Training in progress, epoch 16

config.json

@@ -1,6 +1,6 @@
 {
-  "_name_or_path": "facebook/wav2vec2-base",
-  "activation_dropout": 0.0,
+  "_name_or_path": "ehcalabres/wav2vec2-lg-xlsr-en-speech-emotion-recognition",
+  "activation_dropout": 0.05,
   "adapter_attn_dim": null,
   "adapter_kernel_size": 3,
   "adapter_stride": 2,
@@ -14,7 +14,7 @@
   "classifier_proj_size": 256,
   "codevector_dim": 256,
   "contrastive_logits_temperature": 0.1,
-  "conv_bias": false,
+  "conv_bias": true,
   "conv_dim": [
     512,
     512,
@@ -42,32 +42,33 @@
     2,
     2
   ],
-  "ctc_loss_reduction": "sum",
-  "ctc_zero_infinity": false,
+  "ctc_loss_reduction": "mean",
+  "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,
+  "finetuning_task": "wav2vec2_clf",
   "hidden_act": "gelu",
-  "hidden_dropout": 0.1,
-  "hidden_size": 768,
+  "hidden_dropout": 0.05,
+  "hidden_size": 1024,
   "id2label": {
     "0": "deceptive",
     "1": "truthful"
   },
   "initializer_range": 0.02,
-  "intermediate_size": 3072,
+  "intermediate_size": 4096,
   "label2id": {
     "deceptive": "0",
     "truthful": "1"
   },
   "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,
@@ -83,19 +84,19 @@
   "mask_time_prob": 0.05,
   "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_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": 0,
+  "pooling_mode": "mean",
+  "problem_type": "single_label_classification",
   "proj_codevector_dim": 256,
   "tdnn_dilation": [
     1,
@@ -121,6 +122,6 @@
   "torch_dtype": "float32",
   "transformers_version": "4.46.3",
   "use_weighted_layer_sum": false,
-  "vocab_size": 32,
+  "vocab_size": 33,
   "xvector_output_dim": 512
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d05af6757de1f0012830d48f72038610d971bf5d1b7521b676d96fa864346469
-size 378302360
+oid sha256:68b2855d42a8cb2183d905b820e1c80526f5cd4b9f0560b2ce1d4b18429535b4
+size 1262859296

runs/Dec15_19-27-34_c40aaf450fcd/events.out.tfevents.1734290854.c40aaf450fcd.404.5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:226e077266f6c79ee6b4f3e574383bd40a8b4bc312161936be41cee346dbcedf
-size 15631
+oid sha256:15a53c12940b3ba6ef1335f3830c65d09192dc4e5338443dc8f74b06c7d6061e
+size 16811

runs/Dec15_19-31-56_c40aaf450fcd/events.out.tfevents.1734291117.c40aaf450fcd.404.6

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5d892712438f256eb555674dd284c383c750e820ab82e74895117397fbfac055
+size 14661

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b623cadf1ddbdbfed56a8d4ca82b17034a85f5648091319b4a8f8659a5c4aaf0
+oid sha256:1ad0d5c1cea1ce7c0b4714fd138f0b864e97b854b4236e272bcb440c7df9fffa
 size 5304