End of training

README.md

@@ -0,0 +1,60 @@
+---
+library_name: transformers
+language:
+- multilingual
+license: cc-by-nc-4.0
+base_model: facebook/seamless-m4t-v2-large
+tags:
+- hi,pa,ta,te,ml
+- generated_from_trainer
+datasets:
+- google/fleurs
+model-index:
+- name: "Seamless M4T \u2013 FLEURS Adrican Multilingual Fine\u2011tuning"
+  results: []
+---
+
+<!-- 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. -->
+
+# Seamless M4T – FLEURS Adrican Multilingual Fine‑tuning
+
+This model is a fine-tuned version of [facebook/seamless-m4t-v2-large](https://huggingface.co/facebook/seamless-m4t-v2-large) on the FLEURS dataset.
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 2e-05
+- train_batch_size: 32
+- eval_batch_size: 16
+- seed: 42
+- optimizer: Use adafactor and the args are:
+No additional optimizer arguments
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_steps: 25
+- training_steps: 925
+
+### Training results
+
+
+
+### Framework versions
+
+- Transformers 4.48.3
+- Pytorch 2.6.0+cu124
+- Datasets 3.3.2
+- Tokenizers 0.21.0

config.json

@@ -0,0 +1,118 @@
+{
+  "_name_or_path": "facebook/seamless-m4t-v2-large",
+  "activation_dropout": 0.0,
+  "activation_function": "relu",
+  "adaptor_dropout": 0.1,
+  "adaptor_kernel_size": 8,
+  "adaptor_stride": 8,
+  "add_adapter": true,
+  "architectures": [
+    "SeamlessM4Tv2ForSpeechToText"
+  ],
+  "attention_dropout": 0.1,
+  "bos_token_id": 2,
+  "char_vocab_size": 10943,
+  "conv_depthwise_kernel_size": 31,
+  "decoder_attention_heads": 16,
+  "decoder_ffn_dim": 8192,
+  "decoder_layerdrop": 0.05,
+  "decoder_layers": 24,
+  "decoder_start_token_id": 3,
+  "dropout": 0.1,
+  "encoder_attention_heads": 16,
+  "encoder_ffn_dim": 8192,
+  "encoder_layerdrop": 0.05,
+  "encoder_layers": 24,
+  "eos_token_id": 3,
+  "feature_projection_input_dim": 160,
+  "hidden_size": 1024,
+  "initializer_range": 0.02,
+  "is_encoder_decoder": true,
+  "lang_embed_dim": 256,
+  "layer_norm_eps": 1e-05,
+  "leaky_relu_slope": 0.1,
+  "left_max_position_embeddings": 64,
+  "max_new_tokens": 256,
+  "max_position_embeddings": 4096,
+  "model_type": "seamless_m4t_v2",
+  "num_adapter_layers": 1,
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "pad_token_id": 0,
+  "position_embeddings_type": "relative_key",
+  "resblock_dilation_sizes": [
+    [
+      1,
+      3,
+      5
+    ],
+    [
+      1,
+      3,
+      5
+    ],
+    [
+      1,
+      3,
+      5
+    ]
+  ],
+  "resblock_kernel_sizes": [
+    3,
+    7,
+    11
+  ],
+  "right_max_position_embeddings": 8,
+  "sampling_rate": 16000,
+  "scale_embedding": true,
+  "speech_encoder_attention_heads": 16,
+  "speech_encoder_chunk_size": 20000,
+  "speech_encoder_dropout": 0.0,
+  "speech_encoder_hidden_act": "swish",
+  "speech_encoder_intermediate_size": 4096,
+  "speech_encoder_layerdrop": 0.1,
+  "speech_encoder_layers": 24,
+  "speech_encoder_left_chunk_num": 128,
+  "spkr_embed_dim": 256,
+  "t2u_bos_token_id": 0,
+  "t2u_decoder_attention_heads": 16,
+  "t2u_decoder_ffn_dim": 8192,
+  "t2u_decoder_layers": 6,
+  "t2u_encoder_attention_heads": 16,
+  "t2u_encoder_ffn_dim": 8192,
+  "t2u_encoder_layers": 6,
+  "t2u_eos_token_id": 2,
+  "t2u_max_position_embeddings": 4096,
+  "t2u_pad_token_id": 1,
+  "t2u_variance_pred_dropout": 0.5,
+  "t2u_variance_predictor_embed_dim": 1024,
+  "t2u_variance_predictor_hidden_dim": 256,
+  "t2u_variance_predictor_kernel_size": 3,
+  "t2u_vocab_size": 10082,
+  "torch_dtype": "float32",
+  "transformers_version": "4.48.3",
+  "unit_embed_dim": 1280,
+  "unit_hifi_gan_vocab_size": 10000,
+  "upsample_initial_channel": 512,
+  "upsample_kernel_sizes": [
+    11,
+    8,
+    8,
+    4,
+    4
+  ],
+  "upsample_rates": [
+    5,
+    4,
+    4,
+    2,
+    2
+  ],
+  "use_cache": true,
+  "var_pred_dropout": 0.5,
+  "variance_predictor_kernel_size": 3,
+  "vocab_size": 256102,
+  "vocoder_num_langs": 36,
+  "vocoder_num_spkrs": 200,
+  "vocoder_offset": 4
+}

model-00001-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d3c1294702ae4e079ef12a331034455f1517fd87f7dd65dd2ec22c50a4081ac1
+size 4999937160

model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:44eda871b76ab2ca3d3c1f43a1685a172b2b76b03ee93d8f4006348e5a81d9c5
+size 1007614496

runs/Apr13_10-02-34_cn-g017.server.mila.quebec/events.out.tfevents.1744552965.cn-g017.server.mila.quebec.3425674.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2e285412aefb9d848ac294d71b242934cf6c13ca12f5e51657fa3fe6733bec40
+size 8194

runs/Apr13_12-04-30_cn-g008.server.mila.quebec/events.out.tfevents.1744560294.cn-g008.server.mila.quebec.521158.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a02a3596af34e695f104b843edcc41062236b294bc8344c970500fd100ef5071
+size 8202

runs/Apr13_14-06-56_cn-g008.server.mila.quebec/events.out.tfevents.1744567631.cn-g008.server.mila.quebec.525464.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1ff52b3660c12b8d018b9a6545daafb3664e557e328fbcd8bfddede686247be6
+size 8413

runs/Apr13_16-05-52_cn-g008.server.mila.quebec/events.out.tfevents.1744574767.cn-g008.server.mila.quebec.529659.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d6e72a69d6d1777494cfd8644f3d4af54e3d4f34ddb39dcebee4a04b7f1eefc4
+size 8768

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c9801bc9e9a5d4dc3f458101dfde18a040b1b590cd29c35642e5103a68d5fffd
+size 5496