End of training
Browse files- README.md +1 -1
- all_results.json +13 -0
- eval_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +163 -0
README.md
CHANGED
@@ -31,7 +31,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
31 |
|
32 |
This model is a fine-tuned version of [microsoft/dit-base-finetuned-rvlcdip](https://huggingface.co/microsoft/dit-base-finetuned-rvlcdip) on the imagefolder dataset.
|
33 |
It achieves the following results on the evaluation set:
|
34 |
-
- Loss: 1.
|
35 |
- Accuracy: 0.7576
|
36 |
|
37 |
## Model description
|
|
|
31 |
|
32 |
This model is a fine-tuned version of [microsoft/dit-base-finetuned-rvlcdip](https://huggingface.co/microsoft/dit-base-finetuned-rvlcdip) on the imagefolder dataset.
|
33 |
It achieves the following results on the evaluation set:
|
34 |
+
- Loss: 1.3033
|
35 |
- Accuracy: 0.7576
|
36 |
|
37 |
## Model description
|
all_results.json
ADDED
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 6.72,
|
3 |
+
"eval_accuracy": 0.7575757575757576,
|
4 |
+
"eval_loss": 1.3033273220062256,
|
5 |
+
"eval_runtime": 14.5204,
|
6 |
+
"eval_samples_per_second": 18.181,
|
7 |
+
"eval_steps_per_second": 0.62,
|
8 |
+
"total_flos": 1.2388649195611423e+18,
|
9 |
+
"train_loss": 1.7003454405163962,
|
10 |
+
"train_runtime": 3667.1764,
|
11 |
+
"train_samples_per_second": 4.535,
|
12 |
+
"train_steps_per_second": 0.034
|
13 |
+
}
|
eval_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 6.72,
|
3 |
+
"eval_accuracy": 0.7575757575757576,
|
4 |
+
"eval_loss": 1.3033273220062256,
|
5 |
+
"eval_runtime": 14.5204,
|
6 |
+
"eval_samples_per_second": 18.181,
|
7 |
+
"eval_steps_per_second": 0.62
|
8 |
+
}
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 6.72,
|
3 |
+
"total_flos": 1.2388649195611423e+18,
|
4 |
+
"train_loss": 1.7003454405163962,
|
5 |
+
"train_runtime": 3667.1764,
|
6 |
+
"train_samples_per_second": 4.535,
|
7 |
+
"train_steps_per_second": 0.034
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,163 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.7575757575757576,
|
3 |
+
"best_model_checkpoint": "dit-base-rvlcdip-finetuned-grp-actual/checkpoint-93",
|
4 |
+
"epoch": 6.72,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 126,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.53,
|
13 |
+
"learning_rate": 3.846153846153846e-05,
|
14 |
+
"loss": 2.3577,
|
15 |
+
"step": 10
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"epoch": 0.96,
|
19 |
+
"eval_accuracy": 0.5113636363636364,
|
20 |
+
"eval_loss": 2.086306571960449,
|
21 |
+
"eval_runtime": 240.3077,
|
22 |
+
"eval_samples_per_second": 1.099,
|
23 |
+
"eval_steps_per_second": 0.037,
|
24 |
+
"step": 18
|
25 |
+
},
|
26 |
+
{
|
27 |
+
"epoch": 1.07,
|
28 |
+
"learning_rate": 4.690265486725664e-05,
|
29 |
+
"loss": 2.2163,
|
30 |
+
"step": 20
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 1.6,
|
34 |
+
"learning_rate": 4.247787610619469e-05,
|
35 |
+
"loss": 2.0601,
|
36 |
+
"step": 30
|
37 |
+
},
|
38 |
+
{
|
39 |
+
"epoch": 1.97,
|
40 |
+
"eval_accuracy": 0.6477272727272727,
|
41 |
+
"eval_loss": 1.8153679370880127,
|
42 |
+
"eval_runtime": 14.6331,
|
43 |
+
"eval_samples_per_second": 18.041,
|
44 |
+
"eval_steps_per_second": 0.615,
|
45 |
+
"step": 37
|
46 |
+
},
|
47 |
+
{
|
48 |
+
"epoch": 2.13,
|
49 |
+
"learning_rate": 3.8053097345132744e-05,
|
50 |
+
"loss": 1.943,
|
51 |
+
"step": 40
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 2.67,
|
55 |
+
"learning_rate": 3.3628318584070804e-05,
|
56 |
+
"loss": 1.8068,
|
57 |
+
"step": 50
|
58 |
+
},
|
59 |
+
{
|
60 |
+
"epoch": 2.99,
|
61 |
+
"eval_accuracy": 0.6704545454545454,
|
62 |
+
"eval_loss": 1.5881296396255493,
|
63 |
+
"eval_runtime": 14.8318,
|
64 |
+
"eval_samples_per_second": 17.8,
|
65 |
+
"eval_steps_per_second": 0.607,
|
66 |
+
"step": 56
|
67 |
+
},
|
68 |
+
{
|
69 |
+
"epoch": 3.2,
|
70 |
+
"learning_rate": 2.9203539823008852e-05,
|
71 |
+
"loss": 1.66,
|
72 |
+
"step": 60
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 3.73,
|
76 |
+
"learning_rate": 2.4778761061946905e-05,
|
77 |
+
"loss": 1.5953,
|
78 |
+
"step": 70
|
79 |
+
},
|
80 |
+
{
|
81 |
+
"epoch": 4.0,
|
82 |
+
"eval_accuracy": 0.7159090909090909,
|
83 |
+
"eval_loss": 1.4111517667770386,
|
84 |
+
"eval_runtime": 14.6707,
|
85 |
+
"eval_samples_per_second": 17.995,
|
86 |
+
"eval_steps_per_second": 0.613,
|
87 |
+
"step": 75
|
88 |
+
},
|
89 |
+
{
|
90 |
+
"epoch": 4.27,
|
91 |
+
"learning_rate": 2.0353982300884957e-05,
|
92 |
+
"loss": 1.4929,
|
93 |
+
"step": 80
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 4.8,
|
97 |
+
"learning_rate": 1.592920353982301e-05,
|
98 |
+
"loss": 1.4304,
|
99 |
+
"step": 90
|
100 |
+
},
|
101 |
+
{
|
102 |
+
"epoch": 4.96,
|
103 |
+
"eval_accuracy": 0.7575757575757576,
|
104 |
+
"eval_loss": 1.3033273220062256,
|
105 |
+
"eval_runtime": 14.9724,
|
106 |
+
"eval_samples_per_second": 17.632,
|
107 |
+
"eval_steps_per_second": 0.601,
|
108 |
+
"step": 93
|
109 |
+
},
|
110 |
+
{
|
111 |
+
"epoch": 5.33,
|
112 |
+
"learning_rate": 1.1504424778761062e-05,
|
113 |
+
"loss": 1.3606,
|
114 |
+
"step": 100
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 5.87,
|
118 |
+
"learning_rate": 7.079646017699115e-06,
|
119 |
+
"loss": 1.3458,
|
120 |
+
"step": 110
|
121 |
+
},
|
122 |
+
{
|
123 |
+
"epoch": 5.97,
|
124 |
+
"eval_accuracy": 0.75,
|
125 |
+
"eval_loss": 1.2400753498077393,
|
126 |
+
"eval_runtime": 14.8483,
|
127 |
+
"eval_samples_per_second": 17.78,
|
128 |
+
"eval_steps_per_second": 0.606,
|
129 |
+
"step": 112
|
130 |
+
},
|
131 |
+
{
|
132 |
+
"epoch": 6.4,
|
133 |
+
"learning_rate": 2.6548672566371683e-06,
|
134 |
+
"loss": 1.3523,
|
135 |
+
"step": 120
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 6.72,
|
139 |
+
"eval_accuracy": 0.7575757575757576,
|
140 |
+
"eval_loss": 1.2240339517593384,
|
141 |
+
"eval_runtime": 14.9425,
|
142 |
+
"eval_samples_per_second": 17.668,
|
143 |
+
"eval_steps_per_second": 0.602,
|
144 |
+
"step": 126
|
145 |
+
},
|
146 |
+
{
|
147 |
+
"epoch": 6.72,
|
148 |
+
"step": 126,
|
149 |
+
"total_flos": 1.2388649195611423e+18,
|
150 |
+
"train_loss": 1.7003454405163962,
|
151 |
+
"train_runtime": 3667.1764,
|
152 |
+
"train_samples_per_second": 4.535,
|
153 |
+
"train_steps_per_second": 0.034
|
154 |
+
}
|
155 |
+
],
|
156 |
+
"logging_steps": 10,
|
157 |
+
"max_steps": 126,
|
158 |
+
"num_train_epochs": 7,
|
159 |
+
"save_steps": 500,
|
160 |
+
"total_flos": 1.2388649195611423e+18,
|
161 |
+
"trial_name": null,
|
162 |
+
"trial_params": null
|
163 |
+
}
|