Upload folder using huggingface_hub
Browse files- added_tokens.json +1 -0
- config.json +28 -0
- eval_results.txt +20 -0
- model_args.json +1 -0
- optimizer.pt +3 -0
- pytorch_model.bin +3 -0
- scheduler.pt +3 -0
- sentencepiece.bpe.model +3 -0
- special_tokens_map.json +1 -0
- test_eval_ar.txt +43 -0
- test_eval_en.txt +43 -0
- test_eval_fr.txt +43 -0
- test_eval_ru.txt +43 -0
- test_eval_zh.txt +43 -0
- tokenizer_config.json +1 -0
- training_args.bin +3 -0
added_tokens.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"<e>": 250002, "</e>": 250003}
|
config.json
ADDED
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "xlm-roberta-large",
|
3 |
+
"architectures": [
|
4 |
+
"XLMRobertaForSequenceClassification"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0.1,
|
7 |
+
"bos_token_id": 0,
|
8 |
+
"classifier_dropout": null,
|
9 |
+
"eos_token_id": 2,
|
10 |
+
"hidden_act": "gelu",
|
11 |
+
"hidden_dropout_prob": 0.1,
|
12 |
+
"hidden_size": 1024,
|
13 |
+
"initializer_range": 0.02,
|
14 |
+
"intermediate_size": 4096,
|
15 |
+
"layer_norm_eps": 1e-05,
|
16 |
+
"max_position_embeddings": 514,
|
17 |
+
"model_type": "xlm-roberta",
|
18 |
+
"num_attention_heads": 16,
|
19 |
+
"num_hidden_layers": 24,
|
20 |
+
"output_past": true,
|
21 |
+
"pad_token_id": 1,
|
22 |
+
"position_embedding_type": "absolute",
|
23 |
+
"torch_dtype": "float32",
|
24 |
+
"transformers_version": "4.16.2",
|
25 |
+
"type_vocab_size": 1,
|
26 |
+
"use_cache": true,
|
27 |
+
"vocab_size": 250004
|
28 |
+
}
|
eval_results.txt
ADDED
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
accuracy = 0.7896365042536736
|
2 |
+
cls_report = precision recall f1-score support
|
3 |
+
|
4 |
+
0.0 0.8238 0.7462 0.7831 658
|
5 |
+
1.0 0.7604 0.8346 0.7958 635
|
6 |
+
|
7 |
+
accuracy 0.7896 1293
|
8 |
+
macro avg 0.7921 0.7904 0.7894 1293
|
9 |
+
weighted avg 0.7927 0.7896 0.7893 1293
|
10 |
+
|
11 |
+
eval_loss = 0.4836594340517933
|
12 |
+
fn = 105
|
13 |
+
fp = 167
|
14 |
+
macro_f1 = 0.7894449473396842
|
15 |
+
mcc = 0.5825342982934826
|
16 |
+
tn = 491
|
17 |
+
tp = 530
|
18 |
+
weighted_f1 = 0.7893319778775878
|
19 |
+
weighted_p = 0.7921136125099901
|
20 |
+
weighted_r = 0.7904231385970371
|
model_args.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"adam_epsilon": 1e-08, "begin_tag": "<e>", "best_model_dir": "best_model", "cache_dir": "temp/cache_dir/", "config": {}, "custom_layer_parameters": [], "custom_parameter_groups": [], "dataloader_num_workers": 70, "do_lower_case": false, "dynamic_quantize": false, "early_stopping_consider_epochs": false, "early_stopping_delta": 0, "early_stopping_metric": "eval_loss", "early_stopping_metric_minimize": true, "early_stopping_patience": 10, "encoding": null, "end_tag": "</e>", "eval_batch_size": 8, "evaluate_during_training": true, "evaluate_during_training_silent": false, "evaluate_during_training_steps": 20, "evaluate_during_training_verbose": true, "evaluate_each_epoch": true, "fp16": false, "gradient_accumulation_steps": 1, "learning_rate": 1e-05, "local_rank": -1, "logging_steps": 20, "manual_seed": 777, "max_grad_norm": 1.0, "max_seq_length": 120, "model_name": "xlm-roberta-large", "model_type": "xlmroberta", "multiprocessing_chunksize": 500, "n_gpu": 1, "no_cache": false, "no_save": false, "num_train_epochs": 5, "output_dir": "temp/outputs/", "overwrite_output_dir": true, "process_count": 70, "quantized_model": false, "reprocess_input_data": true, "save_best_model": true, "save_eval_checkpoints": false, "save_model_every_epoch": false, "save_optimizer_and_scheduler": true, "save_steps": 20, "save_recent_only": true, "silent": false, "tensorboard_dir": null, "thread_count": null, "train_batch_size": 8, "train_custom_parameters_only": false, "use_cached_eval_features": false, "use_early_stopping": true, "use_multiprocessing": false, "wandb_kwargs": {"group": "all_xlm-roberta-large_CLS-P_concat", "job_type": "2"}, "wandb_project": "TransWiC-groups", "warmup_ratio": 0.1, "warmup_steps": 729, "weight_decay": 0, "skip_special_tokens": true, "model_class": "ClassificationModel", "labels_list": [0, 1], "labels_map": {}, "lazy_delimiter": "\t", "lazy_labels_column": 1, "lazy_loading": false, "lazy_loading_start_line": 1, "lazy_text_a_column": null, "lazy_text_b_column": null, "lazy_text_column": 0, "onnx": false, "regression": false, "sliding_window": false, "stride": 0.8, "tie_value": 1, "tagging": true, "strategy": "CLS-P", "special_tags": ["<s>", "<e>", "</e>"], "merge_n": 3, "merge_type": "concat"}
|
optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8f7c8b48be14c3f059b21a0d4169c49daaf3d3dfe4313bbe8d3642f73a79b008
|
3 |
+
size 4546546317
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:294a937c1e13bdb907569b589875ee1e6656310cc47055e2b1fba2ca9c24334a
|
3 |
+
size 2277523325
|
scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ae04e07925fc93947130bac6a34e7e16f6615f9f7a1b793cc7d449084fefc608
|
3 |
+
size 627
|
sentencepiece.bpe.model
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
|
3 |
+
size 5069051
|
special_tokens_map.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true}}
|
test_eval_ar.txt
ADDED
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
Default classification report:
|
2 |
+
precision recall f1-score support
|
3 |
+
|
4 |
+
F 0.8012 0.7820 0.7915 500
|
5 |
+
T 0.7871 0.8060 0.7964 500
|
6 |
+
|
7 |
+
accuracy 0.7940 1000
|
8 |
+
macro avg 0.7942 0.7940 0.7940 1000
|
9 |
+
weighted avg 0.7942 0.7940 0.7940 1000
|
10 |
+
|
11 |
+
|
12 |
+
ADJ
|
13 |
+
Accuracy = 0.7755102040816326
|
14 |
+
Weighted Recall = 0.7755102040816326
|
15 |
+
Weighted Precision = 0.776719198317625
|
16 |
+
Weighted F1 = 0.7757917548290219
|
17 |
+
Macro Recall = 0.7756813417190775
|
18 |
+
Macro Precision = 0.7743012098456403
|
19 |
+
Macro F1 = 0.7746655518394648
|
20 |
+
ADV
|
21 |
+
Accuracy = 0.6
|
22 |
+
Weighted Recall = 0.6
|
23 |
+
Weighted Precision = 0.7166666666666667
|
24 |
+
Weighted F1 = 0.6380952380952382
|
25 |
+
Macro Recall = 0.5625
|
26 |
+
Macro Precision = 0.5416666666666667
|
27 |
+
Macro F1 = 0.5238095238095238
|
28 |
+
NOUN
|
29 |
+
Accuracy = 0.7975708502024291
|
30 |
+
Weighted Recall = 0.7975708502024291
|
31 |
+
Weighted Precision = 0.7981559178688616
|
32 |
+
Weighted F1 = 0.7973914388629947
|
33 |
+
Macro Recall = 0.7972459016393443
|
34 |
+
Macro Precision = 0.7983021847854699
|
35 |
+
Macro F1 = 0.7973017331932772
|
36 |
+
VERB
|
37 |
+
Accuracy = 0.7989949748743719
|
38 |
+
Weighted Recall = 0.7989949748743719
|
39 |
+
Weighted Precision = 0.7989949748743719
|
40 |
+
Weighted F1 = 0.7989949748743719
|
41 |
+
Macro Recall = 0.798974669798217
|
42 |
+
Macro Precision = 0.798974669798217
|
43 |
+
Macro F1 = 0.798974669798217
|
test_eval_en.txt
ADDED
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
Default classification report:
|
2 |
+
precision recall f1-score support
|
3 |
+
|
4 |
+
F 0.8917 0.8560 0.8735 500
|
5 |
+
T 0.8615 0.8960 0.8784 500
|
6 |
+
|
7 |
+
accuracy 0.8760 1000
|
8 |
+
macro avg 0.8766 0.8760 0.8760 1000
|
9 |
+
weighted avg 0.8766 0.8760 0.8760 1000
|
10 |
+
|
11 |
+
|
12 |
+
ADJ
|
13 |
+
Accuracy = 0.8611111111111112
|
14 |
+
Weighted Recall = 0.8611111111111112
|
15 |
+
Weighted Precision = 0.8630125010927528
|
16 |
+
Weighted F1 = 0.8605431137076708
|
17 |
+
Macro Recall = 0.858359133126935
|
18 |
+
Macro Precision = 0.8642800944138473
|
19 |
+
Macro F1 = 0.8597857838364169
|
20 |
+
ADV
|
21 |
+
Accuracy = 0.8333333333333334
|
22 |
+
Weighted Recall = 0.8333333333333334
|
23 |
+
Weighted Precision = 0.8371040723981902
|
24 |
+
Weighted F1 = 0.8342857142857143
|
25 |
+
Macro Recall = 0.8333333333333334
|
26 |
+
Macro Precision = 0.8257918552036199
|
27 |
+
Macro F1 = 0.8285714285714285
|
28 |
+
NOUN
|
29 |
+
Accuracy = 0.8806818181818182
|
30 |
+
Weighted Recall = 0.8806818181818182
|
31 |
+
Weighted Precision = 0.8831406377909783
|
32 |
+
Weighted F1 = 0.8805108267244117
|
33 |
+
Macro Recall = 0.8808307626085086
|
34 |
+
Macro Precision = 0.8830235511429231
|
35 |
+
Macro F1 = 0.8805271116251171
|
36 |
+
VERB
|
37 |
+
Accuracy = 0.8791946308724832
|
38 |
+
Weighted Recall = 0.8791946308724832
|
39 |
+
Weighted Precision = 0.880290915661562
|
40 |
+
Weighted F1 = 0.879107505070994
|
41 |
+
Macro Recall = 0.8791946308724832
|
42 |
+
Macro Precision = 0.8802909156615621
|
43 |
+
Macro F1 = 0.8791075050709939
|
test_eval_fr.txt
ADDED
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
Default classification report:
|
2 |
+
precision recall f1-score support
|
3 |
+
|
4 |
+
F 0.8075 0.7720 0.7894 500
|
5 |
+
T 0.7816 0.8160 0.7984 500
|
6 |
+
|
7 |
+
accuracy 0.7940 1000
|
8 |
+
macro avg 0.7946 0.7940 0.7939 1000
|
9 |
+
weighted avg 0.7946 0.7940 0.7939 1000
|
10 |
+
|
11 |
+
|
12 |
+
ADJ
|
13 |
+
Accuracy = 0.7336956521739131
|
14 |
+
Weighted Recall = 0.7336956521739131
|
15 |
+
Weighted Precision = 0.7401544351937506
|
16 |
+
Weighted F1 = 0.7343021962264054
|
17 |
+
Macro Recall = 0.7370273171895503
|
18 |
+
Macro Precision = 0.734869976359338
|
19 |
+
Macro F1 = 0.7333096695950543
|
20 |
+
ADV
|
21 |
+
Accuracy = 0.8333333333333334
|
22 |
+
Weighted Recall = 0.8333333333333334
|
23 |
+
Weighted Precision = 0.8333333333333334
|
24 |
+
Weighted F1 = 0.8222222222222223
|
25 |
+
Macro Recall = 0.753968253968254
|
26 |
+
Macro Precision = 0.8333333333333334
|
27 |
+
Macro F1 = 0.7777777777777779
|
28 |
+
NOUN
|
29 |
+
Accuracy = 0.7879377431906615
|
30 |
+
Weighted Recall = 0.7879377431906615
|
31 |
+
Weighted Precision = 0.793574108409718
|
32 |
+
Weighted F1 = 0.7865909746404117
|
33 |
+
Macro Recall = 0.7867657080550634
|
34 |
+
Macro Precision = 0.7941834451901566
|
35 |
+
Macro F1 = 0.7862997814403576
|
36 |
+
VERB
|
37 |
+
Accuracy = 0.8419117647058824
|
38 |
+
Weighted Recall = 0.8419117647058824
|
39 |
+
Weighted Precision = 0.841672030765168
|
40 |
+
Weighted F1 = 0.8412197346037734
|
41 |
+
Macro Recall = 0.8350978770333609
|
42 |
+
Macro Precision = 0.8410213243546576
|
43 |
+
Macro F1 = 0.8374629997637543
|
test_eval_ru.txt
ADDED
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
Default classification report:
|
2 |
+
precision recall f1-score support
|
3 |
+
|
4 |
+
F 0.7296 0.6960 0.7124 500
|
5 |
+
T 0.7094 0.7420 0.7253 500
|
6 |
+
|
7 |
+
accuracy 0.7190 1000
|
8 |
+
macro avg 0.7195 0.7190 0.7189 1000
|
9 |
+
weighted avg 0.7195 0.7190 0.7189 1000
|
10 |
+
|
11 |
+
|
12 |
+
ADJ
|
13 |
+
Accuracy = 0.7
|
14 |
+
Weighted Recall = 0.7
|
15 |
+
Weighted Precision = 0.7241071428571428
|
16 |
+
Weighted F1 = 0.7051428571428572
|
17 |
+
Macro Recall = 0.7057416267942584
|
18 |
+
Macro Precision = 0.6919642857142857
|
19 |
+
Macro F1 = 0.6914285714285715
|
20 |
+
ADV
|
21 |
+
Accuracy = 0.5
|
22 |
+
Weighted Recall = 0.5
|
23 |
+
Weighted Precision = 0.5666666666666667
|
24 |
+
Weighted F1 = 0.5
|
25 |
+
Macro Recall = 0.5333333333333333
|
26 |
+
Macro Precision = 0.5333333333333333
|
27 |
+
Macro F1 = 0.5
|
28 |
+
NOUN
|
29 |
+
Accuracy = 0.7130584192439863
|
30 |
+
Weighted Recall = 0.7130584192439863
|
31 |
+
Weighted Precision = 0.7145478920529664
|
32 |
+
Weighted F1 = 0.7130423238207809
|
33 |
+
Macro Recall = 0.7137943262411347
|
34 |
+
Macro Precision = 0.7138423523013442
|
35 |
+
Macro F1 = 0.7130575721164492
|
36 |
+
VERB
|
37 |
+
Accuracy = 0.739247311827957
|
38 |
+
Weighted Recall = 0.739247311827957
|
39 |
+
Weighted Precision = 0.7398387096774194
|
40 |
+
Weighted F1 = 0.7388942161134983
|
41 |
+
Macro Recall = 0.7387024026368288
|
42 |
+
Macro Precision = 0.74
|
43 |
+
Macro F1 = 0.7387016184510662
|
test_eval_zh.txt
ADDED
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
Default classification report:
|
2 |
+
precision recall f1-score support
|
3 |
+
|
4 |
+
F 0.6097 0.6060 0.6078 500
|
5 |
+
T 0.6083 0.6120 0.6102 500
|
6 |
+
|
7 |
+
accuracy 0.6090 1000
|
8 |
+
macro avg 0.6090 0.6090 0.6090 1000
|
9 |
+
weighted avg 0.6090 0.6090 0.6090 1000
|
10 |
+
|
11 |
+
|
12 |
+
ADJ
|
13 |
+
Accuracy = 0.5806451612903226
|
14 |
+
Weighted Recall = 0.5806451612903226
|
15 |
+
Weighted Precision = 0.6020161290322581
|
16 |
+
Weighted F1 = 0.5859703020993343
|
17 |
+
Macro Recall = 0.5811403508771931
|
18 |
+
Macro Precision = 0.5770833333333334
|
19 |
+
Macro F1 = 0.5735449735449736
|
20 |
+
ADV
|
21 |
+
Accuracy = 0.5
|
22 |
+
Weighted Recall = 0.5
|
23 |
+
Weighted Precision = 0.8571428571428571
|
24 |
+
Weighted F1 = 0.5252525252525253
|
25 |
+
Macro Recall = 0.6875
|
26 |
+
Macro Precision = 0.6428571428571428
|
27 |
+
Macro F1 = 0.4949494949494949
|
28 |
+
NOUN
|
29 |
+
Accuracy = 0.6191335740072202
|
30 |
+
Weighted Recall = 0.6191335740072202
|
31 |
+
Weighted Precision = 0.6207595224496072
|
32 |
+
Weighted F1 = 0.6188766302094771
|
33 |
+
Macro Recall = 0.6199400104329682
|
34 |
+
Macro Precision = 0.6202646685758373
|
35 |
+
Macro F1 = 0.6190330307820164
|
36 |
+
VERB
|
37 |
+
Accuracy = 0.6043956043956044
|
38 |
+
Weighted Recall = 0.6043956043956044
|
39 |
+
Weighted Precision = 0.6041586804668201
|
40 |
+
Weighted F1 = 0.6042041443130372
|
41 |
+
Macro Recall = 0.6036025145067698
|
42 |
+
Macro Precision = 0.6038032945736435
|
43 |
+
Macro F1 = 0.6036297640653357
|
tokenizer_config.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "cls_token": "<s>", "pad_token": "<pad>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "sp_model_kwargs": {}, "do_lower_case": false, "model_max_length": 512, "special_tokens_map_file": null, "tokenizer_file": "/home/hh2/.cache/huggingface/transformers/7766c86e10505ed9b39af34e456480399bf06e35b36b8f2b917460a2dbe94e59.a984cf52fc87644bd4a2165f1e07e0ac880272c1e82d648b4674907056912bd7", "name_or_path": "xlm-roberta-large", "tokenizer_class": "XLMRobertaTokenizer"}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:fd9fcb3939406d190d6cdc7031742f55bc17db7733ffb0ca7404a404b70ea902
|
3 |
+
size 2875
|