🍻 cheers
Browse files- README.md +15 -11
- all_results.json +14 -14
- eval_results.json +10 -10
- model.safetensors +1 -1
- runs/Jan05_17-18-52_DESKTOP-BDBS5RV/events.out.tfevents.1704442947.DESKTOP-BDBS5RV +2 -2
- runs/Jan05_17-26-06_DESKTOP-BDBS5RV/events.out.tfevents.1704443167.DESKTOP-BDBS5RV +3 -0
- runs/Jan05_17-26-06_DESKTOP-BDBS5RV/events.out.tfevents.1704443688.DESKTOP-BDBS5RV +3 -0
- train_results.json +5 -5
- trainer_state.json +364 -72
- training_args.bin +1 -1
README.md
CHANGED
@@ -24,10 +24,10 @@ model-index:
|
|
24 |
metrics:
|
25 |
- name: Accuracy
|
26 |
type: accuracy
|
27 |
-
value: 0.
|
28 |
- name: F1
|
29 |
type: f1
|
30 |
-
value: 0.
|
31 |
---
|
32 |
|
33 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
@@ -37,12 +37,12 @@ should probably proofread and complete it, then remove this comment. -->
|
|
37 |
|
38 |
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the stool-image dataset.
|
39 |
It achieves the following results on the evaluation set:
|
40 |
-
- Loss: 0.
|
41 |
-
- Auroc: 0.
|
42 |
-
- Accuracy: 0.
|
43 |
-
- Sensitivity: 0.
|
44 |
-
- Specificty: 0.
|
45 |
-
- F1: 0.
|
46 |
|
47 |
## Model description
|
48 |
|
@@ -67,15 +67,19 @@ The following hyperparameters were used during training:
|
|
67 |
- seed: 42
|
68 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
69 |
- lr_scheduler_type: linear
|
70 |
-
- num_epochs:
|
71 |
- mixed_precision_training: Native AMP
|
72 |
|
73 |
### Training results
|
74 |
|
75 |
| Training Loss | Epoch | Step | Validation Loss | Auroc | Accuracy | Sensitivity | Specificty | F1 |
|
76 |
|:-------------:|:-----:|:----:|:---------------:|:------:|:--------:|:-----------:|:----------:|:------:|
|
77 |
-
| 0.
|
78 |
-
| 0.
|
|
|
|
|
|
|
|
|
79 |
|
80 |
|
81 |
### Framework versions
|
|
|
24 |
metrics:
|
25 |
- name: Accuracy
|
26 |
type: accuracy
|
27 |
+
value: 0.8171064604185623
|
28 |
- name: F1
|
29 |
type: f1
|
30 |
+
value: 0.7841031149301826
|
31 |
---
|
32 |
|
33 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
37 |
|
38 |
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the stool-image dataset.
|
39 |
It achieves the following results on the evaluation set:
|
40 |
+
- Loss: 0.4538
|
41 |
+
- Auroc: 0.8897
|
42 |
+
- Accuracy: 0.8171
|
43 |
+
- Sensitivity: 0.8111
|
44 |
+
- Specificty: 0.8213
|
45 |
+
- F1: 0.7841
|
46 |
|
47 |
## Model description
|
48 |
|
|
|
67 |
- seed: 42
|
68 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
69 |
- lr_scheduler_type: linear
|
70 |
+
- num_epochs: 100
|
71 |
- mixed_precision_training: Native AMP
|
72 |
|
73 |
### Training results
|
74 |
|
75 |
| Training Loss | Epoch | Step | Validation Loss | Auroc | Accuracy | Sensitivity | Specificty | F1 |
|
76 |
|:-------------:|:-----:|:----:|:---------------:|:------:|:--------:|:-----------:|:----------:|:------:|
|
77 |
+
| 0.5303 | 0.98 | 100 | 0.4327 | 0.8826 | 0.7942 | 0.7191 | 0.8607 | 0.7665 |
|
78 |
+
| 0.3909 | 1.96 | 200 | 0.5196 | 0.8675 | 0.8047 | 0.8539 | 0.7612 | 0.8042 |
|
79 |
+
| 0.5328 | 2.94 | 300 | 0.4421 | 0.8864 | 0.8074 | 0.7528 | 0.8557 | 0.7859 |
|
80 |
+
| 0.4834 | 3.92 | 400 | 0.4721 | 0.8596 | 0.7757 | 0.7135 | 0.8308 | 0.7493 |
|
81 |
+
| 0.4209 | 4.9 | 500 | 0.4797 | 0.8625 | 0.7863 | 0.6798 | 0.8806 | 0.7492 |
|
82 |
+
| 0.4567 | 5.88 | 600 | 0.5150 | 0.8688 | 0.7942 | 0.6011 | 0.9652 | 0.7329 |
|
83 |
|
84 |
|
85 |
### Framework versions
|
all_results.json
CHANGED
@@ -1,16 +1,16 @@
|
|
1 |
{
|
2 |
-
"epoch":
|
3 |
-
"eval_accuracy": 0.
|
4 |
-
"eval_auroc": 0.
|
5 |
-
"eval_f1": 0.
|
6 |
-
"eval_loss": 0.
|
7 |
-
"eval_runtime":
|
8 |
-
"eval_samples_per_second": 17.
|
9 |
-
"eval_sensitivity": 0.
|
10 |
-
"eval_specificty": 0.
|
11 |
-
"eval_steps_per_second": 2.
|
12 |
-
"train_loss": 0.
|
13 |
-
"train_runtime":
|
14 |
-
"train_samples_per_second":
|
15 |
-
"train_steps_per_second":
|
16 |
}
|
|
|
1 |
{
|
2 |
+
"epoch": 5.88,
|
3 |
+
"eval_accuracy": 0.8171064604185623,
|
4 |
+
"eval_auroc": 0.8896644410203732,
|
5 |
+
"eval_f1": 0.7841031149301826,
|
6 |
+
"eval_loss": 0.45378199219703674,
|
7 |
+
"eval_runtime": 63.9212,
|
8 |
+
"eval_samples_per_second": 17.193,
|
9 |
+
"eval_sensitivity": 0.8111111111111111,
|
10 |
+
"eval_specificty": 0.8212634822804314,
|
11 |
+
"eval_steps_per_second": 2.159,
|
12 |
+
"train_loss": 0.4638279656569163,
|
13 |
+
"train_runtime": 457.0802,
|
14 |
+
"train_samples_per_second": 355.517,
|
15 |
+
"train_steps_per_second": 22.316
|
16 |
}
|
eval_results.json
CHANGED
@@ -1,12 +1,12 @@
|
|
1 |
{
|
2 |
-
"epoch":
|
3 |
-
"eval_accuracy": 0.
|
4 |
-
"eval_auroc": 0.
|
5 |
-
"eval_f1": 0.
|
6 |
-
"eval_loss": 0.
|
7 |
-
"eval_runtime":
|
8 |
-
"eval_samples_per_second": 17.
|
9 |
-
"eval_sensitivity": 0.
|
10 |
-
"eval_specificty": 0.
|
11 |
-
"eval_steps_per_second": 2.
|
12 |
}
|
|
|
1 |
{
|
2 |
+
"epoch": 5.88,
|
3 |
+
"eval_accuracy": 0.8171064604185623,
|
4 |
+
"eval_auroc": 0.8896644410203732,
|
5 |
+
"eval_f1": 0.7841031149301826,
|
6 |
+
"eval_loss": 0.45378199219703674,
|
7 |
+
"eval_runtime": 63.9212,
|
8 |
+
"eval_samples_per_second": 17.193,
|
9 |
+
"eval_sensitivity": 0.8111111111111111,
|
10 |
+
"eval_specificty": 0.8212634822804314,
|
11 |
+
"eval_steps_per_second": 2.159
|
12 |
}
|
model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 343223968
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:bce917709538915003c0c8fe4efef00dae13bd7d04517b013edb5b0d28d71e17
|
3 |
size 343223968
|
runs/Jan05_17-18-52_DESKTOP-BDBS5RV/events.out.tfevents.1704442947.DESKTOP-BDBS5RV
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c3fe79f35723ce861fee77287d519e1c18d951a60862b85645fcd34c53f49b1e
|
3 |
+
size 567
|
runs/Jan05_17-26-06_DESKTOP-BDBS5RV/events.out.tfevents.1704443167.DESKTOP-BDBS5RV
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3cf8bb6f30306e3c6d48bd767fb74682b740f846ee095bd0a651196b77c48999
|
3 |
+
size 17243
|
runs/Jan05_17-26-06_DESKTOP-BDBS5RV/events.out.tfevents.1704443688.DESKTOP-BDBS5RV
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:09b62303b6c67ec8004a1373b7d8162f85881f93156735bc5dc4c83946ad955c
|
3 |
+
size 40
|
train_results.json
CHANGED
@@ -1,7 +1,7 @@
|
|
1 |
{
|
2 |
-
"epoch":
|
3 |
-
"train_loss": 0.
|
4 |
-
"train_runtime":
|
5 |
-
"train_samples_per_second":
|
6 |
-
"train_steps_per_second":
|
7 |
}
|
|
|
1 |
{
|
2 |
+
"epoch": 5.88,
|
3 |
+
"train_loss": 0.4638279656569163,
|
4 |
+
"train_runtime": 457.0802,
|
5 |
+
"train_samples_per_second": 355.517,
|
6 |
+
"train_steps_per_second": 22.316
|
7 |
}
|
trainer_state.json
CHANGED
@@ -1,175 +1,467 @@
|
|
1 |
{
|
2 |
-
"best_metric": 0.
|
3 |
-
"best_model_checkpoint": "./stool-condition-classification\\checkpoint-
|
4 |
-
"epoch":
|
5 |
"eval_steps": 100,
|
6 |
-
"global_step":
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 0.1,
|
13 |
-
"learning_rate": 0.
|
14 |
-
"loss": 0.
|
15 |
"step": 10
|
16 |
},
|
17 |
{
|
18 |
"epoch": 0.2,
|
19 |
-
"learning_rate": 0.
|
20 |
-
"loss": 0.
|
21 |
"step": 20
|
22 |
},
|
23 |
{
|
24 |
"epoch": 0.29,
|
25 |
-
"learning_rate": 0.
|
26 |
-
"loss": 0.
|
27 |
"step": 30
|
28 |
},
|
29 |
{
|
30 |
"epoch": 0.39,
|
31 |
-
"learning_rate": 0.
|
32 |
-
"loss": 0.
|
33 |
"step": 40
|
34 |
},
|
35 |
{
|
36 |
"epoch": 0.49,
|
37 |
-
"learning_rate": 0.
|
38 |
-
"loss": 0.
|
39 |
"step": 50
|
40 |
},
|
41 |
{
|
42 |
"epoch": 0.59,
|
43 |
-
"learning_rate": 0.
|
44 |
-
"loss": 0.
|
45 |
"step": 60
|
46 |
},
|
47 |
{
|
48 |
"epoch": 0.69,
|
49 |
-
"learning_rate": 0.
|
50 |
-
"loss": 0.
|
51 |
"step": 70
|
52 |
},
|
53 |
{
|
54 |
"epoch": 0.78,
|
55 |
-
"learning_rate": 0.
|
56 |
-
"loss": 0.
|
57 |
"step": 80
|
58 |
},
|
59 |
{
|
60 |
"epoch": 0.88,
|
61 |
-
"learning_rate": 0.
|
62 |
-
"loss": 0.
|
63 |
"step": 90
|
64 |
},
|
65 |
{
|
66 |
"epoch": 0.98,
|
67 |
-
"learning_rate": 0.
|
68 |
-
"loss": 0.
|
69 |
"step": 100
|
70 |
},
|
71 |
{
|
72 |
"epoch": 0.98,
|
73 |
-
"eval_accuracy": 0.
|
74 |
-
"eval_auroc": 0.
|
75 |
-
"eval_f1": 0.
|
76 |
-
"eval_loss": 0.
|
77 |
-
"eval_runtime":
|
78 |
-
"eval_samples_per_second":
|
79 |
-
"eval_sensitivity": 0.
|
80 |
-
"eval_specificty": 0.
|
81 |
-
"eval_steps_per_second": 2.
|
82 |
"step": 100
|
83 |
},
|
84 |
{
|
85 |
"epoch": 1.08,
|
86 |
-
"learning_rate":
|
87 |
-
"loss": 0.
|
88 |
"step": 110
|
89 |
},
|
90 |
{
|
91 |
"epoch": 1.18,
|
92 |
-
"learning_rate":
|
93 |
-
"loss": 0.
|
94 |
"step": 120
|
95 |
},
|
96 |
{
|
97 |
"epoch": 1.27,
|
98 |
-
"learning_rate":
|
99 |
-
"loss": 0.
|
100 |
"step": 130
|
101 |
},
|
102 |
{
|
103 |
"epoch": 1.37,
|
104 |
-
"learning_rate":
|
105 |
-
"loss": 0.
|
106 |
"step": 140
|
107 |
},
|
108 |
{
|
109 |
"epoch": 1.47,
|
110 |
-
"learning_rate":
|
111 |
-
"loss": 0.
|
112 |
"step": 150
|
113 |
},
|
114 |
{
|
115 |
"epoch": 1.57,
|
116 |
-
"learning_rate":
|
117 |
-
"loss": 0.
|
118 |
"step": 160
|
119 |
},
|
120 |
{
|
121 |
"epoch": 1.67,
|
122 |
-
"learning_rate":
|
123 |
-
"loss": 0.
|
124 |
"step": 170
|
125 |
},
|
126 |
{
|
127 |
"epoch": 1.76,
|
128 |
-
"learning_rate":
|
129 |
-
"loss": 0.
|
130 |
"step": 180
|
131 |
},
|
132 |
{
|
133 |
"epoch": 1.86,
|
134 |
-
"learning_rate":
|
135 |
-
"loss": 0.
|
136 |
"step": 190
|
137 |
},
|
138 |
{
|
139 |
"epoch": 1.96,
|
140 |
-
"learning_rate":
|
141 |
-
"loss": 0.
|
142 |
"step": 200
|
143 |
},
|
144 |
{
|
145 |
"epoch": 1.96,
|
146 |
-
"eval_accuracy": 0.
|
147 |
-
"eval_auroc": 0.
|
148 |
-
"eval_f1": 0.
|
149 |
-
"eval_loss": 0.
|
150 |
-
"eval_runtime":
|
151 |
-
"eval_samples_per_second":
|
152 |
-
"eval_sensitivity": 0.
|
153 |
-
"eval_specificty": 0.
|
154 |
-
"eval_steps_per_second": 2.
|
155 |
"step": 200
|
156 |
},
|
157 |
{
|
158 |
-
"epoch": 2.
|
159 |
-
"
|
160 |
-
"
|
161 |
-
"
|
162 |
-
|
163 |
-
|
164 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
165 |
}
|
166 |
],
|
167 |
"logging_steps": 10,
|
168 |
-
"max_steps":
|
169 |
"num_input_tokens_seen": 0,
|
170 |
-
"num_train_epochs":
|
171 |
"save_steps": 100,
|
172 |
-
"total_flos":
|
173 |
"train_batch_size": 16,
|
174 |
"trial_name": null,
|
175 |
"trial_params": null
|
|
|
1 |
{
|
2 |
+
"best_metric": 0.8863966683436749,
|
3 |
+
"best_model_checkpoint": "./stool-condition-classification\\checkpoint-300",
|
4 |
+
"epoch": 5.882352941176471,
|
5 |
"eval_steps": 100,
|
6 |
+
"global_step": 600,
|
7 |
"is_hyper_param_search": false,
|
8 |
"is_local_process_zero": true,
|
9 |
"is_world_process_zero": true,
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 0.1,
|
13 |
+
"learning_rate": 0.00019980392156862745,
|
14 |
+
"loss": 0.6109,
|
15 |
"step": 10
|
16 |
},
|
17 |
{
|
18 |
"epoch": 0.2,
|
19 |
+
"learning_rate": 0.00019960784313725492,
|
20 |
+
"loss": 0.5518,
|
21 |
"step": 20
|
22 |
},
|
23 |
{
|
24 |
"epoch": 0.29,
|
25 |
+
"learning_rate": 0.00019941176470588236,
|
26 |
+
"loss": 0.4854,
|
27 |
"step": 30
|
28 |
},
|
29 |
{
|
30 |
"epoch": 0.39,
|
31 |
+
"learning_rate": 0.0001992156862745098,
|
32 |
+
"loss": 0.5984,
|
33 |
"step": 40
|
34 |
},
|
35 |
{
|
36 |
"epoch": 0.49,
|
37 |
+
"learning_rate": 0.00019901960784313727,
|
38 |
+
"loss": 0.5832,
|
39 |
"step": 50
|
40 |
},
|
41 |
{
|
42 |
"epoch": 0.59,
|
43 |
+
"learning_rate": 0.00019882352941176472,
|
44 |
+
"loss": 0.5199,
|
45 |
"step": 60
|
46 |
},
|
47 |
{
|
48 |
"epoch": 0.69,
|
49 |
+
"learning_rate": 0.00019862745098039218,
|
50 |
+
"loss": 0.5347,
|
51 |
"step": 70
|
52 |
},
|
53 |
{
|
54 |
"epoch": 0.78,
|
55 |
+
"learning_rate": 0.00019843137254901963,
|
56 |
+
"loss": 0.5159,
|
57 |
"step": 80
|
58 |
},
|
59 |
{
|
60 |
"epoch": 0.88,
|
61 |
+
"learning_rate": 0.00019823529411764707,
|
62 |
+
"loss": 0.5069,
|
63 |
"step": 90
|
64 |
},
|
65 |
{
|
66 |
"epoch": 0.98,
|
67 |
+
"learning_rate": 0.00019803921568627454,
|
68 |
+
"loss": 0.5303,
|
69 |
"step": 100
|
70 |
},
|
71 |
{
|
72 |
"epoch": 0.98,
|
73 |
+
"eval_accuracy": 0.7941952506596306,
|
74 |
+
"eval_auroc": 0.8826094247861815,
|
75 |
+
"eval_f1": 0.7664670658682635,
|
76 |
+
"eval_loss": 0.43268465995788574,
|
77 |
+
"eval_runtime": 19.5335,
|
78 |
+
"eval_samples_per_second": 19.403,
|
79 |
+
"eval_sensitivity": 0.7191011235955056,
|
80 |
+
"eval_specificty": 0.8606965174129353,
|
81 |
+
"eval_steps_per_second": 2.457,
|
82 |
"step": 100
|
83 |
},
|
84 |
{
|
85 |
"epoch": 1.08,
|
86 |
+
"learning_rate": 0.00019784313725490198,
|
87 |
+
"loss": 0.514,
|
88 |
"step": 110
|
89 |
},
|
90 |
{
|
91 |
"epoch": 1.18,
|
92 |
+
"learning_rate": 0.00019764705882352942,
|
93 |
+
"loss": 0.3919,
|
94 |
"step": 120
|
95 |
},
|
96 |
{
|
97 |
"epoch": 1.27,
|
98 |
+
"learning_rate": 0.0001974509803921569,
|
99 |
+
"loss": 0.5246,
|
100 |
"step": 130
|
101 |
},
|
102 |
{
|
103 |
"epoch": 1.37,
|
104 |
+
"learning_rate": 0.00019725490196078433,
|
105 |
+
"loss": 0.5262,
|
106 |
"step": 140
|
107 |
},
|
108 |
{
|
109 |
"epoch": 1.47,
|
110 |
+
"learning_rate": 0.00019705882352941177,
|
111 |
+
"loss": 0.4092,
|
112 |
"step": 150
|
113 |
},
|
114 |
{
|
115 |
"epoch": 1.57,
|
116 |
+
"learning_rate": 0.00019686274509803922,
|
117 |
+
"loss": 0.6038,
|
118 |
"step": 160
|
119 |
},
|
120 |
{
|
121 |
"epoch": 1.67,
|
122 |
+
"learning_rate": 0.00019666666666666666,
|
123 |
+
"loss": 0.5079,
|
124 |
"step": 170
|
125 |
},
|
126 |
{
|
127 |
"epoch": 1.76,
|
128 |
+
"learning_rate": 0.00019647058823529413,
|
129 |
+
"loss": 0.4992,
|
130 |
"step": 180
|
131 |
},
|
132 |
{
|
133 |
"epoch": 1.86,
|
134 |
+
"learning_rate": 0.00019627450980392157,
|
135 |
+
"loss": 0.4788,
|
136 |
"step": 190
|
137 |
},
|
138 |
{
|
139 |
"epoch": 1.96,
|
140 |
+
"learning_rate": 0.000196078431372549,
|
141 |
+
"loss": 0.3909,
|
142 |
"step": 200
|
143 |
},
|
144 |
{
|
145 |
"epoch": 1.96,
|
146 |
+
"eval_accuracy": 0.8047493403693932,
|
147 |
+
"eval_auroc": 0.8674604505562076,
|
148 |
+
"eval_f1": 0.8042328042328041,
|
149 |
+
"eval_loss": 0.5196136832237244,
|
150 |
+
"eval_runtime": 19.2864,
|
151 |
+
"eval_samples_per_second": 19.651,
|
152 |
+
"eval_sensitivity": 0.8539325842696629,
|
153 |
+
"eval_specificty": 0.7611940298507462,
|
154 |
+
"eval_steps_per_second": 2.489,
|
155 |
"step": 200
|
156 |
},
|
157 |
{
|
158 |
+
"epoch": 2.06,
|
159 |
+
"learning_rate": 0.00019588235294117648,
|
160 |
+
"loss": 0.4421,
|
161 |
+
"step": 210
|
162 |
+
},
|
163 |
+
{
|
164 |
+
"epoch": 2.16,
|
165 |
+
"learning_rate": 0.00019568627450980392,
|
166 |
+
"loss": 0.4479,
|
167 |
+
"step": 220
|
168 |
+
},
|
169 |
+
{
|
170 |
+
"epoch": 2.25,
|
171 |
+
"learning_rate": 0.00019549019607843136,
|
172 |
+
"loss": 0.4655,
|
173 |
+
"step": 230
|
174 |
+
},
|
175 |
+
{
|
176 |
+
"epoch": 2.35,
|
177 |
+
"learning_rate": 0.00019529411764705883,
|
178 |
+
"loss": 0.4479,
|
179 |
+
"step": 240
|
180 |
+
},
|
181 |
+
{
|
182 |
+
"epoch": 2.45,
|
183 |
+
"learning_rate": 0.00019509803921568628,
|
184 |
+
"loss": 0.3894,
|
185 |
+
"step": 250
|
186 |
+
},
|
187 |
+
{
|
188 |
+
"epoch": 2.55,
|
189 |
+
"learning_rate": 0.00019490196078431372,
|
190 |
+
"loss": 0.4315,
|
191 |
+
"step": 260
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 2.65,
|
195 |
+
"learning_rate": 0.0001947058823529412,
|
196 |
+
"loss": 0.4783,
|
197 |
+
"step": 270
|
198 |
+
},
|
199 |
+
{
|
200 |
+
"epoch": 2.75,
|
201 |
+
"learning_rate": 0.00019450980392156863,
|
202 |
+
"loss": 0.3792,
|
203 |
+
"step": 280
|
204 |
+
},
|
205 |
+
{
|
206 |
+
"epoch": 2.84,
|
207 |
+
"learning_rate": 0.0001943137254901961,
|
208 |
+
"loss": 0.5301,
|
209 |
+
"step": 290
|
210 |
+
},
|
211 |
+
{
|
212 |
+
"epoch": 2.94,
|
213 |
+
"learning_rate": 0.00019411764705882354,
|
214 |
+
"loss": 0.5328,
|
215 |
+
"step": 300
|
216 |
+
},
|
217 |
+
{
|
218 |
+
"epoch": 2.94,
|
219 |
+
"eval_accuracy": 0.8073878627968337,
|
220 |
+
"eval_auroc": 0.8863966683436749,
|
221 |
+
"eval_f1": 0.7859237536656891,
|
222 |
+
"eval_loss": 0.4421083927154541,
|
223 |
+
"eval_runtime": 19.1397,
|
224 |
+
"eval_samples_per_second": 19.802,
|
225 |
+
"eval_sensitivity": 0.7528089887640449,
|
226 |
+
"eval_specificty": 0.8557213930348259,
|
227 |
+
"eval_steps_per_second": 2.508,
|
228 |
+
"step": 300
|
229 |
+
},
|
230 |
+
{
|
231 |
+
"epoch": 3.04,
|
232 |
+
"learning_rate": 0.00019392156862745098,
|
233 |
+
"loss": 0.4437,
|
234 |
+
"step": 310
|
235 |
+
},
|
236 |
+
{
|
237 |
+
"epoch": 3.14,
|
238 |
+
"learning_rate": 0.00019372549019607845,
|
239 |
+
"loss": 0.3859,
|
240 |
+
"step": 320
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 3.24,
|
244 |
+
"learning_rate": 0.0001935294117647059,
|
245 |
+
"loss": 0.4995,
|
246 |
+
"step": 330
|
247 |
+
},
|
248 |
+
{
|
249 |
+
"epoch": 3.33,
|
250 |
+
"learning_rate": 0.00019333333333333333,
|
251 |
+
"loss": 0.4282,
|
252 |
+
"step": 340
|
253 |
+
},
|
254 |
+
{
|
255 |
+
"epoch": 3.43,
|
256 |
+
"learning_rate": 0.0001931372549019608,
|
257 |
+
"loss": 0.4742,
|
258 |
+
"step": 350
|
259 |
+
},
|
260 |
+
{
|
261 |
+
"epoch": 3.53,
|
262 |
+
"learning_rate": 0.00019294117647058825,
|
263 |
+
"loss": 0.4219,
|
264 |
+
"step": 360
|
265 |
+
},
|
266 |
+
{
|
267 |
+
"epoch": 3.63,
|
268 |
+
"learning_rate": 0.0001927450980392157,
|
269 |
+
"loss": 0.4531,
|
270 |
+
"step": 370
|
271 |
+
},
|
272 |
+
{
|
273 |
+
"epoch": 3.73,
|
274 |
+
"learning_rate": 0.00019254901960784316,
|
275 |
+
"loss": 0.545,
|
276 |
+
"step": 380
|
277 |
+
},
|
278 |
+
{
|
279 |
+
"epoch": 3.82,
|
280 |
+
"learning_rate": 0.0001923529411764706,
|
281 |
+
"loss": 0.4216,
|
282 |
+
"step": 390
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 3.92,
|
286 |
+
"learning_rate": 0.00019215686274509807,
|
287 |
+
"loss": 0.4834,
|
288 |
+
"step": 400
|
289 |
+
},
|
290 |
+
{
|
291 |
+
"epoch": 3.92,
|
292 |
+
"eval_accuracy": 0.7757255936675461,
|
293 |
+
"eval_auroc": 0.8596483872770976,
|
294 |
+
"eval_f1": 0.7492625368731564,
|
295 |
+
"eval_loss": 0.4720683693885803,
|
296 |
+
"eval_runtime": 19.0669,
|
297 |
+
"eval_samples_per_second": 19.877,
|
298 |
+
"eval_sensitivity": 0.7134831460674157,
|
299 |
+
"eval_specificty": 0.8308457711442786,
|
300 |
+
"eval_steps_per_second": 2.517,
|
301 |
+
"step": 400
|
302 |
+
},
|
303 |
+
{
|
304 |
+
"epoch": 4.02,
|
305 |
+
"learning_rate": 0.0001919607843137255,
|
306 |
+
"loss": 0.3281,
|
307 |
+
"step": 410
|
308 |
+
},
|
309 |
+
{
|
310 |
+
"epoch": 4.12,
|
311 |
+
"learning_rate": 0.00019176470588235295,
|
312 |
+
"loss": 0.5168,
|
313 |
+
"step": 420
|
314 |
+
},
|
315 |
+
{
|
316 |
+
"epoch": 4.22,
|
317 |
+
"learning_rate": 0.00019156862745098042,
|
318 |
+
"loss": 0.5219,
|
319 |
+
"step": 430
|
320 |
+
},
|
321 |
+
{
|
322 |
+
"epoch": 4.31,
|
323 |
+
"learning_rate": 0.00019137254901960786,
|
324 |
+
"loss": 0.4131,
|
325 |
+
"step": 440
|
326 |
+
},
|
327 |
+
{
|
328 |
+
"epoch": 4.41,
|
329 |
+
"learning_rate": 0.0001911764705882353,
|
330 |
+
"loss": 0.4494,
|
331 |
+
"step": 450
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 4.51,
|
335 |
+
"learning_rate": 0.00019098039215686277,
|
336 |
+
"loss": 0.4206,
|
337 |
+
"step": 460
|
338 |
+
},
|
339 |
+
{
|
340 |
+
"epoch": 4.61,
|
341 |
+
"learning_rate": 0.00019078431372549022,
|
342 |
+
"loss": 0.3932,
|
343 |
+
"step": 470
|
344 |
+
},
|
345 |
+
{
|
346 |
+
"epoch": 4.71,
|
347 |
+
"learning_rate": 0.00019058823529411766,
|
348 |
+
"loss": 0.4607,
|
349 |
+
"step": 480
|
350 |
+
},
|
351 |
+
{
|
352 |
+
"epoch": 4.8,
|
353 |
+
"learning_rate": 0.0001903921568627451,
|
354 |
+
"loss": 0.3892,
|
355 |
+
"step": 490
|
356 |
+
},
|
357 |
+
{
|
358 |
+
"epoch": 4.9,
|
359 |
+
"learning_rate": 0.00019019607843137254,
|
360 |
+
"loss": 0.4209,
|
361 |
+
"step": 500
|
362 |
+
},
|
363 |
+
{
|
364 |
+
"epoch": 4.9,
|
365 |
+
"eval_accuracy": 0.7862796833773087,
|
366 |
+
"eval_auroc": 0.8624853261780983,
|
367 |
+
"eval_f1": 0.7492260061919503,
|
368 |
+
"eval_loss": 0.47969380021095276,
|
369 |
+
"eval_runtime": 18.951,
|
370 |
+
"eval_samples_per_second": 19.999,
|
371 |
+
"eval_sensitivity": 0.6797752808988764,
|
372 |
+
"eval_specificty": 0.8805970149253731,
|
373 |
+
"eval_steps_per_second": 2.533,
|
374 |
+
"step": 500
|
375 |
+
},
|
376 |
+
{
|
377 |
+
"epoch": 5.0,
|
378 |
+
"learning_rate": 0.00019,
|
379 |
+
"loss": 0.3681,
|
380 |
+
"step": 510
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 5.1,
|
384 |
+
"learning_rate": 0.00018980392156862745,
|
385 |
+
"loss": 0.397,
|
386 |
+
"step": 520
|
387 |
+
},
|
388 |
+
{
|
389 |
+
"epoch": 5.2,
|
390 |
+
"learning_rate": 0.0001896078431372549,
|
391 |
+
"loss": 0.3833,
|
392 |
+
"step": 530
|
393 |
+
},
|
394 |
+
{
|
395 |
+
"epoch": 5.29,
|
396 |
+
"learning_rate": 0.00018941176470588236,
|
397 |
+
"loss": 0.4202,
|
398 |
+
"step": 540
|
399 |
+
},
|
400 |
+
{
|
401 |
+
"epoch": 5.39,
|
402 |
+
"learning_rate": 0.0001892156862745098,
|
403 |
+
"loss": 0.4304,
|
404 |
+
"step": 550
|
405 |
+
},
|
406 |
+
{
|
407 |
+
"epoch": 5.49,
|
408 |
+
"learning_rate": 0.00018901960784313725,
|
409 |
+
"loss": 0.3883,
|
410 |
+
"step": 560
|
411 |
+
},
|
412 |
+
{
|
413 |
+
"epoch": 5.59,
|
414 |
+
"learning_rate": 0.00018882352941176472,
|
415 |
+
"loss": 0.4737,
|
416 |
+
"step": 570
|
417 |
+
},
|
418 |
+
{
|
419 |
+
"epoch": 5.69,
|
420 |
+
"learning_rate": 0.00018862745098039216,
|
421 |
+
"loss": 0.4661,
|
422 |
+
"step": 580
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 5.78,
|
426 |
+
"learning_rate": 0.0001884313725490196,
|
427 |
+
"loss": 0.3469,
|
428 |
+
"step": 590
|
429 |
+
},
|
430 |
+
{
|
431 |
+
"epoch": 5.88,
|
432 |
+
"learning_rate": 0.00018823529411764707,
|
433 |
+
"loss": 0.4567,
|
434 |
+
"step": 600
|
435 |
+
},
|
436 |
+
{
|
437 |
+
"epoch": 5.88,
|
438 |
+
"eval_accuracy": 0.7941952506596306,
|
439 |
+
"eval_auroc": 0.8687741069931243,
|
440 |
+
"eval_f1": 0.7328767123287672,
|
441 |
+
"eval_loss": 0.514951765537262,
|
442 |
+
"eval_runtime": 18.7983,
|
443 |
+
"eval_samples_per_second": 20.161,
|
444 |
+
"eval_sensitivity": 0.601123595505618,
|
445 |
+
"eval_specificty": 0.9651741293532339,
|
446 |
+
"eval_steps_per_second": 2.553,
|
447 |
+
"step": 600
|
448 |
+
},
|
449 |
+
{
|
450 |
+
"epoch": 5.88,
|
451 |
+
"step": 600,
|
452 |
+
"total_flos": 7.412108806635725e+17,
|
453 |
+
"train_loss": 0.4638279656569163,
|
454 |
+
"train_runtime": 457.0802,
|
455 |
+
"train_samples_per_second": 355.517,
|
456 |
+
"train_steps_per_second": 22.316
|
457 |
}
|
458 |
],
|
459 |
"logging_steps": 10,
|
460 |
+
"max_steps": 10200,
|
461 |
"num_input_tokens_seen": 0,
|
462 |
+
"num_train_epochs": 100,
|
463 |
"save_steps": 100,
|
464 |
+
"total_flos": 7.412108806635725e+17,
|
465 |
"train_batch_size": 16,
|
466 |
"trial_name": null,
|
467 |
"trial_params": null
|
training_args.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 4283
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:00021469e2fc2da6e352c64ba5c1eff722ad7949ad69e804b54677deb2c22f06
|
3 |
size 4283
|