Training in progress, epoch 1

config.json

@@ -1,14 +1,20 @@
 {
-  "_name_or_path": "facebook/hubert-large-ls960-ft",
-  "activation_dropout": 0.1,
+  "_name_or_path": "facebook/wav2vec2-base",
+  "activation_dropout": 0.0,
+  "adapter_attn_dim": null,
+  "adapter_kernel_size": 3,
+  "adapter_stride": 2,
+  "add_adapter": false,
   "apply_spec_augment": true,
   "architectures": [
-    "HubertForSequenceClassification"
+    "Wav2Vec2ForSequenceClassification"
   ],
   "attention_dropout": 0.1,
   "bos_token_id": 1,
   "classifier_proj_size": 256,
-  "conv_bias": true,
+  "codevector_dim": 256,
+  "contrastive_logits_temperature": 0.1,
+  "conv_bias": false,
   "conv_dim": [
     512,
     512,
@@ -39,50 +45,74 @@
   "ctc_loss_reduction": "sum",
   "ctc_zero_infinity": false,
   "diversity_loss_weight": 0.1,
-  "do_stable_layer_norm": true,
+  "do_stable_layer_norm": false,
   "eos_token_id": 2,
   "feat_extract_activation": "gelu",
-  "feat_extract_dropout": 0.0,
-  "feat_extract_norm": "layer",
+  "feat_extract_norm": "group",
   "feat_proj_dropout": 0.1,
-  "feat_proj_layer_norm": true,
-  "final_dropout": 0.1,
-  "gradient_checkpointing": false,
+  "feat_quantizer_dropout": 0.0,
+  "final_dropout": 0.0,
+  "freeze_feat_extract_train": true,
   "hidden_act": "gelu",
   "hidden_dropout": 0.1,
-  "hidden_dropout_prob": 0.1,
-  "hidden_size": 1024,
-  "id2label": {
-    "0": "LABEL_0",
-    "1": "LABEL_1",
-    "2": "LABEL_2",
-    "3": "LABEL_3"
-  },
+  "hidden_size": 768,
   "initializer_range": 0.02,
-  "intermediate_size": 4096,
-  "label2id": {
-    "LABEL_0": 0,
-    "LABEL_1": 1,
-    "LABEL_2": 2,
-    "LABEL_3": 3
-  },
+  "intermediate_size": 3072,
   "layer_norm_eps": 1e-05,
-  "layerdrop": 0.1,
+  "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": 10,
   "mask_feature_min_masks": 0,
   "mask_feature_prob": 0.0,
   "mask_time_length": 10,
   "mask_time_min_masks": 2,
+  "mask_time_min_space": 1,
+  "mask_time_other": 0.0,
   "mask_time_prob": 0.05,
-  "model_type": "hubert",
-  "num_attention_heads": 16,
+  "mask_time_selection": "static",
+  "model_type": "wav2vec2",
+  "no_mask_channel_overlap": false,
+  "no_mask_time_overlap": false,
+  "num_adapter_layers": 3,
+  "num_attention_heads": 12,
+  "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_hidden_layers": 12,
+  "num_negatives": 100,
+  "output_hidden_size": 768,
   "pad_token_id": 0,
+  "proj_codevector_dim": 256,
+  "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": "4.40.2",
   "use_weighted_layer_sum": false,
-  "vocab_size": 32
+  "vocab_size": 32,
+  "xvector_output_dim": 512
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:231f889082b255f7f3f712f308ad3a0428f34fc67c69c1b344e876dcffb72a2c
-size 1262860464
+oid sha256:06f971f9a2989c75c25fae5f70361706066378ae05a97a07c9a2693b7c179772
+size 378302312

preprocessor_config.json

@@ -3,7 +3,7 @@
   "feature_extractor_type": "Wav2Vec2FeatureExtractor",
   "feature_size": 1,
   "padding_side": "right",
-  "padding_value": 0,
-  "return_attention_mask": true,
+  "padding_value": 0.0,
+  "return_attention_mask": false,
   "sampling_rate": 16000
 }

runs/Jun14_15-20-00_WIN-CEJF2NB9KFH/events.out.tfevents.1718349601.WIN-CEJF2NB9KFH.46136.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1d62a333e058b2e35eb5b2bae2f2349b013ed7ca00e866e23ef94b452d356c28
+size 12014

runs/Jun14_15-27-37_WIN-CEJF2NB9KFH/events.out.tfevents.1718350058.WIN-CEJF2NB9KFH.80928.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:906091bbd9af1257df3c80b4ca30d303865a1ffe30de8b246788357d63a03787
+size 12015

runs/Jun14_15-41-15_WIN-CEJF2NB9KFH/events.out.tfevents.1718350876.WIN-CEJF2NB9KFH.86172.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5734e8286e5d749a13f11e125a8af81a671812c39eee10b2348668af13491a42
+size 6951

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:64023c86a070381403f232ee499e8e1c68923eb78496f8efc186e7732aba0afd
+oid sha256:3d0200d5cc3fe54ceee0bdf76642da780274759690aa46be812caad6e62f3e66
 size 4539