Vui Seng Chua
commited on
Commit
•
de654c9
1
Parent(s):
f9a16c1
add model collaterals
Browse files- README.md +57 -0
- all_results.json +8 -0
- config.json +37 -0
- eval_mnli-mm_results.json +8 -0
- eval_mnli_results.json +8 -0
- eval_results.json +9 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +1 -0
- tokenizer.json +0 -0
- tokenizer_config.json +1 -0
- train_results.json +8 -0
- trainer_state.json +0 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
README.md
ADDED
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
This model is developed with transformers v4.9.1.
|
2 |
+
|
3 |
+
```
|
4 |
+
m = 0.8444
|
5 |
+
eval_samples = 9815
|
6 |
+
|
7 |
+
mm = 0.8495
|
8 |
+
eval_samples = 9832
|
9 |
+
```
|
10 |
+
|
11 |
+
# Train
|
12 |
+
```bash
|
13 |
+
#!/usr/bin/env bash
|
14 |
+
|
15 |
+
export CUDA_VISIBLE_DEVICES=0
|
16 |
+
|
17 |
+
OUTDIR=bert-mnli
|
18 |
+
NEPOCH=3
|
19 |
+
|
20 |
+
WORKDIR=transformers/examples/pytorch/text-classification
|
21 |
+
cd $WORKDIR
|
22 |
+
|
23 |
+
python run_glue.py \
|
24 |
+
--model_name_or_path bert-base-uncased \
|
25 |
+
--task_name mnli \
|
26 |
+
--max_seq_length 128 \
|
27 |
+
--do_train \
|
28 |
+
--per_device_train_batch_size 32 \
|
29 |
+
--learning_rate 2e-5 \
|
30 |
+
--num_train_epochs $NEPOCH \
|
31 |
+
--logging_steps 1 \
|
32 |
+
--evaluation_strategy steps \
|
33 |
+
--save_steps 3000 \
|
34 |
+
--do_eval \
|
35 |
+
--per_device_eval_batch_size 128 \
|
36 |
+
--eval_steps 250 \
|
37 |
+
--output_dir $OUTDIR
|
38 |
+
--overwrite_output_dir
|
39 |
+
```
|
40 |
+
|
41 |
+
# Eval
|
42 |
+
```bash
|
43 |
+
export CUDA_VISIBLE_DEVICES=0
|
44 |
+
|
45 |
+
OUTDIR=eval-bert-mnli
|
46 |
+
WORKDIR=transformers/examples/pytorch/text-classification
|
47 |
+
cd $WORKDIR
|
48 |
+
|
49 |
+
nohup python run_glue.py \
|
50 |
+
--model_name_or_path vuiseng9/bert-mnli \
|
51 |
+
--task_name mnli \
|
52 |
+
--do_eval \
|
53 |
+
--per_device_eval_batch_size 128 \
|
54 |
+
--max_seq_length 128 \
|
55 |
+
--overwrite_output_dir \
|
56 |
+
--output_dir $OUTDIR 2>&1 | tee $OUTDIR/run.log &
|
57 |
+
```
|
all_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"eval_accuracy": 0.8494711147274207,
|
3 |
+
"eval_loss": 0.46933791041374207,
|
4 |
+
"eval_runtime": 22.6994,
|
5 |
+
"eval_samples": 9832,
|
6 |
+
"eval_samples_per_second": 433.139,
|
7 |
+
"eval_steps_per_second": 3.392
|
8 |
+
}
|
config.json
ADDED
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "bert-base-uncased",
|
3 |
+
"architectures": [
|
4 |
+
"BertForSequenceClassification"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0.1,
|
7 |
+
"finetuning_task": "mnli",
|
8 |
+
"gradient_checkpointing": false,
|
9 |
+
"hidden_act": "gelu",
|
10 |
+
"hidden_dropout_prob": 0.1,
|
11 |
+
"hidden_size": 768,
|
12 |
+
"id2label": {
|
13 |
+
"0": "LABEL_0",
|
14 |
+
"1": "LABEL_1",
|
15 |
+
"2": "LABEL_2"
|
16 |
+
},
|
17 |
+
"initializer_range": 0.02,
|
18 |
+
"intermediate_size": 3072,
|
19 |
+
"label2id": {
|
20 |
+
"LABEL_0": 0,
|
21 |
+
"LABEL_1": 1,
|
22 |
+
"LABEL_2": 2
|
23 |
+
},
|
24 |
+
"layer_norm_eps": 1e-12,
|
25 |
+
"max_position_embeddings": 512,
|
26 |
+
"model_type": "bert",
|
27 |
+
"num_attention_heads": 12,
|
28 |
+
"num_hidden_layers": 12,
|
29 |
+
"pad_token_id": 0,
|
30 |
+
"position_embedding_type": "absolute",
|
31 |
+
"problem_type": "single_label_classification",
|
32 |
+
"torch_dtype": "float32",
|
33 |
+
"transformers_version": "4.9.1",
|
34 |
+
"type_vocab_size": 2,
|
35 |
+
"use_cache": true,
|
36 |
+
"vocab_size": 30522
|
37 |
+
}
|
eval_mnli-mm_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"eval_accuracy": 0.8494711147274207,
|
3 |
+
"eval_loss": 0.46933791041374207,
|
4 |
+
"eval_runtime": 22.6994,
|
5 |
+
"eval_samples": 9832,
|
6 |
+
"eval_samples_per_second": 433.139,
|
7 |
+
"eval_steps_per_second": 3.392
|
8 |
+
}
|
eval_mnli_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"eval_accuracy": 0.8444218033622007,
|
3 |
+
"eval_loss": 0.49247315526008606,
|
4 |
+
"eval_runtime": 22.5125,
|
5 |
+
"eval_samples": 9815,
|
6 |
+
"eval_samples_per_second": 435.98,
|
7 |
+
"eval_steps_per_second": 3.42
|
8 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 3.0,
|
3 |
+
"eval_accuracy": 0.8494711147274207,
|
4 |
+
"eval_loss": 0.46933791041374207,
|
5 |
+
"eval_runtime": 22.4675,
|
6 |
+
"eval_samples": 9832,
|
7 |
+
"eval_samples_per_second": 437.61,
|
8 |
+
"eval_steps_per_second": 3.427
|
9 |
+
}
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f7ecc6a182477feca62817a4e028dbd3a949fac885da7b457b4517ad086774c9
|
3 |
+
size 438022317
|
special_tokens_map.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]"}
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
tokenizer_config.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"do_lower_case": true, "unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "tokenize_chinese_chars": true, "strip_accents": null, "model_max_length": 512, "special_tokens_map_file": null, "name_or_path": "bert-base-uncased", "tokenizer_class": "BertTokenizer"}
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 3.0,
|
3 |
+
"train_loss": 0.3605867331176782,
|
4 |
+
"train_runtime": 11742.5051,
|
5 |
+
"train_samples": 392702,
|
6 |
+
"train_samples_per_second": 100.328,
|
7 |
+
"train_steps_per_second": 3.135
|
8 |
+
}
|
trainer_state.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2d6f3cbe28aea9c1b0c5b889ec91bec153ace6d29b73b98f5a59d8812b2be826
|
3 |
+
size 2735
|
vocab.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|