Training in progress epoch 0

.gitattributes

@@ -32,3 +32,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+source.spm filter=lfs diff=lfs merge=lfs -text
+target.spm filter=lfs diff=lfs merge=lfs -text

.gitignore

@@ -0,0 +1,2 @@
+step_*
+epoch_*

added_tokens.json

@@ -0,0 +1,3 @@
+{
+  "<brk>": 53017
+}

config.json

@@ -0,0 +1,46 @@
+{
+  "_name_or_path": "context-mt/iwslt17-marian-big-ctx4-cwd1-en-fr",
+  "activation_dropout": 0.0,
+  "activation_function": "relu",
+  "architectures": [
+    "MarianMTModel"
+  ],
+  "attention_dropout": 0.0,
+  "bad_words_ids": [
+    [
+      53016
+    ]
+  ],
+  "bos_token_id": 0,
+  "classifier_dropout": 0.0,
+  "d_model": 1024,
+  "decoder_attention_heads": 16,
+  "decoder_ffn_dim": 4096,
+  "decoder_layerdrop": 0.0,
+  "decoder_layers": 6,
+  "decoder_start_token_id": 53016,
+  "decoder_vocab_size": 53018,
+  "dropout": 0.1,
+  "encoder_attention_heads": 16,
+  "encoder_ffn_dim": 4096,
+  "encoder_layerdrop": 0.0,
+  "encoder_layers": 6,
+  "eos_token_id": 43311,
+  "forced_eos_token_id": 43311,
+  "init_std": 0.02,
+  "is_encoder_decoder": true,
+  "max_length": 512,
+  "max_position_embeddings": 1024,
+  "model_type": "marian",
+  "normalize_embedding": false,
+  "num_beams": 4,
+  "num_hidden_layers": 6,
+  "pad_token_id": 53016,
+  "scale_embedding": true,
+  "share_encoder_decoder_embeddings": true,
+  "static_position_embeddings": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.28.0",
+  "use_cache": true,
+  "vocab_size": 53018
+}

generation_config.json

@@ -0,0 +1,16 @@
+{
+  "_from_model_config": true,
+  "bad_words_ids": [
+    [
+      53016
+    ]
+  ],
+  "bos_token_id": 0,
+  "decoder_start_token_id": 53016,
+  "eos_token_id": 43311,
+  "forced_eos_token_id": 43311,
+  "max_length": 512,
+  "num_beams": 4,
+  "pad_token_id": 53016,
+  "transformers_version": "4.28.0"
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:708602effb1344627cc9ef18610f0b753d2af28b760f3f2b78e76170f1faeabb
+size 922889797

source.spm

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:18bb5a422a53a0bd699b005c8086cbd58a85c4be427fd57ca687c958588edcc9
+size 802408

special_tokens_map.json

@@ -0,0 +1,8 @@
+{
+  "additional_special_tokens": [
+    "<brk>"
+  ],
+  "eos_token": "</s>",
+  "pad_token": "<pad>",
+  "unk_token": "<unk>"
+}

target.spm

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6437b5600d2ae14888939e589773b09bc667c77e81326e20e3f57698a4078593
+size 819955

tokenizer_config.json

@@ -0,0 +1,12 @@
+{
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "</s>",
+  "model_max_length": 512,
+  "pad_token": "<pad>",
+  "separate_vocabs": false,
+  "source_lang": "en",
+  "sp_model_kwargs": {},
+  "target_lang": "fr",
+  "tokenizer_class": "MarianTokenizer",
+  "unk_token": "<unk>"
+}

translation_no_trainer/1686672825.1023133/events.out.tfevents.1686672825.a100gpu1.2120556.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c99959508e113708df845499d112e9099919e8f9e6577f6ebcfc50db73c02a94
+size 2237

translation_no_trainer/1686672825.1280608/hparams.yml

@@ -0,0 +1,43 @@
+checkpointing_steps: '1000'
+config_name: null
+context_size: 4
+context_word_dropout: 0.1
+dataset_config_name: sentences
+dataset_name: inseq/scat
+gradient_accumulation_steps: 2
+hub_model_id: context-mt/scat-marian-big-ctx4-cwd1-en-fr
+hub_token: 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
+ignore_pad_token_for_loss: true
+learning_rate: 5.0e-05
+logging_steps: '200'
+lr_scheduler_type: linear
+max_length: 128
+max_source_length: 512
+max_target_length: 128
+max_train_steps: 1388
+model_name_or_path: context-mt/iwslt17-marian-big-ctx4-cwd1-en-fr
+model_type: null
+num_beams: 5
+num_train_epochs: 2
+num_warmup_steps: 0
+output_dir: /scratch/p305238/scat-marian-big-ctx4-cwd1-en-fr
+overwrite_cache: false
+pad_to_max_length: true
+per_device_eval_batch_size: 8
+per_device_train_batch_size: 8
+predict_with_generate: true
+preprocessing_num_workers: null
+push_to_hub: true
+report_to: tensorboard
+resume_from_checkpoint: null
+sample_context: true
+seed: null
+source_lang: en_XX
+target_lang: fr_XX
+tokenizer_name: null
+train_file: null
+use_slow_tokenizer: false
+val_max_target_length: null
+validation_file: null
+weight_decay: 0.0
+with_tracking: true

translation_no_trainer/events.out.tfevents.1686672825.a100gpu1.2120556.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c28e1c66d07ad6ff1c4a3235badd0cd49a0357a7129f81c905bb63cfcac4f7d8
+size 1083