Training in progress, step 40

config.json

@@ -1,6 +1,6 @@
 {
-  "_name_or_path": "w11wo/wav2vec2-xls-r-300m-zh-HK-lm-v2",
-  "activation_dropout": 0.1,
+  "_name_or_path": "ctl/wav2vec2-large-xlsr-cantonese",
+  "activation_dropout": 0.0,
   "adapter_attn_dim": null,
   "adapter_kernel_size": 3,
   "adapter_stride": 2,
@@ -12,7 +12,7 @@
   "attention_dropout": 0.0,
   "bos_token_id": 1,
   "classifier_proj_size": 256,
-  "codevector_dim": 768,
+  "codevector_dim": 256,
   "contrastive_logits_temperature": 0.1,
   "conv_bias": true,
   "conv_dim": [
@@ -60,12 +60,20 @@
   "intermediate_size": 4096,
   "layer_norm_eps": 1e-05,
   "layerdrop": 0.0,
-  "mask_feature_length": 64,
+  "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": 0.25,
+  "mask_feature_prob": 0.0,
   "mask_time_length": 10,
   "mask_time_min_masks": 2,
-  "mask_time_prob": 0.0,
+  "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": 16,
@@ -77,8 +85,8 @@
   "num_hidden_layers": 24,
   "num_negatives": 100,
   "output_hidden_size": 1024,
-  "pad_token_id": 3580,
-  "proj_codevector_dim": 768,
+  "pad_token_id": 50001,
+  "proj_codevector_dim": 256,
   "tdnn_dilation": [
     1,
     2,
@@ -103,6 +111,6 @@
   "torch_dtype": "float32",
   "transformers_version": "4.35.2",
   "use_weighted_layer_sum": false,
-  "vocab_size": 3583,
+  "vocab_size": 50002,
   "xvector_output_dim": 512
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:138922249ab18bee86ade0ea53c2d60f6e66a1dda527b02042014a32de9bd208
-size 1276497820
+oid sha256:4d55b7c2729dd00b0c41cef376a9a4cb50d6f0c02cb49ee756cd71efb86b7846
+size 1466815904

runs/Dec17_21-23-45_1d2c540adda9/events.out.tfevents.1702848241.1d2c540adda9.3058.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d912a2e8b9a0e6c1b0adc736943b60b67ca8b1663ca5689cc845dc5b24a188c9
+size 6373

runs/Dec17_21-25-51_1d2c540adda9/events.out.tfevents.1702848363.1d2c540adda9.3058.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ee322a4486c14b77840a2dab0b0dfb0f20c21ddcc29dbf49260fe2f41d8794de
+size 6025

runs/Dec17_21-27-14_1d2c540adda9/events.out.tfevents.1702848509.1d2c540adda9.3058.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c612af638ab61bfc25a88c6647d7f21d758414bba7a2f758a29eecb3772bd7a
+size 6378

runs/Dec17_21-33-47_1d2c540adda9/events.out.tfevents.1702848840.1d2c540adda9.3058.3

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e760ce57d31896f3a0a4b8fc6756bc3ba2aa97e8fa950bc385ccd04315a68d23
+size 6487

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6f437f05240d6cabc870e307679be862f3c6e7e2e5c2f4afe923ef56011fb65f
+oid sha256:18f172ebe09e4f010e32b630c02cbefe3b4d4549671684204f01a54ba6044d71
 size 4728