-sv-robust" \
 	--dataset_config_name="sv-SE" \
 	--train_split_name="train+validation" \
 	--eval_split_name="test" \
 	--output_dir="./" \
-	--num_train_epochs="10" \
+	--overwrite_output_dir \
+	--num_train_epochs="50" \
 	--per_device_train_batch_size="32" \
 	--per_device_eval_batch_size="32" \
 	--gradient_accumulation_steps="4" \
 	--learning_rate="7.5e-5" \
-	--warmup_steps="400" \
-	--max_train_samples="10000" \
-    --max_eval_samples="100" \
+	--warmup_steps="2000" \
 	--length_column_name="input_length" \
 	--evaluation_strategy="steps" \
 	--save_strategy="steps" \
-	--eval_steps="20" \
-	--save_steps="20" \
+	--eval_steps="500" \
+	--save_steps="500" \
 	--text_column_name="sentence" \
 	--chars_to_ignore , ? . ! \- \; \: \" “ % ‘ ” � — ’ … – \
-	--logging_steps="10" \
+	--logging_steps="100" \
 	--layerdrop="0.0" \
 	--activation_dropout="0.1" \
 	--save_total_limit="2" \

special_tokens_map.json

@@ -1 +1 @@
-{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "[UNK]", "pad_token": "[PAD]", "additional_special_tokens": [{"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}]}
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "[UNK]", "pad_token": "[PAD]", "additional_special_tokens": [{"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "<s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, {"content": "</s>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}]}

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ad06bcea7129bc90fba0c606327162a14f89fd4622d165a2ff161b26e6fbb59e
+oid sha256:5806ee760f93397337bb9a8c9234fc7d900c26bb857abd80187ccde2f46dde70
 size 3055