DrishtiSharma
commited on
Commit
•
a0435a3
1
Parent(s):
1dadbca
End of training
Browse files- all_results.json +17 -0
- eval_results.json +17 -0
- trainer_state.json +154 -0
all_results.json
ADDED
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 5.0,
|
3 |
+
"eval_Macro F1": 0.9831697750089234,
|
4 |
+
"eval_Macro Precision": 0.9845093632958801,
|
5 |
+
"eval_Macro Recall": 0.9820926966292135,
|
6 |
+
"eval_Micro F1": 0.9837133550488599,
|
7 |
+
"eval_Micro Precision": 0.9837133550488599,
|
8 |
+
"eval_Micro Recall": 0.9837133550488599,
|
9 |
+
"eval_Weighted F1": 0.983695143647141,
|
10 |
+
"eval_Weighted Precision": 0.9839478827361564,
|
11 |
+
"eval_Weighted Recall": 0.9837133550488599,
|
12 |
+
"eval_accuracy": 0.9837133550488599,
|
13 |
+
"eval_loss": 0.029342563822865486,
|
14 |
+
"eval_runtime": 1.7305,
|
15 |
+
"eval_samples_per_second": 177.408,
|
16 |
+
"eval_steps_per_second": 11.558
|
17 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 5.0,
|
3 |
+
"eval_Macro F1": 0.9831697750089234,
|
4 |
+
"eval_Macro Precision": 0.9845093632958801,
|
5 |
+
"eval_Macro Recall": 0.9820926966292135,
|
6 |
+
"eval_Micro F1": 0.9837133550488599,
|
7 |
+
"eval_Micro Precision": 0.9837133550488599,
|
8 |
+
"eval_Micro Recall": 0.9837133550488599,
|
9 |
+
"eval_Weighted F1": 0.983695143647141,
|
10 |
+
"eval_Weighted Precision": 0.9839478827361564,
|
11 |
+
"eval_Weighted Recall": 0.9837133550488599,
|
12 |
+
"eval_accuracy": 0.9837133550488599,
|
13 |
+
"eval_loss": 0.029342563822865486,
|
14 |
+
"eval_runtime": 1.7305,
|
15 |
+
"eval_samples_per_second": 177.408,
|
16 |
+
"eval_steps_per_second": 11.558
|
17 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,154 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.029342563822865486,
|
3 |
+
"best_model_checkpoint": "roberta-large-hate-offensive-normal-speech-lr-2e-05/checkpoint-765",
|
4 |
+
"epoch": 5.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 765,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.01,
|
13 |
+
"learning_rate": 1.997385620915033e-05,
|
14 |
+
"loss": 1.1222,
|
15 |
+
"step": 1
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"epoch": 1.0,
|
19 |
+
"learning_rate": 1.6000000000000003e-05,
|
20 |
+
"loss": 0.5253,
|
21 |
+
"step": 153
|
22 |
+
},
|
23 |
+
{
|
24 |
+
"epoch": 1.0,
|
25 |
+
"eval_Macro F1": 0.9632836299502966,
|
26 |
+
"eval_Macro Precision": 0.9633333333333334,
|
27 |
+
"eval_Macro Recall": 0.9662454462659381,
|
28 |
+
"eval_Micro F1": 0.9641693811074918,
|
29 |
+
"eval_Micro Precision": 0.9641693811074918,
|
30 |
+
"eval_Micro Recall": 0.9641693811074918,
|
31 |
+
"eval_Weighted F1": 0.9647479825674883,
|
32 |
+
"eval_Weighted Precision": 0.9681107491856679,
|
33 |
+
"eval_Weighted Recall": 0.9641693811074918,
|
34 |
+
"eval_accuracy": 0.9641693811074918,
|
35 |
+
"eval_loss": 0.12700782716274261,
|
36 |
+
"eval_runtime": 1.6798,
|
37 |
+
"eval_samples_per_second": 182.759,
|
38 |
+
"eval_steps_per_second": 11.906,
|
39 |
+
"step": 153
|
40 |
+
},
|
41 |
+
{
|
42 |
+
"epoch": 2.0,
|
43 |
+
"learning_rate": 1.2e-05,
|
44 |
+
"loss": 0.0921,
|
45 |
+
"step": 306
|
46 |
+
},
|
47 |
+
{
|
48 |
+
"epoch": 2.0,
|
49 |
+
"eval_Macro F1": 0.980327393886716,
|
50 |
+
"eval_Macro Precision": 0.9817575757575758,
|
51 |
+
"eval_Macro Recall": 0.9790873600622173,
|
52 |
+
"eval_Micro F1": 0.9804560260586319,
|
53 |
+
"eval_Micro Precision": 0.9804560260586319,
|
54 |
+
"eval_Micro Recall": 0.9804560260586319,
|
55 |
+
"eval_Weighted F1": 0.9804746376641527,
|
56 |
+
"eval_Weighted Precision": 0.9806946994373705,
|
57 |
+
"eval_Weighted Recall": 0.9804560260586319,
|
58 |
+
"eval_accuracy": 0.9804560260586319,
|
59 |
+
"eval_loss": 0.08777838945388794,
|
60 |
+
"eval_runtime": 1.6847,
|
61 |
+
"eval_samples_per_second": 182.225,
|
62 |
+
"eval_steps_per_second": 11.871,
|
63 |
+
"step": 306
|
64 |
+
},
|
65 |
+
{
|
66 |
+
"epoch": 3.0,
|
67 |
+
"learning_rate": 8.000000000000001e-06,
|
68 |
+
"loss": 0.0413,
|
69 |
+
"step": 459
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 3.0,
|
73 |
+
"eval_Macro F1": 0.9860112452705044,
|
74 |
+
"eval_Macro Precision": 0.985663082437276,
|
75 |
+
"eval_Macro Recall": 0.9868510928961749,
|
76 |
+
"eval_Micro F1": 0.9869706840390879,
|
77 |
+
"eval_Micro Precision": 0.9869706840390879,
|
78 |
+
"eval_Micro Recall": 0.9869706840390879,
|
79 |
+
"eval_Weighted F1": 0.9870296055601882,
|
80 |
+
"eval_Weighted Precision": 0.9875310847255788,
|
81 |
+
"eval_Weighted Recall": 0.9869706840390879,
|
82 |
+
"eval_accuracy": 0.9869706840390879,
|
83 |
+
"eval_loss": 0.05901196971535683,
|
84 |
+
"eval_runtime": 1.7013,
|
85 |
+
"eval_samples_per_second": 180.445,
|
86 |
+
"eval_steps_per_second": 11.755,
|
87 |
+
"step": 459
|
88 |
+
},
|
89 |
+
{
|
90 |
+
"epoch": 4.0,
|
91 |
+
"learning_rate": 4.000000000000001e-06,
|
92 |
+
"loss": 0.0261,
|
93 |
+
"step": 612
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 4.0,
|
97 |
+
"eval_Macro F1": 0.989644746787604,
|
98 |
+
"eval_Macro Precision": 0.9899639655737217,
|
99 |
+
"eval_Macro Recall": 0.9895833333333334,
|
100 |
+
"eval_Micro F1": 0.990228013029316,
|
101 |
+
"eval_Micro Precision": 0.990228013029316,
|
102 |
+
"eval_Micro Recall": 0.990228013029316,
|
103 |
+
"eval_Weighted F1": 0.9901932977316877,
|
104 |
+
"eval_Weighted Precision": 0.9903976746751036,
|
105 |
+
"eval_Weighted Recall": 0.990228013029316,
|
106 |
+
"eval_accuracy": 0.990228013029316,
|
107 |
+
"eval_loss": 0.05230216309428215,
|
108 |
+
"eval_runtime": 1.6848,
|
109 |
+
"eval_samples_per_second": 182.218,
|
110 |
+
"eval_steps_per_second": 11.871,
|
111 |
+
"step": 612
|
112 |
+
},
|
113 |
+
{
|
114 |
+
"epoch": 5.0,
|
115 |
+
"learning_rate": 0.0,
|
116 |
+
"loss": 0.012,
|
117 |
+
"step": 765
|
118 |
+
},
|
119 |
+
{
|
120 |
+
"epoch": 5.0,
|
121 |
+
"eval_Macro F1": 0.9831697750089234,
|
122 |
+
"eval_Macro Precision": 0.9845093632958801,
|
123 |
+
"eval_Macro Recall": 0.9820926966292135,
|
124 |
+
"eval_Micro F1": 0.9837133550488599,
|
125 |
+
"eval_Micro Precision": 0.9837133550488599,
|
126 |
+
"eval_Micro Recall": 0.9837133550488599,
|
127 |
+
"eval_Weighted F1": 0.983695143647141,
|
128 |
+
"eval_Weighted Precision": 0.9839478827361564,
|
129 |
+
"eval_Weighted Recall": 0.9837133550488599,
|
130 |
+
"eval_accuracy": 0.9837133550488599,
|
131 |
+
"eval_loss": 0.029342563822865486,
|
132 |
+
"eval_runtime": 1.6892,
|
133 |
+
"eval_samples_per_second": 181.741,
|
134 |
+
"eval_steps_per_second": 11.84,
|
135 |
+
"step": 765
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 5.0,
|
139 |
+
"step": 765,
|
140 |
+
"total_flos": 2539812775083840.0,
|
141 |
+
"train_loss": 0.14014302266189477,
|
142 |
+
"train_runtime": 281.4865,
|
143 |
+
"train_samples_per_second": 43.483,
|
144 |
+
"train_steps_per_second": 2.718
|
145 |
+
}
|
146 |
+
],
|
147 |
+
"logging_steps": 500,
|
148 |
+
"max_steps": 765,
|
149 |
+
"num_train_epochs": 5,
|
150 |
+
"save_steps": 500,
|
151 |
+
"total_flos": 2539812775083840.0,
|
152 |
+
"trial_name": null,
|
153 |
+
"trial_params": null
|
154 |
+
}
|