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 = 0.8259860788863109
|
2 |
+
cls_report = precision recall f1-score support
|
3 |
+
|
4 |
+
0.0 0.8315 0.8252 0.8284 658
|
5 |
+
1.0 0.8203 0.8268 0.8235 635
|
6 |
+
|
7 |
+
accuracy 0.8260 1293
|
8 |
+
macro avg 0.8259 0.8260 0.8260 1293
|
9 |
+
weighted avg 0.8260 0.8260 0.8260 1293
|
10 |
+
|
11 |
+
eval_loss = 0.4010350993478004
|
12 |
+
fn = 110
|
13 |
+
fp = 115
|
14 |
+
macro_f1 = 0.8259523489029479
|
15 |
+
mcc = 0.6519294085063897
|
16 |
+
tn = 543
|
17 |
+
tp = 525
|
18 |
+
weighted_f1 = 0.825995448326134
|
19 |
+
weighted_p = 0.8259296037519143
|
20 |
+
weighted_r = 0.8259998085345714
|
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_ET_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": "ET", "special_tags": ["</e>"], "merge_n": 2, "merge_type": "concat"}
|
optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1797589fc44bb93a9cd1adbca56605528b25cf694bc5779a646f41ecb1d40337
|
3 |
+
size 202375168
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:30b9ca91460ac511db5ae2f077ca4e63636440b883e2ec87537f255c51bc1963
|
3 |
+
size 2256539453
|
scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f6bc4e6cc022a189642a4ae14da94cafc01f88094a2eb890f4f74a63dbc33871
|
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 0.8689 0.7160 0.7851 500
|
5 |
+
T 0.7585 0.8920 0.8199 500
|
6 |
+
|
7 |
+
accuracy 0.8040 1000
|
8 |
+
macro avg 0.8137 0.8040 0.8025 1000
|
9 |
+
weighted avg 0.8137 0.8040 0.8025 1000
|
10 |
+
|
11 |
+
|
12 |
+
ADJ
|
13 |
+
Accuracy = 0.7653061224489796
|
14 |
+
Weighted Recall = 0.7653061224489796
|
15 |
+
Weighted Precision = 0.7779066171923315
|
16 |
+
Weighted F1 = 0.7650861656752672
|
17 |
+
Macro Recall = 0.7712788259958071
|
18 |
+
Macro Precision = 0.7723063973063973
|
19 |
+
Macro F1 = 0.765281682807456
|
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 = 0.8178137651821862
|
30 |
+
Weighted Recall = 0.8178137651821862
|
31 |
+
Weighted Precision = 0.8260395879697684
|
32 |
+
Weighted F1 = 0.8164280894544051
|
33 |
+
Macro Recall = 0.8168032786885246
|
34 |
+
Macro Precision = 0.8266565246788371
|
35 |
+
Macro F1 = 0.8162202380952381
|
36 |
+
VERB
|
37 |
+
Accuracy = 0.7964824120603015
|
38 |
+
Weighted Recall = 0.7964824120603015
|
39 |
+
Weighted Precision = 0.807451369836466
|
40 |
+
Weighted F1 = 0.7949028107834089
|
41 |
+
Macro Recall = 0.7973962673939945
|
42 |
+
Macro Precision = 0.8068584531999166
|
43 |
+
Macro F1 = 0.7950735784890188
|
test_eval_en.txt
ADDED
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
Default classification report:
|
2 |
+
precision recall f1-score support
|
3 |
+
|
4 |
+
F 0.8918 0.8900 0.8909 500
|
5 |
+
T 0.8902 0.8920 0.8911 500
|
6 |
+
|
7 |
+
accuracy 0.8910 1000
|
8 |
+
macro avg 0.8910 0.8910 0.8910 1000
|
9 |
+
weighted avg 0.8910 0.8910 0.8910 1000
|
10 |
+
|
11 |
+
|
12 |
+
ADJ
|
13 |
+
Accuracy = 0.8611111111111112
|
14 |
+
Weighted Recall = 0.8611111111111112
|
15 |
+
Weighted Precision = 0.8616044616044616
|
16 |
+
Weighted F1 = 0.8611916264090177
|
17 |
+
Macro Recall = 0.8614551083591331
|
18 |
+
Macro Precision = 0.8606177606177606
|
19 |
+
Macro F1 = 0.8608695652173913
|
20 |
+
ADV
|
21 |
+
Accuracy = 0.7666666666666667
|
22 |
+
Weighted Recall = 0.7666666666666667
|
23 |
+
Weighted Precision = 0.8126696832579187
|
24 |
+
Weighted F1 = 0.7679644048943269
|
25 |
+
Macro Recall = 0.7916666666666666
|
26 |
+
Macro Precision = 0.7850678733031675
|
27 |
+
Macro F1 = 0.7664071190211346
|
28 |
+
NOUN
|
29 |
+
Accuracy = 0.8920454545454546
|
30 |
+
Weighted Recall = 0.8920454545454546
|
31 |
+
Weighted Precision = 0.892332073969671
|
32 |
+
Weighted F1 = 0.892031900152266
|
33 |
+
Macro Recall = 0.8920941243991678
|
34 |
+
Macro Precision = 0.89229112833764
|
35 |
+
Macro F1 = 0.8920357728360341
|
36 |
+
VERB
|
37 |
+
Accuracy = 0.9161073825503355
|
38 |
+
Weighted Recall = 0.9161073825503355
|
39 |
+
Weighted Precision = 0.9176311030741412
|
40 |
+
Weighted F1 = 0.9160307924664405
|
41 |
+
Macro Recall = 0.9161073825503356
|
42 |
+
Macro Precision = 0.9176311030741411
|
43 |
+
Macro F1 = 0.9160307924664405
|
test_eval_fr.txt
ADDED
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
Default classification report:
|
2 |
+
precision recall f1-score support
|
3 |
+
|
4 |
+
F 0.8229 0.8180 0.8205 500
|
5 |
+
T 0.8191 0.8240 0.8215 500
|
6 |
+
|
7 |
+
accuracy 0.8210 1000
|
8 |
+
macro avg 0.8210 0.8210 0.8210 1000
|
9 |
+
weighted avg 0.8210 0.8210 0.8210 1000
|
10 |
+
|
11 |
+
|
12 |
+
ADJ
|
13 |
+
Accuracy = 0.7717391304347826
|
14 |
+
Weighted Recall = 0.7717391304347826
|
15 |
+
Weighted Precision = 0.7866528533419082
|
16 |
+
Weighted F1 = 0.7717391304347826
|
17 |
+
Macro Recall = 0.7791959918883454
|
18 |
+
Macro Precision = 0.7791959918883454
|
19 |
+
Macro F1 = 0.7717391304347826
|
20 |
+
ADV
|
21 |
+
Accuracy = 0.9333333333333333
|
22 |
+
Weighted Recall = 0.9333333333333333
|
23 |
+
Weighted Precision = 0.9391304347826086
|
24 |
+
Weighted F1 = 0.930681818181818
|
25 |
+
Macro Recall = 0.8888888888888888
|
26 |
+
Macro Precision = 0.9565217391304348
|
27 |
+
Macro F1 = 0.9147727272727273
|
28 |
+
NOUN
|
29 |
+
Accuracy = 0.8093385214007782
|
30 |
+
Weighted Recall = 0.8093385214007782
|
31 |
+
Weighted Precision = 0.8122439235818186
|
32 |
+
Weighted F1 = 0.8086977643026878
|
33 |
+
Macro Recall = 0.8085048384898459
|
34 |
+
Macro Precision = 0.8126909085327481
|
35 |
+
Macro F1 = 0.808500608272506
|
36 |
+
VERB
|
37 |
+
Accuracy = 0.8639705882352942
|
38 |
+
Weighted Recall = 0.8639705882352942
|
39 |
+
Weighted Precision = 0.8639561464316365
|
40 |
+
Weighted F1 = 0.8633751204730145
|
41 |
+
Macro Recall = 0.8575958092087124
|
42 |
+
Macro Precision = 0.8639169472502806
|
43 |
+
Macro F1 = 0.8601425811920678
|
test_eval_ru.txt
ADDED
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
Default classification report:
|
2 |
+
precision recall f1-score support
|
3 |
+
|
4 |
+
F 0.7434 0.7360 0.7397 500
|
5 |
+
T 0.7386 0.7460 0.7423 500
|
6 |
+
|
7 |
+
accuracy 0.7410 1000
|
8 |
+
macro avg 0.7410 0.7410 0.7410 1000
|
9 |
+
weighted avg 0.7410 0.7410 0.7410 1000
|
10 |
+
|
11 |
+
|
12 |
+
ADJ
|
13 |
+
Accuracy = 0.8
|
14 |
+
Weighted Recall = 0.8
|
15 |
+
Weighted Precision = 0.8355555555555555
|
16 |
+
Weighted F1 = 0.8036199095022625
|
17 |
+
Macro Recall = 0.8229665071770335
|
18 |
+
Macro Precision = 0.8
|
19 |
+
Macro F1 = 0.7963800904977375
|
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 = 0.7405498281786942
|
30 |
+
Weighted Recall = 0.7405498281786942
|
31 |
+
Weighted Precision = 0.742101056792141
|
32 |
+
Weighted F1 = 0.7405352748319636
|
33 |
+
Macro Recall = 0.7413120567375886
|
34 |
+
Macro Precision = 0.7413662642910346
|
35 |
+
Macro F1 = 0.7405490622130768
|
36 |
+
VERB
|
37 |
+
Accuracy = 0.7473118279569892
|
38 |
+
Weighted Recall = 0.7473118279569892
|
39 |
+
Weighted Precision = 0.7477007490228008
|
40 |
+
Weighted F1 = 0.7473118279569892
|
41 |
+
Macro Recall = 0.7475062884898951
|
42 |
+
Macro Precision = 0.7475062884898951
|
43 |
+
Macro F1 = 0.7473118279569891
|
test_eval_zh.txt
ADDED
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
Default classification report:
|
2 |
+
precision recall f1-score support
|
3 |
+
|
4 |
+
F 0.6245 0.6320 0.6282 500
|
5 |
+
T 0.6275 0.6200 0.6237 500
|
6 |
+
|
7 |
+
accuracy 0.6260 1000
|
8 |
+
macro avg 0.6260 0.6260 0.6260 1000
|
9 |
+
weighted avg 0.6260 0.6260 0.6260 1000
|
10 |
+
|
11 |
+
|
12 |
+
ADJ
|
13 |
+
Accuracy = 0.5645161290322581
|
14 |
+
Weighted Recall = 0.5645161290322581
|
15 |
+
Weighted Precision = 0.6191455972101134
|
16 |
+
Weighted F1 = 0.5659892724937674
|
17 |
+
Macro Recall = 0.5910087719298246
|
18 |
+
Macro Precision = 0.5897297297297297
|
19 |
+
Macro F1 = 0.5644028103044496
|
20 |
+
ADV
|
21 |
+
Accuracy = 0.7
|
22 |
+
Weighted Recall = 0.7
|
23 |
+
Weighted Precision = 0.8800000000000001
|
24 |
+
Weighted F1 = 0.7296703296703297
|
25 |
+
Macro Recall = 0.8125
|
26 |
+
Macro Precision = 0.7
|
27 |
+
Macro F1 = 0.6703296703296704
|
28 |
+
NOUN
|
29 |
+
Accuracy = 0.628158844765343
|
30 |
+
Weighted Recall = 0.628158844765343
|
31 |
+
Weighted Precision = 0.6309777120497468
|
32 |
+
Weighted F1 = 0.627460259983564
|
33 |
+
Macro Recall = 0.6293818466353678
|
34 |
+
Macro Precision = 0.6303610848312835
|
35 |
+
Macro F1 = 0.6277658908255923
|
36 |
+
VERB
|
37 |
+
Accuracy = 0.6291208791208791
|
38 |
+
Weighted Recall = 0.6291208791208791
|
39 |
+
Weighted Precision = 0.6292541883790845
|
40 |
+
Weighted F1 = 0.6274601211310071
|
41 |
+
Macro Recall = 0.6269946808510638
|
42 |
+
Macro Precision = 0.6292962860395704
|
43 |
+
Macro F1 = 0.6264112213479301
|
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 sha256:1138f7df8badeff6e636af9da2bb3d6bb0e97fd127c0af9928592de320bb3432
|
3 |
+
size 2811
|