End of training
Browse files- README.md +8 -6
- all_results.json +29 -0
- eval_results.json +13 -0
- generated_predictions.txt +0 -0
- predict_results.json +12 -0
- train_results.json +9 -0
- trainer_state.json +130 -0
README.md
CHANGED
@@ -1,4 +1,6 @@
|
|
1 |
---
|
|
|
|
|
2 |
license: apache-2.0
|
3 |
base_model: LazarusNLP/IndoNanoT5-base
|
4 |
tags:
|
@@ -17,12 +19,12 @@ should probably proofread and complete it, then remove this comment. -->
|
|
17 |
|
18 |
This model is a fine-tuned version of [LazarusNLP/IndoNanoT5-base](https://huggingface.co/LazarusNLP/IndoNanoT5-base) on an unknown dataset.
|
19 |
It achieves the following results on the evaluation set:
|
20 |
-
- Loss: 0.
|
21 |
-
- Rouge1: 72.
|
22 |
-
- Rouge2:
|
23 |
-
- Rougel:
|
24 |
-
- Rougelsum: 71.
|
25 |
-
- Gen Len:
|
26 |
|
27 |
## Model description
|
28 |
|
|
|
1 |
---
|
2 |
+
language:
|
3 |
+
- id
|
4 |
license: apache-2.0
|
5 |
base_model: LazarusNLP/IndoNanoT5-base
|
6 |
tags:
|
|
|
19 |
|
20 |
This model is a fine-tuned version of [LazarusNLP/IndoNanoT5-base](https://huggingface.co/LazarusNLP/IndoNanoT5-base) on an unknown dataset.
|
21 |
It achieves the following results on the evaluation set:
|
22 |
+
- Loss: 0.5442
|
23 |
+
- Rouge1: 72.6224
|
24 |
+
- Rouge2: 65.7072
|
25 |
+
- Rougel: 69.4293
|
26 |
+
- Rougelsum: 71.6325
|
27 |
+
- Gen Len: 101.7893
|
28 |
|
29 |
## Model description
|
30 |
|
all_results.json
ADDED
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 5.0,
|
3 |
+
"eval_gen_len": 101.78933333333333,
|
4 |
+
"eval_loss": 0.5441932678222656,
|
5 |
+
"eval_rouge1": 72.6224,
|
6 |
+
"eval_rouge2": 65.7072,
|
7 |
+
"eval_rougeL": 69.4293,
|
8 |
+
"eval_rougeLsum": 71.6325,
|
9 |
+
"eval_runtime": 988.0304,
|
10 |
+
"eval_samples": 750,
|
11 |
+
"eval_samples_per_second": 0.759,
|
12 |
+
"eval_steps_per_second": 0.024,
|
13 |
+
"predict_gen_len": 101.4952025586354,
|
14 |
+
"predict_loss": 0.5291196703910828,
|
15 |
+
"predict_rouge1": 72.796,
|
16 |
+
"predict_rouge2": 66.0161,
|
17 |
+
"predict_rougeL": 69.9142,
|
18 |
+
"predict_rougeLsum": 71.9662,
|
19 |
+
"predict_runtime": 4936.0723,
|
20 |
+
"predict_samples": 3752,
|
21 |
+
"predict_samples_per_second": 0.76,
|
22 |
+
"predict_steps_per_second": 0.024,
|
23 |
+
"total_flos": 4.90582346563584e+16,
|
24 |
+
"train_loss": 0.6255005293362879,
|
25 |
+
"train_runtime": 4838.9085,
|
26 |
+
"train_samples": 14272,
|
27 |
+
"train_samples_per_second": 14.747,
|
28 |
+
"train_steps_per_second": 0.922
|
29 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 5.0,
|
3 |
+
"eval_gen_len": 101.78933333333333,
|
4 |
+
"eval_loss": 0.5441932678222656,
|
5 |
+
"eval_rouge1": 72.6224,
|
6 |
+
"eval_rouge2": 65.7072,
|
7 |
+
"eval_rougeL": 69.4293,
|
8 |
+
"eval_rougeLsum": 71.6325,
|
9 |
+
"eval_runtime": 988.0304,
|
10 |
+
"eval_samples": 750,
|
11 |
+
"eval_samples_per_second": 0.759,
|
12 |
+
"eval_steps_per_second": 0.024
|
13 |
+
}
|
generated_predictions.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
predict_results.json
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"predict_gen_len": 101.4952025586354,
|
3 |
+
"predict_loss": 0.5291196703910828,
|
4 |
+
"predict_rouge1": 72.796,
|
5 |
+
"predict_rouge2": 66.0161,
|
6 |
+
"predict_rougeL": 69.9142,
|
7 |
+
"predict_rougeLsum": 71.9662,
|
8 |
+
"predict_runtime": 4936.0723,
|
9 |
+
"predict_samples": 3752,
|
10 |
+
"predict_samples_per_second": 0.76,
|
11 |
+
"predict_steps_per_second": 0.024
|
12 |
+
}
|
train_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 5.0,
|
3 |
+
"total_flos": 4.90582346563584e+16,
|
4 |
+
"train_loss": 0.6255005293362879,
|
5 |
+
"train_runtime": 4838.9085,
|
6 |
+
"train_samples": 14272,
|
7 |
+
"train_samples_per_second": 14.747,
|
8 |
+
"train_steps_per_second": 0.922
|
9 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,130 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 72.3075,
|
3 |
+
"best_model_checkpoint": "bin/indosum-lora-3/checkpoint-2676",
|
4 |
+
"epoch": 5.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 4460,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 1.0,
|
13 |
+
"grad_norm": 0.5876893401145935,
|
14 |
+
"learning_rate": 0.0008,
|
15 |
+
"loss": 0.8274,
|
16 |
+
"step": 892
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 1.0,
|
20 |
+
"eval_gen_len": 100.152,
|
21 |
+
"eval_loss": 0.5684335827827454,
|
22 |
+
"eval_rouge1": 69.5991,
|
23 |
+
"eval_rouge2": 62.0811,
|
24 |
+
"eval_rougeL": 66.2103,
|
25 |
+
"eval_rougeLsum": 68.5179,
|
26 |
+
"eval_runtime": 219.0061,
|
27 |
+
"eval_samples_per_second": 3.425,
|
28 |
+
"eval_steps_per_second": 0.11,
|
29 |
+
"step": 892
|
30 |
+
},
|
31 |
+
{
|
32 |
+
"epoch": 2.0,
|
33 |
+
"grad_norm": 0.4317479431629181,
|
34 |
+
"learning_rate": 0.0006,
|
35 |
+
"loss": 0.6237,
|
36 |
+
"step": 1784
|
37 |
+
},
|
38 |
+
{
|
39 |
+
"epoch": 2.0,
|
40 |
+
"eval_gen_len": 104.78133333333334,
|
41 |
+
"eval_loss": 0.5486577153205872,
|
42 |
+
"eval_rouge1": 71.3379,
|
43 |
+
"eval_rouge2": 64.0461,
|
44 |
+
"eval_rougeL": 68.1399,
|
45 |
+
"eval_rougeLsum": 70.3424,
|
46 |
+
"eval_runtime": 219.5178,
|
47 |
+
"eval_samples_per_second": 3.417,
|
48 |
+
"eval_steps_per_second": 0.109,
|
49 |
+
"step": 1784
|
50 |
+
},
|
51 |
+
{
|
52 |
+
"epoch": 3.0,
|
53 |
+
"grad_norm": 0.5349451303482056,
|
54 |
+
"learning_rate": 0.0004,
|
55 |
+
"loss": 0.5824,
|
56 |
+
"step": 2676
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 3.0,
|
60 |
+
"eval_gen_len": 101.356,
|
61 |
+
"eval_loss": 0.5441932678222656,
|
62 |
+
"eval_rouge1": 72.3075,
|
63 |
+
"eval_rouge2": 65.3703,
|
64 |
+
"eval_rougeL": 69.1638,
|
65 |
+
"eval_rougeLsum": 71.3776,
|
66 |
+
"eval_runtime": 219.0561,
|
67 |
+
"eval_samples_per_second": 3.424,
|
68 |
+
"eval_steps_per_second": 0.11,
|
69 |
+
"step": 2676
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 4.0,
|
73 |
+
"grad_norm": 0.44907253980636597,
|
74 |
+
"learning_rate": 0.0002,
|
75 |
+
"loss": 0.555,
|
76 |
+
"step": 3568
|
77 |
+
},
|
78 |
+
{
|
79 |
+
"epoch": 4.0,
|
80 |
+
"eval_gen_len": 103.64666666666666,
|
81 |
+
"eval_loss": 0.5253363251686096,
|
82 |
+
"eval_rouge1": 72.2139,
|
83 |
+
"eval_rouge2": 65.2916,
|
84 |
+
"eval_rougeL": 69.2278,
|
85 |
+
"eval_rougeLsum": 71.2919,
|
86 |
+
"eval_runtime": 219.9659,
|
87 |
+
"eval_samples_per_second": 3.41,
|
88 |
+
"eval_steps_per_second": 0.109,
|
89 |
+
"step": 3568
|
90 |
+
},
|
91 |
+
{
|
92 |
+
"epoch": 5.0,
|
93 |
+
"grad_norm": 0.3915654420852661,
|
94 |
+
"learning_rate": 0.0,
|
95 |
+
"loss": 0.5391,
|
96 |
+
"step": 4460
|
97 |
+
},
|
98 |
+
{
|
99 |
+
"epoch": 5.0,
|
100 |
+
"eval_gen_len": 102.11733333333333,
|
101 |
+
"eval_loss": 0.5204364657402039,
|
102 |
+
"eval_rouge1": 72.0194,
|
103 |
+
"eval_rouge2": 64.9755,
|
104 |
+
"eval_rougeL": 68.9244,
|
105 |
+
"eval_rougeLsum": 71.0549,
|
106 |
+
"eval_runtime": 218.95,
|
107 |
+
"eval_samples_per_second": 3.425,
|
108 |
+
"eval_steps_per_second": 0.11,
|
109 |
+
"step": 4460
|
110 |
+
},
|
111 |
+
{
|
112 |
+
"epoch": 5.0,
|
113 |
+
"step": 4460,
|
114 |
+
"total_flos": 4.90582346563584e+16,
|
115 |
+
"train_loss": 0.6255005293362879,
|
116 |
+
"train_runtime": 4838.9085,
|
117 |
+
"train_samples_per_second": 14.747,
|
118 |
+
"train_steps_per_second": 0.922
|
119 |
+
}
|
120 |
+
],
|
121 |
+
"logging_steps": 500,
|
122 |
+
"max_steps": 4460,
|
123 |
+
"num_input_tokens_seen": 0,
|
124 |
+
"num_train_epochs": 5,
|
125 |
+
"save_steps": 500,
|
126 |
+
"total_flos": 4.90582346563584e+16,
|
127 |
+
"train_batch_size": 16,
|
128 |
+
"trial_name": null,
|
129 |
+
"trial_params": null
|
130 |
+
}
|