Upload folder using huggingface_hub

all_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 12.54,
+    "perplexity": 1.7641247398494213,
+    "test_loss": 0.5676546692848206,
+    "test_runtime": 45.537,
+    "test_samples_per_second": 1407.449,
+    "test_steps_per_second": 2.46
+}

config.json

@@ -0,0 +1,27 @@
+{
+  "architectures": [
+    "RobertaForMaskedLM"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 514,
+  "model_type": "roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.25.1",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 32002
+}

french_log_final.out

@@ -0,0 +1,116 @@
+nohup: ignoring input
+sentencepiece_trainer.cc(77) LOG(INFO) Starts training with : 
+trainer_spec {
+  input: /home/jknafou/Corpus_translation/translated_corpus/cat_title_abstract_10M.txt
+  input_format: 
+  model_prefix: /home/jknafou/Language_model_training/French/Tokenizer/bio_french
+  model_type: UNIGRAM
+  vocab_size: 32000
+  self_test_sample_size: 0
+  character_coverage: 0.9995
+  input_sentence_size: 0
+  shuffle_input_sentence: 1
+  seed_sentencepiece_size: 1000000
+  shrinking_factor: 0.75
+  max_sentence_length: 4192
+  num_threads: 16
+  num_sub_iterations: 2
+  max_sentencepiece_length: 16
+  split_by_unicode_script: 1
+  split_by_number: 1
+  split_by_whitespace: 1
+  split_digits: 0
+  treat_whitespace_as_suffix: 0
+  required_chars: 
+  byte_fallback: 0
+  vocabulary_output_piece_score: 1
+  train_extremely_large_corpus: 1
+  hard_vocab_limit: 1
+  use_all_vocab: 0
+  unk_id: 0
+  bos_id: 1
+  eos_id: 2
+  pad_id: -1
+  unk_piece: <unk>
+  bos_piece: <s>
+  eos_piece: </s>
+  pad_piece: <pad>
+  unk_surface:  ⁇ 
+}
+normalizer_spec {
+  name: nmt_nfkc
+  add_dummy_prefix: 1
+  remove_extra_whitespaces: 1
+  escape_whitespaces: 1
+  normalization_rule_tsv: 
+}
+denormalizer_spec {}
+trainer_interface.cc(319) LOG(INFO) SentenceIterator is not specified. Using MultiFileSentenceIterator.
+trainer_interface.cc(174) LOG(INFO) Loading corpus: /home/jknafou/Corpus_translation/translated_corpus/cat_title_abstract_10M.txt
+trainer_interface.cc(346) LOG(WARNING) Found too long line (7312 > 4192).
+trainer_interface.cc(348) LOG(WARNING) Too long lines are skipped in the training.
+trainer_interface.cc(349) LOG(WARNING) The maximum length can be changed with --max_sentence_length=<size> flag.
+trainer_interface.cc(136) LOG(INFO) Loaded 1000000 lines
+trainer_interface.cc(136) LOG(INFO) Loaded 2000000 lines
+trainer_interface.cc(136) LOG(INFO) Loaded 3000000 lines
+trainer_interface.cc(136) LOG(INFO) Loaded 4000000 lines
+trainer_interface.cc(136) LOG(INFO) Loaded 5000000 lines
+trainer_interface.cc(136) LOG(INFO) Loaded 6000000 lines
+trainer_interface.cc(136) LOG(INFO) Loaded 7000000 lines
+trainer_interface.cc(136) LOG(INFO) Loaded 8000000 lines
+trainer_interface.cc(136) LOG(INFO) Loaded 9000000 lines
+trainer_interface.cc(113) LOG(WARNING) Too many sentences are loaded! (9978301), which may slow down training.
+trainer_interface.cc(115) LOG(WARNING) Consider using --input_sentence_size=<size> and --shuffle_input_sentence=true.
+trainer_interface.cc(118) LOG(WARNING) They allow to randomly sample <size> sentences from the entire corpus.
+trainer_interface.cc(375) LOG(INFO) Loaded all 9978301 sentences
+trainer_interface.cc(381) LOG(INFO) Skipped 21699 too long sentences.
+trainer_interface.cc(390) LOG(INFO) Adding meta_piece: <unk>
+trainer_interface.cc(390) LOG(INFO) Adding meta_piece: <s>
+trainer_interface.cc(390) LOG(INFO) Adding meta_piece: </s>
+trainer_interface.cc(395) LOG(INFO) Normalizing sentences...
+trainer_interface.cc(456) LOG(INFO) all chars count=16138226709
+trainer_interface.cc(467) LOG(INFO) Done: 99.9527% characters are covered.
+trainer_interface.cc(477) LOG(INFO) Alphabet size=95
+trainer_interface.cc(478) LOG(INFO) Final character coverage=0.999527
+trainer_interface.cc(510) LOG(INFO) Done! preprocessed 9978301 sentences.
+tcmalloc: large alloc 1073741824 bytes == 0x561532c8c000 @ 
+tcmalloc: large alloc 2147483648 bytes == 0x561572c8c000 @ 
+tcmalloc: large alloc 4294967296 bytes == 0x5615f3508000 @ 
+tcmalloc: large alloc 8589934592 bytes == 0x5616f3508000 @ 
+tcmalloc: large alloc 17179869184 bytes == 0x5618f3d08000 @ 
+tcmalloc: large alloc 34359738368 bytes == 0x561cf4d08000 @ 
+tcmalloc: large alloc 68719476736 bytes == 0x5624f6d08000 @ 
+tcmalloc: large alloc 129105821696 bytes == 0x5634fad08000 @ 
+tcmalloc: large alloc 129105821696 bytes == 0x5653119f0000 @ 
+tcmalloc: large alloc 129105821696 bytes == 0x5671286d8000 @ 
+tcmalloc: large alloc 129105821696 bytes == 0x568f3f3c0000 @ 
+unigram_model_trainer.cc(138) LOG(INFO) Making suffix array...
+unigram_model_trainer.cc(142) LOG(INFO) Extracting frequent sub strings...
+unigram_model_trainer.cc(193) LOG(INFO) Initialized 1000000 seed sentencepieces
+trainer_interface.cc(516) LOG(INFO) Tokenizing input sentences with whitespace: 9978301
+trainer_interface.cc(526) LOG(INFO) Done! 19224817
+unigram_model_trainer.cc(488) LOG(INFO) Using 19224817 sentences for EM training
+unigram_model_trainer.cc(504) LOG(INFO) EM sub_iter=0 size=739481 obj=12.0696 num_tokens=74362684 num_tokens/piece=100.561
+unigram_model_trainer.cc(504) LOG(INFO) EM sub_iter=1 size=599525 obj=9.09794 num_tokens=74303077 num_tokens/piece=123.937
+unigram_model_trainer.cc(504) LOG(INFO) EM sub_iter=0 size=449629 obj=9.06759 num_tokens=74685579 num_tokens/piece=166.105
+unigram_model_trainer.cc(504) LOG(INFO) EM sub_iter=1 size=449389 obj=9.06467 num_tokens=74741532 num_tokens/piece=166.318
+unigram_model_trainer.cc(504) LOG(INFO) EM sub_iter=0 size=337040 obj=9.05777 num_tokens=75208218 num_tokens/piece=223.143
+unigram_model_trainer.cc(504) LOG(INFO) EM sub_iter=1 size=337037 obj=9.06272 num_tokens=75221866 num_tokens/piece=223.186
+unigram_model_trainer.cc(504) LOG(INFO) EM sub_iter=0 size=252777 obj=9.05783 num_tokens=76327358 num_tokens/piece=301.955
+unigram_model_trainer.cc(504) LOG(INFO) EM sub_iter=1 size=252776 obj=9.06287 num_tokens=76319490 num_tokens/piece=301.925
+unigram_model_trainer.cc(504) LOG(INFO) EM sub_iter=0 size=189582 obj=9.07482 num_tokens=78333478 num_tokens/piece=413.19
+unigram_model_trainer.cc(504) LOG(INFO) EM sub_iter=1 size=189582 obj=9.0738 num_tokens=78313493 num_tokens/piece=413.085
+unigram_model_trainer.cc(504) LOG(INFO) EM sub_iter=0 size=142186 obj=9.10085 num_tokens=80614240 num_tokens/piece=566.963
+unigram_model_trainer.cc(504) LOG(INFO) EM sub_iter=1 size=142186 obj=9.09887 num_tokens=80596413 num_tokens/piece=566.838
+unigram_model_trainer.cc(504) LOG(INFO) EM sub_iter=0 size=106639 obj=9.13963 num_tokens=83102337 num_tokens/piece=779.287
+unigram_model_trainer.cc(504) LOG(INFO) EM sub_iter=1 size=106639 obj=9.12951 num_tokens=83086218 num_tokens/piece=779.135
+unigram_model_trainer.cc(504) LOG(INFO) EM sub_iter=0 size=79979 obj=9.19014 num_tokens=85866144 num_tokens/piece=1073.61
+unigram_model_trainer.cc(504) LOG(INFO) EM sub_iter=1 size=79979 obj=9.18657 num_tokens=85860214 num_tokens/piece=1073.53
+unigram_model_trainer.cc(504) LOG(INFO) EM sub_iter=0 size=59984 obj=9.25486 num_tokens=88825440 num_tokens/piece=1480.82
+unigram_model_trainer.cc(504) LOG(INFO) EM sub_iter=1 size=59984 obj=9.23964 num_tokens=88813706 num_tokens/piece=1480.62
+unigram_model_trainer.cc(504) LOG(INFO) EM sub_iter=0 size=44988 obj=9.33847 num_tokens=92202568 num_tokens/piece=2049.49
+unigram_model_trainer.cc(504) LOG(INFO) EM sub_iter=1 size=44988 obj=9.32441 num_tokens=92203259 num_tokens/piece=2049.51
+unigram_model_trainer.cc(504) LOG(INFO) EM sub_iter=0 size=35200 obj=9.4222 num_tokens=95280571 num_tokens/piece=2706.83
+unigram_model_trainer.cc(504) LOG(INFO) EM sub_iter=1 size=35200 obj=9.40217 num_tokens=95275736 num_tokens/piece=2706.7
+trainer_interface.cc(604) LOG(INFO) Saving model: /home/jknafou/Language_model_training/French/Tokenizer/bio_french.model
+trainer_interface.cc(615) LOG(INFO) Saving vocabs: /home/jknafou/Language_model_training/French/Tokenizer/bio_french.vocab

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:39e9a675ca00ae7e81b96bf1bd1ecd166ff01f05cc10d91ef69346509eaa038b
+size 442686457

sentencepiece.bpe.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14fa8a8b41dab2349cb013afd529b785bb2285db9c60018cc83b0124e36ce090
+size 830480

special_tokens_map.json

@@ -0,0 +1,15 @@
+{
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "unk_token": "<unk>"
+}

test_set_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 12.54,
+    "perplexity": 1.7641247398494213,
+    "test_loss": 0.5676546692848206,
+    "test_runtime": 45.537,
+    "test_samples_per_second": 1407.449,
+    "test_steps_per_second": 2.46
+}

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14fa8a8b41dab2349cb013afd529b785bb2285db9c60018cc83b0124e36ce090
+size 830480

tokenizer_config.json

@@ -0,0 +1,21 @@
+{
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": {
+    "__type": "AddedToken",
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "max_len": 512,
+  "model_max_length": 512,
+  "name_or_path": ".",
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "tokenizer_class": "XLMRobertaTokenizer",
+  "special_tokens_map_file": "./special_tokens_map.json",
+  "unk_token": "<unk>"
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fcf18dceeae42053b5fc7b3d47198f44bef599f599f61c36a74c2161f1a6833e
+size 3387