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 = 0.8081979891724671
|
2 |
+
cls_report = precision recall f1-score support
|
3 |
+
|
4 |
+
0.0 0.8328 0.7796 0.8053 658
|
5 |
+
1.0 0.7858 0.8378 0.8110 635
|
6 |
+
|
7 |
+
accuracy 0.8082 1293
|
8 |
+
macro avg 0.8093 0.8087 0.8082 1293
|
9 |
+
weighted avg 0.8097 0.8082 0.8081 1293
|
10 |
+
|
11 |
+
eval_loss = 0.42931405634239866
|
12 |
+
fn = 103
|
13 |
+
fp = 145
|
14 |
+
macro_f1 = 0.8081565646896658
|
15 |
+
mcc = 0.6180209851479913
|
16 |
+
tn = 513
|
17 |
+
tp = 532
|
18 |
+
weighted_f1 = 0.8081064192631168
|
19 |
+
weighted_p = 0.8093060004987627
|
20 |
+
weighted_r = 0.8087152669746069
|
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-E_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-E", "special_tags": ["<s>", "</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 sha256:4bf5151cd5a000debcb638afd0ba7f370479b66af82266b7cca3763a3e94dd93
|
3 |
+
size 4546546317
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:108de007bdafee022da6bba2a213ecdb2a7267d9b3eb6f821aa704976a0f3944
|
3 |
+
size 2277523261
|
scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d982096ff59f135a044ca1a194f1a77d0ca582f006400eac142ca7a5bec50d6f
|
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 0.8816 0.7300 0.7987 500
|
5 |
+
T 0.7696 0.9020 0.8306 500
|
6 |
+
|
7 |
+
accuracy 0.8160 1000
|
8 |
+
macro avg 0.8256 0.8160 0.8146 1000
|
9 |
+
weighted avg 0.8256 0.8160 0.8146 1000
|
10 |
+
|
11 |
+
|
12 |
+
ADJ
|
13 |
+
Accuracy = 0.7755102040816326
|
14 |
+
Weighted Recall = 0.7755102040816326
|
15 |
+
Weighted Precision = 0.790956551753894
|
16 |
+
Weighted F1 = 0.7750425170068028
|
17 |
+
Macro Recall = 0.7823899371069183
|
18 |
+
Macro Precision = 0.7847780126849895
|
19 |
+
Macro F1 = 0.7754166666666668
|
20 |
+
ADV
|
21 |
+
Accuracy = 0.8
|
22 |
+
Weighted Recall = 0.8
|
23 |
+
Weighted Precision = 0.64
|
24 |
+
Weighted F1 = 0.7111111111111111
|
25 |
+
Macro Recall = 0.5
|
26 |
+
Macro Precision = 0.4
|
27 |
+
Macro F1 = 0.4444444444444445
|
28 |
+
NOUN
|
29 |
+
Accuracy = 0.819838056680162
|
30 |
+
Weighted Recall = 0.819838056680162
|
31 |
+
Weighted Precision = 0.8260981862089894
|
32 |
+
Weighted F1 = 0.8187713058862275
|
33 |
+
Macro Recall = 0.8189508196721311
|
34 |
+
Macro Precision = 0.8266347687400319
|
35 |
+
Macro F1 = 0.818588434321553
|
36 |
+
VERB
|
37 |
+
Accuracy = 0.821608040201005
|
38 |
+
Weighted Recall = 0.821608040201005
|
39 |
+
Weighted Precision = 0.834766661304009
|
40 |
+
Weighted F1 = 0.820058819089243
|
41 |
+
Macro Recall = 0.8225749425461525
|
42 |
+
Macro Precision = 0.8340918602218037
|
43 |
+
Macro F1 = 0.8202177135622444
|
test_eval_en.txt
ADDED
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
Default classification report:
|
2 |
+
precision recall f1-score support
|
3 |
+
|
4 |
+
F 0.8980 0.8980 0.8980 500
|
5 |
+
T 0.8980 0.8980 0.8980 500
|
6 |
+
|
7 |
+
accuracy 0.8980 1000
|
8 |
+
macro avg 0.8980 0.8980 0.8980 1000
|
9 |
+
weighted avg 0.8980 0.8980 0.8980 1000
|
10 |
+
|
11 |
+
|
12 |
+
ADJ
|
13 |
+
Accuracy = 0.8819444444444444
|
14 |
+
Weighted Recall = 0.8819444444444444
|
15 |
+
Weighted Precision = 0.883352929649226
|
16 |
+
Weighted F1 = 0.8815713413612962
|
17 |
+
Macro Recall = 0.8796439628482973
|
18 |
+
Macro Precision = 0.8844797178130511
|
19 |
+
Macro F1 = 0.8809743764282589
|
20 |
+
ADV
|
21 |
+
Accuracy = 0.8
|
22 |
+
Weighted Recall = 0.8
|
23 |
+
Weighted Precision = 0.8107142857142856
|
24 |
+
Weighted F1 = 0.8018099547511313
|
25 |
+
Macro Recall = 0.8055555555555556
|
26 |
+
Macro Precision = 0.7946428571428572
|
27 |
+
Macro F1 = 0.7963800904977376
|
28 |
+
NOUN
|
29 |
+
Accuracy = 0.9015151515151515
|
30 |
+
Weighted Recall = 0.9015151515151515
|
31 |
+
Weighted Precision = 0.9017142072729536
|
32 |
+
Weighted F1 = 0.9014981908638625
|
33 |
+
Macro Recall = 0.901470693736997
|
34 |
+
Macro Precision = 0.9017473832325872
|
35 |
+
Macro F1 = 0.9014925373134328
|
36 |
+
VERB
|
37 |
+
Accuracy = 0.9093959731543624
|
38 |
+
Weighted Recall = 0.9093959731543624
|
39 |
+
Weighted Precision = 0.9094144144144144
|
40 |
+
Weighted F1 = 0.9093949528732137
|
41 |
+
Macro Recall = 0.9093959731543624
|
42 |
+
Macro Precision = 0.9094144144144144
|
43 |
+
Macro F1 = 0.9093949528732137
|
test_eval_fr.txt
ADDED
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
Default classification report:
|
2 |
+
precision recall f1-score support
|
3 |
+
|
4 |
+
F 0.8220 0.8220 0.8220 500
|
5 |
+
T 0.8220 0.8220 0.8220 500
|
6 |
+
|
7 |
+
accuracy 0.8220 1000
|
8 |
+
macro avg 0.8220 0.8220 0.8220 1000
|
9 |
+
weighted avg 0.8220 0.8220 0.8220 1000
|
10 |
+
|
11 |
+
|
12 |
+
ADJ
|
13 |
+
Accuracy = 0.7717391304347826
|
14 |
+
Weighted Recall = 0.7717391304347826
|
15 |
+
Weighted Precision = 0.781458497555346
|
16 |
+
Weighted F1 = 0.7721168670983658
|
17 |
+
Macro Recall = 0.777048789216271
|
18 |
+
Macro Precision = 0.775210332977841
|
19 |
+
Macro F1 = 0.7716312056737589
|
20 |
+
ADV
|
21 |
+
Accuracy = 0.9666666666666667
|
22 |
+
Weighted Recall = 0.9666666666666667
|
23 |
+
Weighted Precision = 0.9700000000000001
|
24 |
+
Weighted F1 = 0.9671373555840822
|
25 |
+
Macro Recall = 0.9761904761904762
|
26 |
+
Macro Precision = 0.95
|
27 |
+
Macro F1 = 0.9614890885750963
|
28 |
+
NOUN
|
29 |
+
Accuracy = 0.8093385214007782
|
30 |
+
Weighted Recall = 0.8093385214007782
|
31 |
+
Weighted Precision = 0.8103589227297893
|
32 |
+
Weighted F1 = 0.8090607958572607
|
33 |
+
Macro Recall = 0.8088077173534445
|
34 |
+
Macro Precision = 0.810614023062042
|
35 |
+
Macro F1 = 0.8089219330855018
|
36 |
+
VERB
|
37 |
+
Accuracy = 0.8639705882352942
|
38 |
+
Weighted Recall = 0.8639705882352942
|
39 |
+
Weighted Precision = 0.8638478766838301
|
40 |
+
Weighted F1 = 0.8638973764298372
|
41 |
+
Macro Recall = 0.8607389026743866
|
42 |
+
Macro Precision = 0.8615163572060124
|
43 |
+
Macro F1 = 0.861115327822475
|
test_eval_ru.txt
ADDED
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
Default classification report:
|
2 |
+
precision recall f1-score support
|
3 |
+
|
4 |
+
F 0.7588 0.7300 0.7441 500
|
5 |
+
T 0.7399 0.7680 0.7537 500
|
6 |
+
|
7 |
+
accuracy 0.7490 1000
|
8 |
+
macro avg 0.7494 0.7490 0.7489 1000
|
9 |
+
weighted avg 0.7494 0.7490 0.7489 1000
|
10 |
+
|
11 |
+
|
12 |
+
ADJ
|
13 |
+
Accuracy = 0.6666666666666666
|
14 |
+
Weighted Recall = 0.6666666666666666
|
15 |
+
Weighted Precision = 0.7022222222222222
|
16 |
+
Weighted F1 = 0.6726998491704373
|
17 |
+
Macro Recall = 0.6794258373205742
|
18 |
+
Macro Precision = 0.6666666666666667
|
19 |
+
Macro F1 = 0.6606334841628959
|
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 = 0.7542955326460481
|
30 |
+
Weighted Recall = 0.7542955326460481
|
31 |
+
Weighted Precision = 0.7553633641533805
|
32 |
+
Weighted F1 = 0.7543281757266053
|
33 |
+
Macro Recall = 0.7548581560283688
|
34 |
+
Macro Precision = 0.754722665248981
|
35 |
+
Macro F1 = 0.7542890040299366
|
36 |
+
VERB
|
37 |
+
Accuracy = 0.7580645161290323
|
38 |
+
Weighted Recall = 0.7580645161290323
|
39 |
+
Weighted Precision = 0.7588922957369351
|
40 |
+
Weighted F1 = 0.7576859990747361
|
41 |
+
Macro Recall = 0.7574811345303148
|
42 |
+
Macro Precision = 0.7590992406389108
|
43 |
+
Macro F1 = 0.757496740547588
|
test_eval_zh.txt
ADDED
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
Default classification report:
|
2 |
+
precision recall f1-score support
|
3 |
+
|
4 |
+
F 0.6122 0.6220 0.6171 500
|
5 |
+
T 0.6159 0.6060 0.6109 500
|
6 |
+
|
7 |
+
accuracy 0.6140 1000
|
8 |
+
macro avg 0.6140 0.6140 0.6140 1000
|
9 |
+
weighted avg 0.6140 0.6140 0.6140 1000
|
10 |
+
|
11 |
+
|
12 |
+
ADJ
|
13 |
+
Accuracy = 0.6129032258064516
|
14 |
+
Weighted Recall = 0.6129032258064516
|
15 |
+
Weighted Precision = 0.6532258064516129
|
16 |
+
Weighted F1 = 0.6169481241783799
|
17 |
+
Macro Recall = 0.6304824561403509
|
18 |
+
Macro Precision = 0.625
|
19 |
+
Macro F1 = 0.6112852664576802
|
20 |
+
ADV
|
21 |
+
Accuracy = 0.55
|
22 |
+
Weighted Recall = 0.55
|
23 |
+
Weighted Precision = 0.8615384615384615
|
24 |
+
Weighted F1 = 0.581074168797954
|
25 |
+
Macro Recall = 0.71875
|
26 |
+
Macro Precision = 0.6538461538461539
|
27 |
+
Macro F1 = 0.5396419437340154
|
28 |
+
NOUN
|
29 |
+
Accuracy = 0.6155234657039711
|
30 |
+
Weighted Recall = 0.6155234657039711
|
31 |
+
Weighted Precision = 0.6175788138872322
|
32 |
+
Weighted F1 = 0.6150835224819089
|
33 |
+
Macro Recall = 0.616510172143975
|
34 |
+
Macro Precision = 0.6170459458397202
|
35 |
+
Macro F1 = 0.6153116411896449
|
36 |
+
VERB
|
37 |
+
Accuracy = 0.6153846153846154
|
38 |
+
Weighted Recall = 0.6153846153846154
|
39 |
+
Weighted Precision = 0.615158371040724
|
40 |
+
Weighted F1 = 0.6140763997906855
|
41 |
+
Macro Recall = 0.6135154738878144
|
42 |
+
Macro Precision = 0.6150735294117647
|
43 |
+
Macro F1 = 0.613095238095238
|
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:ab934f00a1e0bd6b9aaa5a48a6db395ea72de95ef3b0ce3815634aaacb33ab16
|
3 |
+
size 2875
|