upload model

config.json

@@ -0,0 +1,79 @@
+{
+  "_name_or_path": "facebook/wav2vec2-xls-r-300m",
+  "activation_dropout": 0.0,
+  "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,
+  "gradient_checkpointing": false,
+  "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_prob": 0.0,
+  "mask_time_length": 10,
+  "mask_time_prob": 0.05,
+  "model_type": "wav2vec2",
+  "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,
+  "pad_token_id": 38,
+  "proj_codevector_dim": 768,
+  "torch_dtype": "float32",
+  "transformers_version": "4.11.3",
+  "use_weighted_layer_sum": false,
+  "vocab_size": 39
+}

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:86fb6c8e8c72fb44c88e5eb037c501bdad9057c468ed9349a8015af5bc28560b
+size 2490378769

preprocessor_config.json

@@ -0,0 +1,9 @@
+{
+  "do_normalize": true,
+  "feature_extractor_type": "Wav2Vec2FeatureExtractor",
+  "feature_size": 1,
+  "padding_side": "right",
+  "padding_value": 0.0,
+  "return_attention_mask": true,
+  "sampling_rate": 16000
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d9fffa4618c81a76a00f8398728b08f87d1a1ed13e9dd91b23ed79d51430b795
+size 1262083569

rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:052df295d6910707b016bd0c631851bdbf6b8b0af7e4593fab5a737ef60ec8de
+size 14567

scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e9fcd9e904053b8ddce97cd0923edfb28f14e8fd90cc8f5438d41189ff5c0dae
+size 559

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eeae617e986d14c33d4b005d55f686f355ecb9b630291edd636a0941bdfb1f75
+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": "|", "tokenizer_class": "Wav2Vec2CTCTokenizer"}

trainer_state.json

@@ -0,0 +1,133 @@
+{
+  "best_metric": 0.3982403874397278,
+  "best_model_checkpoint": "XLS-R-tamil-ipa/checkpoint-1500",
+  "epoch": 1.0706512042818912,
+  "global_step": 1500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.07,
+      "learning_rate": 2.94e-05,
+      "loss": 7.2232,
+      "step": 100
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 5.91e-05,
+      "loss": 3.0769,
+      "step": 200
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 8.909999999999998e-05,
+      "loss": 2.616,
+      "step": 300
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 0.0001188,
+      "loss": 2.3839,
+      "step": 400
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.00014849999999999998,
+      "loss": 1.0712,
+      "step": 500
+    },
+    {
+      "epoch": 0.36,
+      "eval_cer": 0.3590590901797176,
+      "eval_loss": 0.6315754652023315,
+      "eval_runtime": 1320.9999,
+      "eval_samples_per_second": 9.121,
+      "eval_steps_per_second": 1.141,
+      "step": 500
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 0.00017819999999999997,
+      "loss": 0.4126,
+      "step": 600
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.00020789999999999996,
+      "loss": 0.3474,
+      "step": 700
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.00023789999999999998,
+      "loss": 0.3079,
+      "step": 800
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 0.0002673,
+      "loss": 0.3011,
+      "step": 900
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 0.00029729999999999996,
+      "loss": 0.2795,
+      "step": 1000
+    },
+    {
+      "epoch": 0.71,
+      "eval_cer": 0.25890564826700896,
+      "eval_loss": 0.44443580508232117,
+      "eval_runtime": 1282.5832,
+      "eval_samples_per_second": 9.394,
+      "eval_steps_per_second": 1.175,
+      "step": 1000
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 0.00029898963730569947,
+      "loss": 0.2838,
+      "step": 1100
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 0.00029790155440414503,
+      "loss": 0.2674,
+      "step": 1200
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 0.00029681347150259065,
+      "loss": 0.2531,
+      "step": 1300
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.00029570318282753515,
+      "loss": 0.2508,
+      "step": 1400
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 0.0002945928941524796,
+      "loss": 0.2403,
+      "step": 1500
+    },
+    {
+      "epoch": 1.07,
+      "eval_cer": 0.2333796333440308,
+      "eval_loss": 0.3982403874397278,
+      "eval_runtime": 1264.6744,
+      "eval_samples_per_second": 9.527,
+      "eval_steps_per_second": 1.192,
+      "step": 1500
+    }
+  ],
+  "max_steps": 28020,
+  "num_train_epochs": 20,
+  "total_flos": 9.531198923471569e+18,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:23d29bd3f888a21db2f4b36895391dbf3ad074043982a91045cb2ba6375e8801
+size 2863

vocab.json

@@ -0,0 +1 @@
+{"n": 1, "oː": 2, "o": 3, "kʂ": 4, "r": 5, "uː": 6, "aʋ": 7, "a": 8, "u": 9, "d͡ʒ": 10, "ɭ": 11, "s": 12, "h": 13, "ɻ": 14, "p": 15, "்": 16, "ɳ": 17, "aː": 18, "ʈ": 19, "n̪": 20, "k": 21, "e": 22, "m": 23, "iː": 24, "j": 25, "ɾ": 26, "i": 27, "eː": 28, "t͡ʃ": 29, "l": 30, "ʋ": 31, "ʂ": 32, "t̪": 33, "aj": 34, "ŋ": 35, "ɲ": 36, "|": 0, "[UNK]": 37, "[PAD]": 38}