upload beta version

added_tokens.json

@@ -0,0 +1 @@
+{"<s>": 44, "</s>": 45}

config.json

@@ -0,0 +1,107 @@
+{
+  "_name_or_path": "facebook/wav2vec2-xls-r-300m",
+  "activation_dropout": 0.0,
+  "adapter_kernel_size": 3,
+  "adapter_stride": 2,
+  "add_adapter": false,
+  "apply_spec_augment": true,
+  "architectures": [
+    "Wav2Vec2ForPreTraining"
+  ],
+  "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_feature_length": 10,
+  "mask_feature_min_masks": 0,
+  "mask_feature_prob": 0.0,
+  "mask_time_length": 10,
+  "mask_time_min_masks": 2,
+  "mask_time_prob": 0.05,
+  "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": 43,
+  "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": "4.16.2",
+  "use_weighted_layer_sum": false,
+  "vocab_size": 46,
+  "xvector_output_dim": 512
+}

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2720401989ca4dbdc57ad7199cdc4116d02c36c374567e9a1295b01cd6c45b3
+size 625461417

preprocessor_config.json

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

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3299e75237418a55c403fc9c78801934d3057ff076e484879bc40bfd516f13ba
+size 1262112241

rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:39223200ea0fd8350b41518c6492d3d14d96d823ca3f6e9a374864389e6dbb1b
+size 14503

rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5091f67e4a1caff5227497b7c37712739cd8baf0783971573e46569913fdfff3
+size 14567

rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:16de60f262d42506479df2d553aa073b64cf621ae9eebfc35fbc2b23021cee2d
+size 14503

rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a641931222bcf2b19d6a073bd70c446b59abafb3a4077c2d6a5aea1f4001e06a
+size 14503

rng_state_4.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:da3f79cf3a4b4d9379e47a4353e6555235b22515af826d448849df6d7a5f04f0
+size 14567

rng_state_5.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:16352761baa415d3af2ff3cea0ef555419a76b51449f27bcec56bcdba9a15ff5
+size 14567

rng_state_6.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8ea93d555065f78f93a91b315329c1c392289d935801de6b732e124d18b1586a
+size 14567

rng_state_7.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:46629c3fa6cb14bd2274cb93b2b9a613c84794d3912a7fb43cf8af9b51524544
+size 14503

scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:431319fd21daa87636d8253400f763a0a1a5400306ce1db9e67b38942c76551d
+size 559

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6f90f0da980bf8cd346163915335c996f7bbdbcfe93a2e624a8b02f902aa6d01
+size 623

special_tokens_map.json

@@ -0,0 +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}]}

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": "|", "special_tokens_map_file": null, "tokenizer_file": null, "name_or_path": "/workspace/cv-corpus-8.0-2022-01-19/output", "tokenizer_class": "Wav2Vec2CTCTokenizer"}

trainer_state.json

@@ -0,0 +1,289 @@
+{
+  "best_metric": 0.4493387111903199,
+  "best_model_checkpoint": "/workspace/cv-corpus-8.0-2022-01-19/output/checkpoint-1000",
+  "epoch": 8.317338451695457,
+  "global_step": 13000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.32,
+      "learning_rate": 9.92e-05,
+      "loss": 6.7916,
+      "step": 500
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 0.00019920000000000002,
+      "loss": 1.354,
+      "step": 1000
+    },
+    {
+      "epoch": 0.64,
+      "eval_loss": 0.4108898937702179,
+      "eval_runtime": 71.0896,
+      "eval_samples_per_second": 146.125,
+      "eval_steps_per_second": 1.829,
+      "eval_wer": 0.4493387111903199,
+      "step": 1000
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 0.00019321941216678059,
+      "loss": 0.7084,
+      "step": 1500
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 0.00018638414217361587,
+      "loss": 0.5886,
+      "step": 2000
+    },
+    {
+      "epoch": 1.28,
+      "eval_loss": 0.2797781527042389,
+      "eval_runtime": 71.0704,
+      "eval_samples_per_second": 146.165,
+      "eval_steps_per_second": 1.829,
+      "eval_wer": 0.3099334021198762,
+      "step": 2000
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 0.00017954887218045112,
+      "loss": 0.5386,
+      "step": 2500
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 0.0001727136021872864,
+      "loss": 0.4977,
+      "step": 3000
+    },
+    {
+      "epoch": 1.92,
+      "eval_loss": 0.23867885768413544,
+      "eval_runtime": 71.0833,
+      "eval_samples_per_second": 146.138,
+      "eval_steps_per_second": 1.829,
+      "eval_wer": 0.2673295188068662,
+      "step": 3000
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 0.00016587833219412168,
+      "loss": 0.4531,
+      "step": 3500
+    },
+    {
+      "epoch": 2.56,
+      "learning_rate": 0.00015905673274094328,
+      "loss": 0.4253,
+      "step": 4000
+    },
+    {
+      "epoch": 2.56,
+      "eval_loss": 0.22657370567321777,
+      "eval_runtime": 71.5875,
+      "eval_samples_per_second": 145.109,
+      "eval_steps_per_second": 1.816,
+      "eval_wer": 0.2523215458212175,
+      "step": 4000
+    },
+    {
+      "epoch": 2.88,
+      "learning_rate": 0.00015222146274777856,
+      "loss": 0.413,
+      "step": 4500
+    },
+    {
+      "epoch": 3.2,
+      "learning_rate": 0.0001453861927546138,
+      "loss": 0.3942,
+      "step": 5000
+    },
+    {
+      "epoch": 3.2,
+      "eval_loss": 0.21706202626228333,
+      "eval_runtime": 72.0993,
+      "eval_samples_per_second": 144.079,
+      "eval_steps_per_second": 1.803,
+      "eval_wer": 0.2437294812869337,
+      "step": 5000
+    },
+    {
+      "epoch": 3.52,
+      "learning_rate": 0.00013855092276144907,
+      "loss": 0.3741,
+      "step": 5500
+    },
+    {
+      "epoch": 3.84,
+      "learning_rate": 0.00013171565276828435,
+      "loss": 0.3619,
+      "step": 6000
+    },
+    {
+      "epoch": 3.84,
+      "eval_loss": 0.20762862265110016,
+      "eval_runtime": 72.0546,
+      "eval_samples_per_second": 144.168,
+      "eval_steps_per_second": 1.804,
+      "eval_wer": 0.22530719444705,
+      "step": 6000
+    },
+    {
+      "epoch": 4.16,
+      "learning_rate": 0.00012489405331510595,
+      "loss": 0.3435,
+      "step": 6500
+    },
+    {
+      "epoch": 4.48,
+      "learning_rate": 0.00011805878332194122,
+      "loss": 0.3245,
+      "step": 7000
+    },
+    {
+      "epoch": 4.48,
+      "eval_loss": 0.2087564468383789,
+      "eval_runtime": 71.9965,
+      "eval_samples_per_second": 144.285,
+      "eval_steps_per_second": 1.806,
+      "eval_wer": 0.21862864646843636,
+      "step": 7000
+    },
+    {
+      "epoch": 4.8,
+      "learning_rate": 0.0001112235133287765,
+      "loss": 0.3135,
+      "step": 7500
+    },
+    {
+      "epoch": 5.12,
+      "learning_rate": 0.0001044019138755981,
+      "loss": 0.308,
+      "step": 8000
+    },
+    {
+      "epoch": 5.12,
+      "eval_loss": 0.2086208015680313,
+      "eval_runtime": 68.9232,
+      "eval_samples_per_second": 150.718,
+      "eval_steps_per_second": 1.886,
+      "eval_wer": 0.22063596285526685,
+      "step": 8000
+    },
+    {
+      "epoch": 5.44,
+      "learning_rate": 9.756664388243337e-05,
+      "loss": 0.292,
+      "step": 8500
+    },
+    {
+      "epoch": 5.76,
+      "learning_rate": 9.073137388926864e-05,
+      "loss": 0.2881,
+      "step": 9000
+    },
+    {
+      "epoch": 5.76,
+      "eval_loss": 0.20888157188892365,
+      "eval_runtime": 70.2391,
+      "eval_samples_per_second": 147.895,
+      "eval_steps_per_second": 1.851,
+      "eval_wer": 0.21048682112372197,
+      "step": 9000
+    },
+    {
+      "epoch": 6.08,
+      "learning_rate": 8.389610389610389e-05,
+      "loss": 0.2717,
+      "step": 9500
+    },
+    {
+      "epoch": 6.4,
+      "learning_rate": 7.706083390293917e-05,
+      "loss": 0.2557,
+      "step": 10000
+    },
+    {
+      "epoch": 6.4,
+      "eval_loss": 0.20148096978664398,
+      "eval_runtime": 70.1726,
+      "eval_samples_per_second": 148.035,
+      "eval_steps_per_second": 1.853,
+      "eval_wer": 0.20035643935840916,
+      "step": 10000
+    },
+    {
+      "epoch": 6.72,
+      "learning_rate": 7.022556390977444e-05,
+      "loss": 0.2536,
+      "step": 10500
+    },
+    {
+      "epoch": 7.04,
+      "learning_rate": 6.33902939166097e-05,
+      "loss": 0.248,
+      "step": 11000
+    },
+    {
+      "epoch": 7.04,
+      "eval_loss": 0.2043762356042862,
+      "eval_runtime": 70.2184,
+      "eval_samples_per_second": 147.938,
+      "eval_steps_per_second": 1.851,
+      "eval_wer": 0.19529124847575274,
+      "step": 11000
+    },
+    {
+      "epoch": 7.36,
+      "learning_rate": 5.655502392344498e-05,
+      "loss": 0.2308,
+      "step": 11500
+    },
+    {
+      "epoch": 7.68,
+      "learning_rate": 4.971975393028025e-05,
+      "loss": 0.2251,
+      "step": 12000
+    },
+    {
+      "epoch": 7.68,
+      "eval_loss": 0.20575200021266937,
+      "eval_runtime": 70.8946,
+      "eval_samples_per_second": 146.527,
+      "eval_steps_per_second": 1.834,
+      "eval_wer": 0.19315261232529782,
+      "step": 12000
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 4.291182501708818e-05,
+      "loss": 0.2207,
+      "step": 12500
+    },
+    {
+      "epoch": 8.32,
+      "learning_rate": 3.6076555023923446e-05,
+      "loss": 0.2052,
+      "step": 13000
+    },
+    {
+      "epoch": 8.32,
+      "eval_loss": 0.21170856058597565,
+      "eval_runtime": 70.7957,
+      "eval_samples_per_second": 146.732,
+      "eval_steps_per_second": 1.836,
+      "eval_wer": 0.18776850201669637,
+      "step": 13000
+    }
+  ],
+  "max_steps": 15630,
+  "num_train_epochs": 10,
+  "total_flos": 8.268631785694088e+20,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1a1b80bcfc69cf9fa9cf637cef6f631649190d0c441246f20b284d86514490bf
+size 3055

vocab.json

@@ -0,0 +1 @@
+{"\"": 1, "،": 2, "ء": 3, "آ": 4, "أ": 5, "ؤ": 6, "إ": 7, "ئ": 8, "ا": 9, "ب": 10, "ة": 11, "ت": 12, "ث": 13, "ج": 14, "ح": 15, "خ": 16, "د": 17, "ذ": 18, "ر": 19, "ز": 20, "س": 21, "ش": 22, "ص": 23, "ض": 24, "ط": 25, "ظ": 26, "ع": 27, "غ": 28, "ـ": 29, "ف": 30, "ق": 31, "ك": 32, "ل": 33, "م": 34, "ن": 35, "ه": 36, "و": 37, "ى": 38, "ي": 39, "ّ": 40, "ی": 41, "|": 0, "[UNK]": 42, "[PAD]": 43}