End of training
Browse files- all_results.json +15 -0
- eval_results.json +9 -0
- runs/May09_11-10-30_ml-machine/events.out.tfevents.1715259169.ml-machine.2004.1 +3 -0
- train_results.json +9 -0
- trainer_state.json +155 -0
all_results.json
ADDED
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 5.0,
|
3 |
+
"eval_exact_match": 61.125,
|
4 |
+
"eval_f1": 75.55577348115898,
|
5 |
+
"eval_runtime": 87.5964,
|
6 |
+
"eval_samples": 3858,
|
7 |
+
"eval_samples_per_second": 44.043,
|
8 |
+
"eval_steps_per_second": 1.575,
|
9 |
+
"total_flos": 6.354793682009856e+16,
|
10 |
+
"train_loss": 1.2508642091341544,
|
11 |
+
"train_runtime": 5889.9617,
|
12 |
+
"train_samples": 18247,
|
13 |
+
"train_samples_per_second": 15.49,
|
14 |
+
"train_steps_per_second": 0.553
|
15 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 5.0,
|
3 |
+
"eval_exact_match": 61.125,
|
4 |
+
"eval_f1": 75.55577348115898,
|
5 |
+
"eval_runtime": 87.5964,
|
6 |
+
"eval_samples": 3858,
|
7 |
+
"eval_samples_per_second": 44.043,
|
8 |
+
"eval_steps_per_second": 1.575
|
9 |
+
}
|
runs/May09_11-10-30_ml-machine/events.out.tfevents.1715259169.ml-machine.2004.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3314bba6a3beb8cc4ac8ab0473b10db1f2b1e1d2a03f0305751f6bd1a7afba57
|
3 |
+
size 412
|
train_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 5.0,
|
3 |
+
"total_flos": 6.354793682009856e+16,
|
4 |
+
"train_loss": 1.2508642091341544,
|
5 |
+
"train_runtime": 5889.9617,
|
6 |
+
"train_samples": 18247,
|
7 |
+
"train_samples_per_second": 15.49,
|
8 |
+
"train_steps_per_second": 0.553
|
9 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,155 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 75.55577348115898,
|
3 |
+
"best_model_checkpoint": "/root/turkic_qa/en_kaz_models/en_kaz_xlm_roberta_large_model/checkpoint-3260",
|
4 |
+
"epoch": 5.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 3260,
|
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 |
+
"step": 652,
|
14 |
+
"train_exact_match": 55.94405594405595,
|
15 |
+
"train_f1": 73.63671036317213,
|
16 |
+
"train_runtime": 28.1019,
|
17 |
+
"train_samples_per_second": 43.663,
|
18 |
+
"train_steps_per_second": 1.566
|
19 |
+
},
|
20 |
+
{
|
21 |
+
"epoch": 1.0,
|
22 |
+
"grad_norm": 41.142127990722656,
|
23 |
+
"learning_rate": 1e-05,
|
24 |
+
"loss": 3.2057,
|
25 |
+
"step": 652
|
26 |
+
},
|
27 |
+
{
|
28 |
+
"epoch": 1.0,
|
29 |
+
"eval_exact_match": 51.65625,
|
30 |
+
"eval_f1": 68.43543312331303,
|
31 |
+
"eval_runtime": 88.3369,
|
32 |
+
"eval_samples_per_second": 43.674,
|
33 |
+
"eval_steps_per_second": 1.562,
|
34 |
+
"step": 652
|
35 |
+
},
|
36 |
+
{
|
37 |
+
"epoch": 2.0,
|
38 |
+
"step": 1304,
|
39 |
+
"train_exact_match": 67.53246753246754,
|
40 |
+
"train_f1": 82.24342151350066,
|
41 |
+
"train_runtime": 28.5968,
|
42 |
+
"train_samples_per_second": 43.396,
|
43 |
+
"train_steps_per_second": 1.574
|
44 |
+
},
|
45 |
+
{
|
46 |
+
"epoch": 2.0,
|
47 |
+
"grad_norm": 36.88103103637695,
|
48 |
+
"learning_rate": 7.500000000000001e-06,
|
49 |
+
"loss": 1.172,
|
50 |
+
"step": 1304
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 2.0,
|
54 |
+
"eval_exact_match": 56.875,
|
55 |
+
"eval_f1": 73.03642440401033,
|
56 |
+
"eval_runtime": 88.3326,
|
57 |
+
"eval_samples_per_second": 43.676,
|
58 |
+
"eval_steps_per_second": 1.562,
|
59 |
+
"step": 1304
|
60 |
+
},
|
61 |
+
{
|
62 |
+
"epoch": 3.0,
|
63 |
+
"step": 1956,
|
64 |
+
"train_exact_match": 78.92107892107892,
|
65 |
+
"train_f1": 90.2235322381411,
|
66 |
+
"train_runtime": 27.7189,
|
67 |
+
"train_samples_per_second": 43.508,
|
68 |
+
"train_steps_per_second": 1.587
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 3.0,
|
72 |
+
"grad_norm": 38.58055877685547,
|
73 |
+
"learning_rate": 5e-06,
|
74 |
+
"loss": 0.8237,
|
75 |
+
"step": 1956
|
76 |
+
},
|
77 |
+
{
|
78 |
+
"epoch": 3.0,
|
79 |
+
"eval_exact_match": 60.5625,
|
80 |
+
"eval_f1": 75.47199499569918,
|
81 |
+
"eval_runtime": 88.109,
|
82 |
+
"eval_samples_per_second": 43.787,
|
83 |
+
"eval_steps_per_second": 1.566,
|
84 |
+
"step": 1956
|
85 |
+
},
|
86 |
+
{
|
87 |
+
"epoch": 4.0,
|
88 |
+
"step": 2608,
|
89 |
+
"train_exact_match": 82.11788211788212,
|
90 |
+
"train_f1": 92.67663132485467,
|
91 |
+
"train_runtime": 29.1938,
|
92 |
+
"train_samples_per_second": 43.16,
|
93 |
+
"train_steps_per_second": 1.541
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 4.0,
|
97 |
+
"grad_norm": 34.8208122253418,
|
98 |
+
"learning_rate": 2.5e-06,
|
99 |
+
"loss": 0.588,
|
100 |
+
"step": 2608
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 4.0,
|
104 |
+
"eval_exact_match": 60.65625,
|
105 |
+
"eval_f1": 75.1934488142968,
|
106 |
+
"eval_runtime": 89.1778,
|
107 |
+
"eval_samples_per_second": 43.262,
|
108 |
+
"eval_steps_per_second": 1.547,
|
109 |
+
"step": 2608
|
110 |
+
},
|
111 |
+
{
|
112 |
+
"epoch": 5.0,
|
113 |
+
"step": 3260,
|
114 |
+
"train_exact_match": 85.61438561438561,
|
115 |
+
"train_f1": 94.19636283438837,
|
116 |
+
"train_runtime": 27.8587,
|
117 |
+
"train_samples_per_second": 43.433,
|
118 |
+
"train_steps_per_second": 1.579
|
119 |
+
},
|
120 |
+
{
|
121 |
+
"epoch": 5.0,
|
122 |
+
"grad_norm": 31.479175567626953,
|
123 |
+
"learning_rate": 0.0,
|
124 |
+
"loss": 0.465,
|
125 |
+
"step": 3260
|
126 |
+
},
|
127 |
+
{
|
128 |
+
"epoch": 5.0,
|
129 |
+
"eval_exact_match": 61.125,
|
130 |
+
"eval_f1": 75.55577348115898,
|
131 |
+
"eval_runtime": 88.1654,
|
132 |
+
"eval_samples_per_second": 43.759,
|
133 |
+
"eval_steps_per_second": 1.565,
|
134 |
+
"step": 3260
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 5.0,
|
138 |
+
"step": 3260,
|
139 |
+
"total_flos": 6.354793682009856e+16,
|
140 |
+
"train_loss": 1.2508642091341544,
|
141 |
+
"train_runtime": 5889.9617,
|
142 |
+
"train_samples_per_second": 15.49,
|
143 |
+
"train_steps_per_second": 0.553
|
144 |
+
}
|
145 |
+
],
|
146 |
+
"logging_steps": 500,
|
147 |
+
"max_steps": 3260,
|
148 |
+
"num_input_tokens_seen": 0,
|
149 |
+
"num_train_epochs": 5,
|
150 |
+
"save_steps": 500,
|
151 |
+
"total_flos": 6.354793682009856e+16,
|
152 |
+
"train_batch_size": 28,
|
153 |
+
"trial_name": null,
|
154 |
+
"trial_params": null
|
155 |
+
}
|