Training in progress, step 1600

README.md

@@ -1,10 +1,4 @@
----
-license: apache-2.0
-tags:
-- generated_from_trainer
-model-index:
-  name: d-l-dl
----
+
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 should probably proofread and complete it, then remove this comment. -->

config.json

@@ -1,6 +1,6 @@
 {
-  "_name_or_path": "facebook/wav2vec2-base-960h",
-  "activation_dropout": 0.1,
+  "_name_or_path": "jonatasgrosman/wav2vec2-large-xlsr-53-german",
+  "activation_dropout": 0.05,
   "apply_spec_augment": true,
   "architectures": [
     "Wav2Vec2ForCTC"
@@ -8,9 +8,9 @@
   "attention_dropout": 0.5,
   "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,
@@ -39,42 +39,48 @@
     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_dropout": 0.0,
-  "feat_extract_norm": "group",
+  "feat_extract_norm": "layer",
   "feat_proj_dropout": 0.0,
   "feat_quantizer_dropout": 0.0,
-  "final_dropout": 0.1,
-  "gradient_checkpointing": false,
+  "final_dropout": 0.0,
   "hidden_act": "gelu",
   "hidden_dropout": 0.0,
-  "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_prob": 0.0,
   "mask_time_length": 10,
+  "mask_time_min_space": 1,
+  "mask_time_other": 0.0,
   "mask_time_prob": 0.05,
+  "mask_time_selection": "static",
   "model_type": "wav2vec2",
-  "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": 24,
-  "proj_codevector_dim": 256,
+  "proj_codevector_dim": 768,
   "torch_dtype": "float32",
   "transformers_version": "4.11.3",
   "use_weighted_layer_sum": false,
-  "vocab_size": 32
+  "vocab_size": 38
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:af2867f74cc5fb8521e8001af14d411e5fcf0f238969256724df83dec44cadbc
-size 377670039
+oid sha256:3d79e5fd2754fa90ea4955d2e1eb959dac8e0055d81761ad729d3ffbfd110abc
+size 1262079473

runs/May17_01-31-58_0d0f38c34c7b/1652751127.5893614/events.out.tfevents.1652751127.0d0f38c34c7b.1404.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bcb35f8a1c11ec7a28550c52897db6ebc944e6524467107219928c0b33895f6a
+size 4483

runs/May17_01-31-58_0d0f38c34c7b/events.out.tfevents.1652751127.0d0f38c34c7b.1404.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:904ae5272ffe16c7c21ca65d7e3190c38e14af4441a6fba4354f63069808b004
+size 11921

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:19f0c582e5da6613da0735544fb140f0f7217fc13b5e3427db24f1d7c8b22440
+oid sha256:e9eb099522910f5f0deff4d2bd83132cc3a06dd1e0148901d06b58b6760016e1
 size 2799