Training in progress, step 100

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "nguyenvulebinh/wav2vec2-base-vi-vlsp2020",
+  "_name_or_path": "nguyenvulebinh/wav2vec2-large-vi-vlsp2020",
   "activation_dropout": 0.0,
   "adapter_kernel_size": 3,
   "adapter_stride": 2,
@@ -11,9 +11,9 @@
   "attention_dropout": 0.05,
   "bos_token_id": 1,
   "classifier_proj_size": 256,
-  "codevector_dim": 256,
+  "codevector_dim": 768,
   "contrastive_logits_temperature": 0.1,
-  "conv_bias": false,
+  "conv_bias": true,
   "conv_dim": [
     512,
     512,
@@ -44,41 +44,48 @@
   "ctc_loss_reduction": "mean",
   "ctc_zero_infinity": false,
   "diversity_loss_weight": 0.1,
-  "do_stable_layer_norm": false,
+  "do_stable_layer_norm": true,
   "eos_token_id": 2,
   "feat_extract_activation": "gelu",
   "feat_extract_dropout": 0.0,
-  "feat_extract_norm": "group",
+  "feat_extract_norm": "layer",
   "feat_proj_dropout": 0.0,
   "feat_quantizer_dropout": 0.0,
   "final_dropout": 0.0,
   "hidden_act": "gelu",
   "hidden_dropout": 0.05,
-  "hidden_dropout_prob": 0.1,
-  "hidden_size": 768,
+  "hidden_size": 1024,
   "initializer_range": 0.02,
-  "intermediate_size": 3072,
+  "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": 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,
+  "mask_time_selection": "static",
   "model_type": "wav2vec2",
   "num_adapter_layers": 3,
-  "num_attention_heads": 12,
+  "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": 12,
+  "num_hidden_layers": 24,
   "num_negatives": 100,
-  "output_hidden_size": 768,
+  "output_hidden_size": 1024,
   "pad_token_id": 95,
-  "proj_codevector_dim": 256,
+  "proj_codevector_dim": 768,
   "tdnn_dilation": [
     1,
     2,

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e7ebd332cd92c4cbbfa0078b9bc9f036b61b97fe8fb8d4f4b1bfce78f26307d5
-size 384990573
+oid sha256:42f9cfc71ec6e451fd7741cc82b40dfb85104a0bec800f9a18082eeebc6f3adf
+size 1274951623

runs/Jan17_07-59-45_58e570142ef9/1673942703.3857741/events.out.tfevents.1673942703.58e570142ef9.23.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2686713659850bfb5ac9339b784b1b6c1d5cb33f0f6ca30ab3d91a70136197c4
+size 5323

runs/Jan17_07-59-45_58e570142ef9/events.out.tfevents.1673942703.58e570142ef9.23.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:55a51f09b9f8fb1f91384e7e2376a44ec95ec25208a5a63d536fc29717d3806d
+size 5703

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:83c094825a2dc05c9cc06527ba032c5aec49a72dc58f517badb41d17c72f093e
+oid sha256:be9ee12cc1a397b69e63a8850184d38d83d15216b9e7fd6d5331a2ef3d2da5bc
 size 3247