José Ángel González
commited on
Commit
•
841bbde
1
Parent(s):
85c8f2c
add model
Browse files- .gitignore +1 -0
- README.md +81 -0
- all_results.json +18 -0
- config.json +73 -0
- eval_results.json +13 -0
- merges.txt +0 -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 +305 -0
- training_args.bin +3 -0
- vocab.json +0 -0
.gitignore
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
checkpoint-*/
|
README.md
ADDED
@@ -0,0 +1,81 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
tags:
|
4 |
+
- generated_from_trainer
|
5 |
+
metrics:
|
6 |
+
- rouge
|
7 |
+
model-index:
|
8 |
+
- name: POCTS
|
9 |
+
results:
|
10 |
+
- task:
|
11 |
+
name: Summarization
|
12 |
+
type: summarization
|
13 |
+
metrics:
|
14 |
+
- name: Rouge1
|
15 |
+
type: rouge
|
16 |
+
value: 23.9947
|
17 |
+
---
|
18 |
+
|
19 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
20 |
+
should probably proofread and complete it, then remove this comment. -->
|
21 |
+
|
22 |
+
# POCTS
|
23 |
+
|
24 |
+
This model is a fine-tuned version of [facebook/bart-large](https://huggingface.co/facebook/bart-large) on an unknown dataset.
|
25 |
+
It achieves the following results on the evaluation set:
|
26 |
+
- Loss: 4.6366
|
27 |
+
- Rouge1: 23.9947
|
28 |
+
- Rouge2: 5.3034
|
29 |
+
- Rougel: 16.3635
|
30 |
+
- Rougelsum: 19.7575
|
31 |
+
- Gen Len: 55.25
|
32 |
+
|
33 |
+
## Model description
|
34 |
+
|
35 |
+
More information needed
|
36 |
+
|
37 |
+
## Intended uses & limitations
|
38 |
+
|
39 |
+
More information needed
|
40 |
+
|
41 |
+
## Training and evaluation data
|
42 |
+
|
43 |
+
More information needed
|
44 |
+
|
45 |
+
## Training procedure
|
46 |
+
|
47 |
+
### Training hyperparameters
|
48 |
+
|
49 |
+
The following hyperparameters were used during training:
|
50 |
+
- learning_rate: 5e-05
|
51 |
+
- train_batch_size: 8
|
52 |
+
- eval_batch_size: 8
|
53 |
+
- seed: 42
|
54 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
55 |
+
- lr_scheduler_type: linear
|
56 |
+
- lr_scheduler_warmup_ratio: 0.15
|
57 |
+
- num_epochs: 10.0
|
58 |
+
- mixed_precision_training: Native AMP
|
59 |
+
|
60 |
+
### Training results
|
61 |
+
|
62 |
+
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|
63 |
+
|:-------------:|:-----:|:----:|:---------------:|:-------:|:------:|:-------:|:---------:|:-------:|
|
64 |
+
| 3.6608 | 1.0 | 125 | 3.3274 | 15.2816 | 4.0723 | 12.0873 | 13.6532 | 19.23 |
|
65 |
+
| 3.3135 | 2.0 | 250 | 3.3133 | 16.5105 | 4.4915 | 13.8639 | 14.8435 | 19.31 |
|
66 |
+
| 2.7732 | 3.0 | 375 | 3.3856 | 16.8687 | 4.904 | 13.4615 | 14.7432 | 19.93 |
|
67 |
+
| 2.2759 | 4.0 | 500 | 3.5508 | 15.4023 | 4.1761 | 12.5635 | 13.7679 | 19.27 |
|
68 |
+
| 1.8199 | 5.0 | 625 | 3.7591 | 17.7686 | 4.532 | 13.9086 | 15.3305 | 19.93 |
|
69 |
+
| 1.4575 | 6.0 | 750 | 3.9726 | 16.4133 | 4.359 | 13.4621 | 14.5896 | 19.92 |
|
70 |
+
| 1.126 | 7.0 | 875 | 4.2964 | 17.3934 | 3.6935 | 13.7934 | 14.9719 | 19.43 |
|
71 |
+
| 0.9073 | 8.0 | 1000 | 4.4205 | 17.4328 | 3.8734 | 13.4282 | 14.7105 | 19.83 |
|
72 |
+
| 0.7925 | 9.0 | 1125 | 4.5501 | 17.3798 | 3.9775 | 13.3141 | 14.7692 | 19.87 |
|
73 |
+
| 0.6844 | 10.0 | 1250 | 4.6366 | 17.3867 | 4.2671 | 13.7924 | 15.1543 | 19.81 |
|
74 |
+
|
75 |
+
|
76 |
+
### Framework versions
|
77 |
+
|
78 |
+
- Transformers 4.10.2
|
79 |
+
- Pytorch 1.7.1+cu110
|
80 |
+
- Datasets 1.11.0
|
81 |
+
- Tokenizers 0.10.3
|
all_results.json
ADDED
@@ -0,0 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 10.0,
|
3 |
+
"eval_gen_len": 55.25,
|
4 |
+
"eval_loss": 4.636623382568359,
|
5 |
+
"eval_rouge1": 23.9947,
|
6 |
+
"eval_rouge2": 5.3034,
|
7 |
+
"eval_rougeL": 16.3635,
|
8 |
+
"eval_rougeLsum": 19.7575,
|
9 |
+
"eval_runtime": 30.2029,
|
10 |
+
"eval_samples": 100,
|
11 |
+
"eval_samples_per_second": 3.311,
|
12 |
+
"eval_steps_per_second": 0.43,
|
13 |
+
"train_loss": 1.877090069580078,
|
14 |
+
"train_runtime": 1007.5303,
|
15 |
+
"train_samples": 995,
|
16 |
+
"train_samples_per_second": 9.876,
|
17 |
+
"train_steps_per_second": 1.241
|
18 |
+
}
|
config.json
ADDED
@@ -0,0 +1,73 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "facebook/bart-large",
|
3 |
+
"activation_dropout": 0.1,
|
4 |
+
"activation_function": "gelu",
|
5 |
+
"add_bias_logits": false,
|
6 |
+
"add_final_layer_norm": false,
|
7 |
+
"architectures": [
|
8 |
+
"BartForConditionalGeneration"
|
9 |
+
],
|
10 |
+
"attention_dropout": 0.1,
|
11 |
+
"bos_token_id": 0,
|
12 |
+
"classif_dropout": 0.1,
|
13 |
+
"classifier_dropout": 0.0,
|
14 |
+
"d_model": 1024,
|
15 |
+
"decoder_attention_heads": 16,
|
16 |
+
"decoder_ffn_dim": 4096,
|
17 |
+
"decoder_layerdrop": 0.0,
|
18 |
+
"decoder_layers": 12,
|
19 |
+
"decoder_start_token_id": 2,
|
20 |
+
"dropout": 0.1,
|
21 |
+
"early_stopping": true,
|
22 |
+
"encoder_attention_heads": 16,
|
23 |
+
"encoder_ffn_dim": 4096,
|
24 |
+
"encoder_layerdrop": 0.0,
|
25 |
+
"encoder_layers": 12,
|
26 |
+
"eos_token_id": 2,
|
27 |
+
"forced_eos_token_id": 2,
|
28 |
+
"gradient_checkpointing": false,
|
29 |
+
"id2label": {
|
30 |
+
"0": "LABEL_0",
|
31 |
+
"1": "LABEL_1",
|
32 |
+
"2": "LABEL_2"
|
33 |
+
},
|
34 |
+
"init_std": 0.02,
|
35 |
+
"is_encoder_decoder": true,
|
36 |
+
"label2id": {
|
37 |
+
"LABEL_0": 0,
|
38 |
+
"LABEL_1": 1,
|
39 |
+
"LABEL_2": 2
|
40 |
+
},
|
41 |
+
"max_position_embeddings": 1024,
|
42 |
+
"model_type": "bart",
|
43 |
+
"no_repeat_ngram_size": 3,
|
44 |
+
"normalize_before": false,
|
45 |
+
"num_beams": 4,
|
46 |
+
"num_hidden_layers": 12,
|
47 |
+
"pad_token_id": 1,
|
48 |
+
"scale_embedding": false,
|
49 |
+
"task_specific_params": {
|
50 |
+
"summarization": {
|
51 |
+
"length_penalty": 1.0,
|
52 |
+
"max_length": 128,
|
53 |
+
"min_length": 12,
|
54 |
+
"num_beams": 4
|
55 |
+
},
|
56 |
+
"summarization_cnn": {
|
57 |
+
"length_penalty": 2.0,
|
58 |
+
"max_length": 142,
|
59 |
+
"min_length": 56,
|
60 |
+
"num_beams": 4
|
61 |
+
},
|
62 |
+
"summarization_xsum": {
|
63 |
+
"length_penalty": 1.0,
|
64 |
+
"max_length": 62,
|
65 |
+
"min_length": 11,
|
66 |
+
"num_beams": 6
|
67 |
+
}
|
68 |
+
},
|
69 |
+
"torch_dtype": "float32",
|
70 |
+
"transformers_version": "4.10.2",
|
71 |
+
"use_cache": true,
|
72 |
+
"vocab_size": 50265
|
73 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 10.0,
|
3 |
+
"eval_gen_len": 55.25,
|
4 |
+
"eval_loss": 4.636623382568359,
|
5 |
+
"eval_rouge1": 23.9947,
|
6 |
+
"eval_rouge2": 5.3034,
|
7 |
+
"eval_rougeL": 16.3635,
|
8 |
+
"eval_rougeLsum": 19.7575,
|
9 |
+
"eval_runtime": 30.2029,
|
10 |
+
"eval_samples": 100,
|
11 |
+
"eval_samples_per_second": 3.311,
|
12 |
+
"eval_steps_per_second": 0.43
|
13 |
+
}
|
merges.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:90eee62817bf936a7493476fc3439c5f8844d7965c7e38dcfe0ac1244ea91248
|
3 |
+
size 1625569391
|
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": false}}
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
tokenizer_config.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"unk_token": "<unk>", "bos_token": "<s>", "eos_token": "</s>", "add_prefix_space": false, "errors": "replace", "sep_token": "</s>", "cls_token": "<s>", "pad_token": "<pad>", "mask_token": "<mask>", "model_max_length": 1024, "special_tokens_map_file": null, "name_or_path": "facebook/bart-large", "tokenizer_class": "BartTokenizer"}
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 10.0,
|
3 |
+
"train_loss": 1.877090069580078,
|
4 |
+
"train_runtime": 1007.5303,
|
5 |
+
"train_samples": 995,
|
6 |
+
"train_samples_per_second": 9.876,
|
7 |
+
"train_steps_per_second": 1.241
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,305 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 10.0,
|
5 |
+
"global_step": 1250,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.4,
|
12 |
+
"learning_rate": 1.223404255319149e-05,
|
13 |
+
"loss": 3.8907,
|
14 |
+
"step": 50
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.8,
|
18 |
+
"learning_rate": 2.5265957446808515e-05,
|
19 |
+
"loss": 3.6608,
|
20 |
+
"step": 100
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 1.0,
|
24 |
+
"eval_gen_len": 19.23,
|
25 |
+
"eval_loss": 3.32743763923645,
|
26 |
+
"eval_rouge1": 15.2816,
|
27 |
+
"eval_rouge2": 4.0723,
|
28 |
+
"eval_rougeL": 12.0873,
|
29 |
+
"eval_rougeLsum": 13.6532,
|
30 |
+
"eval_runtime": 11.1064,
|
31 |
+
"eval_samples_per_second": 9.004,
|
32 |
+
"eval_steps_per_second": 1.17,
|
33 |
+
"step": 125
|
34 |
+
},
|
35 |
+
{
|
36 |
+
"epoch": 1.2,
|
37 |
+
"learning_rate": 3.8563829787234045e-05,
|
38 |
+
"loss": 3.4017,
|
39 |
+
"step": 150
|
40 |
+
},
|
41 |
+
{
|
42 |
+
"epoch": 1.6,
|
43 |
+
"learning_rate": 4.967043314500942e-05,
|
44 |
+
"loss": 3.3923,
|
45 |
+
"step": 200
|
46 |
+
},
|
47 |
+
{
|
48 |
+
"epoch": 2.0,
|
49 |
+
"learning_rate": 4.7316384180790966e-05,
|
50 |
+
"loss": 3.3135,
|
51 |
+
"step": 250
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 2.0,
|
55 |
+
"eval_gen_len": 19.31,
|
56 |
+
"eval_loss": 3.313326358795166,
|
57 |
+
"eval_rouge1": 16.5105,
|
58 |
+
"eval_rouge2": 4.4915,
|
59 |
+
"eval_rougeL": 13.8639,
|
60 |
+
"eval_rougeLsum": 14.8435,
|
61 |
+
"eval_runtime": 10.7765,
|
62 |
+
"eval_samples_per_second": 9.279,
|
63 |
+
"eval_steps_per_second": 1.206,
|
64 |
+
"step": 250
|
65 |
+
},
|
66 |
+
{
|
67 |
+
"epoch": 2.4,
|
68 |
+
"learning_rate": 4.4962335216572505e-05,
|
69 |
+
"loss": 2.7445,
|
70 |
+
"step": 300
|
71 |
+
},
|
72 |
+
{
|
73 |
+
"epoch": 2.8,
|
74 |
+
"learning_rate": 4.260828625235405e-05,
|
75 |
+
"loss": 2.7732,
|
76 |
+
"step": 350
|
77 |
+
},
|
78 |
+
{
|
79 |
+
"epoch": 3.0,
|
80 |
+
"eval_gen_len": 19.93,
|
81 |
+
"eval_loss": 3.3855550289154053,
|
82 |
+
"eval_rouge1": 16.8687,
|
83 |
+
"eval_rouge2": 4.904,
|
84 |
+
"eval_rougeL": 13.4615,
|
85 |
+
"eval_rougeLsum": 14.7432,
|
86 |
+
"eval_runtime": 10.8555,
|
87 |
+
"eval_samples_per_second": 9.212,
|
88 |
+
"eval_steps_per_second": 1.198,
|
89 |
+
"step": 375
|
90 |
+
},
|
91 |
+
{
|
92 |
+
"epoch": 3.2,
|
93 |
+
"learning_rate": 4.025423728813559e-05,
|
94 |
+
"loss": 2.5055,
|
95 |
+
"step": 400
|
96 |
+
},
|
97 |
+
{
|
98 |
+
"epoch": 3.6,
|
99 |
+
"learning_rate": 3.790018832391714e-05,
|
100 |
+
"loss": 2.1993,
|
101 |
+
"step": 450
|
102 |
+
},
|
103 |
+
{
|
104 |
+
"epoch": 4.0,
|
105 |
+
"learning_rate": 3.554613935969868e-05,
|
106 |
+
"loss": 2.2759,
|
107 |
+
"step": 500
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 4.0,
|
111 |
+
"eval_gen_len": 19.27,
|
112 |
+
"eval_loss": 3.5507638454437256,
|
113 |
+
"eval_rouge1": 15.4023,
|
114 |
+
"eval_rouge2": 4.1761,
|
115 |
+
"eval_rougeL": 12.5635,
|
116 |
+
"eval_rougeLsum": 13.7679,
|
117 |
+
"eval_runtime": 10.8402,
|
118 |
+
"eval_samples_per_second": 9.225,
|
119 |
+
"eval_steps_per_second": 1.199,
|
120 |
+
"step": 500
|
121 |
+
},
|
122 |
+
{
|
123 |
+
"epoch": 4.4,
|
124 |
+
"learning_rate": 3.319209039548023e-05,
|
125 |
+
"loss": 1.7219,
|
126 |
+
"step": 550
|
127 |
+
},
|
128 |
+
{
|
129 |
+
"epoch": 4.8,
|
130 |
+
"learning_rate": 3.0838041431261774e-05,
|
131 |
+
"loss": 1.8199,
|
132 |
+
"step": 600
|
133 |
+
},
|
134 |
+
{
|
135 |
+
"epoch": 5.0,
|
136 |
+
"eval_gen_len": 19.93,
|
137 |
+
"eval_loss": 3.7590889930725098,
|
138 |
+
"eval_rouge1": 17.7686,
|
139 |
+
"eval_rouge2": 4.532,
|
140 |
+
"eval_rougeL": 13.9086,
|
141 |
+
"eval_rougeLsum": 15.3305,
|
142 |
+
"eval_runtime": 10.8632,
|
143 |
+
"eval_samples_per_second": 9.205,
|
144 |
+
"eval_steps_per_second": 1.197,
|
145 |
+
"step": 625
|
146 |
+
},
|
147 |
+
{
|
148 |
+
"epoch": 5.2,
|
149 |
+
"learning_rate": 2.8483992467043313e-05,
|
150 |
+
"loss": 1.5748,
|
151 |
+
"step": 650
|
152 |
+
},
|
153 |
+
{
|
154 |
+
"epoch": 5.6,
|
155 |
+
"learning_rate": 2.612994350282486e-05,
|
156 |
+
"loss": 1.3965,
|
157 |
+
"step": 700
|
158 |
+
},
|
159 |
+
{
|
160 |
+
"epoch": 6.0,
|
161 |
+
"learning_rate": 2.3775894538606405e-05,
|
162 |
+
"loss": 1.4575,
|
163 |
+
"step": 750
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 6.0,
|
167 |
+
"eval_gen_len": 19.92,
|
168 |
+
"eval_loss": 3.9725918769836426,
|
169 |
+
"eval_rouge1": 16.4133,
|
170 |
+
"eval_rouge2": 4.359,
|
171 |
+
"eval_rougeL": 13.4621,
|
172 |
+
"eval_rougeLsum": 14.5896,
|
173 |
+
"eval_runtime": 10.8642,
|
174 |
+
"eval_samples_per_second": 9.205,
|
175 |
+
"eval_steps_per_second": 1.197,
|
176 |
+
"step": 750
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 6.4,
|
180 |
+
"learning_rate": 2.1421845574387948e-05,
|
181 |
+
"loss": 1.1037,
|
182 |
+
"step": 800
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 6.8,
|
186 |
+
"learning_rate": 1.906779661016949e-05,
|
187 |
+
"loss": 1.126,
|
188 |
+
"step": 850
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 7.0,
|
192 |
+
"eval_gen_len": 19.43,
|
193 |
+
"eval_loss": 4.29640007019043,
|
194 |
+
"eval_rouge1": 17.3934,
|
195 |
+
"eval_rouge2": 3.6935,
|
196 |
+
"eval_rougeL": 13.7934,
|
197 |
+
"eval_rougeLsum": 14.9719,
|
198 |
+
"eval_runtime": 10.8699,
|
199 |
+
"eval_samples_per_second": 9.2,
|
200 |
+
"eval_steps_per_second": 1.196,
|
201 |
+
"step": 875
|
202 |
+
},
|
203 |
+
{
|
204 |
+
"epoch": 7.2,
|
205 |
+
"learning_rate": 1.6713747645951036e-05,
|
206 |
+
"loss": 1.0553,
|
207 |
+
"step": 900
|
208 |
+
},
|
209 |
+
{
|
210 |
+
"epoch": 7.6,
|
211 |
+
"learning_rate": 1.435969868173258e-05,
|
212 |
+
"loss": 0.9473,
|
213 |
+
"step": 950
|
214 |
+
},
|
215 |
+
{
|
216 |
+
"epoch": 8.0,
|
217 |
+
"learning_rate": 1.2005649717514125e-05,
|
218 |
+
"loss": 0.9073,
|
219 |
+
"step": 1000
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 8.0,
|
223 |
+
"eval_gen_len": 19.83,
|
224 |
+
"eval_loss": 4.420531749725342,
|
225 |
+
"eval_rouge1": 17.4328,
|
226 |
+
"eval_rouge2": 3.8734,
|
227 |
+
"eval_rougeL": 13.4282,
|
228 |
+
"eval_rougeLsum": 14.7105,
|
229 |
+
"eval_runtime": 10.9454,
|
230 |
+
"eval_samples_per_second": 9.136,
|
231 |
+
"eval_steps_per_second": 1.188,
|
232 |
+
"step": 1000
|
233 |
+
},
|
234 |
+
{
|
235 |
+
"epoch": 8.4,
|
236 |
+
"learning_rate": 9.65160075329567e-06,
|
237 |
+
"loss": 0.7636,
|
238 |
+
"step": 1050
|
239 |
+
},
|
240 |
+
{
|
241 |
+
"epoch": 8.8,
|
242 |
+
"learning_rate": 7.297551789077213e-06,
|
243 |
+
"loss": 0.7925,
|
244 |
+
"step": 1100
|
245 |
+
},
|
246 |
+
{
|
247 |
+
"epoch": 9.0,
|
248 |
+
"eval_gen_len": 19.87,
|
249 |
+
"eval_loss": 4.550107955932617,
|
250 |
+
"eval_rouge1": 17.3798,
|
251 |
+
"eval_rouge2": 3.9775,
|
252 |
+
"eval_rougeL": 13.3141,
|
253 |
+
"eval_rougeLsum": 14.7692,
|
254 |
+
"eval_runtime": 10.8417,
|
255 |
+
"eval_samples_per_second": 9.224,
|
256 |
+
"eval_steps_per_second": 1.199,
|
257 |
+
"step": 1125
|
258 |
+
},
|
259 |
+
{
|
260 |
+
"epoch": 9.2,
|
261 |
+
"learning_rate": 4.990583804143127e-06,
|
262 |
+
"loss": 0.7396,
|
263 |
+
"step": 1150
|
264 |
+
},
|
265 |
+
{
|
266 |
+
"epoch": 9.6,
|
267 |
+
"learning_rate": 2.6365348399246707e-06,
|
268 |
+
"loss": 0.6792,
|
269 |
+
"step": 1200
|
270 |
+
},
|
271 |
+
{
|
272 |
+
"epoch": 10.0,
|
273 |
+
"learning_rate": 2.8248587570621473e-07,
|
274 |
+
"loss": 0.6844,
|
275 |
+
"step": 1250
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 10.0,
|
279 |
+
"eval_gen_len": 19.81,
|
280 |
+
"eval_loss": 4.636623382568359,
|
281 |
+
"eval_rouge1": 17.3867,
|
282 |
+
"eval_rouge2": 4.2671,
|
283 |
+
"eval_rougeL": 13.7924,
|
284 |
+
"eval_rougeLsum": 15.1543,
|
285 |
+
"eval_runtime": 10.8893,
|
286 |
+
"eval_samples_per_second": 9.183,
|
287 |
+
"eval_steps_per_second": 1.194,
|
288 |
+
"step": 1250
|
289 |
+
},
|
290 |
+
{
|
291 |
+
"epoch": 10.0,
|
292 |
+
"step": 1250,
|
293 |
+
"total_flos": 2.278181410848768e+16,
|
294 |
+
"train_loss": 1.877090069580078,
|
295 |
+
"train_runtime": 1007.5303,
|
296 |
+
"train_samples_per_second": 9.876,
|
297 |
+
"train_steps_per_second": 1.241
|
298 |
+
}
|
299 |
+
],
|
300 |
+
"max_steps": 1250,
|
301 |
+
"num_train_epochs": 10,
|
302 |
+
"total_flos": 2.278181410848768e+16,
|
303 |
+
"trial_name": null,
|
304 |
+
"trial_params": null
|
305 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:70f5cbd5c3fdf1dff1e73912d6b5f5e3bf6950bf344cc4feff7de3eec81290ac
|
3 |
+
size 2799
|
vocab.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|