End of training

config.json

@@ -1,6 +1,6 @@
 {
-  "_name_or_path": "facebook/wav2vec2-base",
-  "activation_dropout": 0.0,
+  "_name_or_path": "jonatasgrosman/wav2vec2-large-xlsr-53-chinese-zh-cn",
+  "activation_dropout": 0.05,
   "apply_spec_augment": true,
   "architectures": [
     "Wav2Vec2ForCTC"
@@ -10,7 +10,7 @@
   "classifier_proj_size": 256,
   "codevector_dim": 256,
   "contrastive_logits_temperature": 0.1,
-  "conv_bias": false,
+  "conv_bias": true,
   "conv_dim": [
     512,
     512,
@@ -39,23 +39,23 @@
     2
   ],
   "ctc_loss_reduction": "mean",
-  "ctc_zero_infinity": false,
+  "ctc_zero_infinity": true,
   "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_norm": "group",
-  "feat_proj_dropout": 0.1,
+  "feat_extract_dropout": 0.0,
+  "feat_extract_norm": "layer",
+  "feat_proj_dropout": 0.05,
   "feat_quantizer_dropout": 0.0,
   "final_dropout": 0.0,
-  "freeze_feat_extract_train": true,
   "hidden_act": "gelu",
-  "hidden_dropout": 0.1,
-  "hidden_size": 768,
+  "hidden_dropout": 0.05,
+  "hidden_size": 1024,
   "initializer_range": 0.02,
-  "intermediate_size": 3072,
+  "intermediate_size": 4096,
   "layer_norm_eps": 1e-05,
-  "layerdrop": 0.0,
+  "layerdrop": 0.05,
   "mask_channel_length": 10,
   "mask_channel_min_space": 1,
   "mask_channel_other": 0.0,
@@ -69,20 +69,18 @@
   "mask_time_prob": 0.05,
   "mask_time_selection": "static",
   "model_type": "wav2vec2",
-  "no_mask_channel_overlap": false,
-  "no_mask_time_overlap": false,
-  "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,
   "pad_token_id": 29,
   "proj_codevector_dim": 256,
   "torch_dtype": "float32",
   "transformers_version": "4.11.3",
   "use_weighted_layer_sum": false,
-  "vocab_size": 32
+  "vocab_size": 3503
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8d6d3ece8da8c8a36ef3cfcbbba81aacf4d0b0c218dea7607f8ce03c0c1fda3f
-size 377670039
+oid sha256:939abe444d3fe2192f8e9ca63213193d5eee0f949ff0bc098e51be6691ffcd47
+size 1276286001

runs/Feb17_10-00-25_aa4c2d3bb432/1645092102.5238657/events.out.tfevents.1645092102.aa4c2d3bb432.98.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:15bb714c21327cbd326b6fb872f37c405874f90babd80c081348017f84fbdd97
+size 4555

runs/Feb17_10-00-25_aa4c2d3bb432/events.out.tfevents.1645092102.aa4c2d3bb432.98.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4a5eb0fc53dc397301726757ebb4acd33d337d681927010a0ac2396e3174d271
+size 4756

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a7470f214e490e7dce62e9b00056ce488cdb892f835feaf7400745eb7cc41c8b
+oid sha256:dd8327c8cfc13350ca25565d9eddb50858f1ee45fbe7580bf2942994829dc303
 size 2863