Training in progress, step 7000, checkpoint
Browse files
checkpoint-7000/model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 1711848436
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0142364f978126efa7e67a89547cb98f2a0d6f9bd160513701f0d7975d87ff86
|
3 |
size 1711848436
|
checkpoint-7000/optimizer.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 3424043887
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a08ab3cc21222d06d6be5ae1d578f753ed0e8e977db01279b32cbb2deed24e71
|
3 |
size 3424043887
|
checkpoint-7000/trainer_state.json
CHANGED
@@ -1,6 +1,6 @@
|
|
1 |
{
|
2 |
-
"best_metric": 0.
|
3 |
-
"best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e4l57-l/checkpoint-
|
4 |
"epoch": 0.7465870307167235,
|
5 |
"eval_steps": 500,
|
6 |
"global_step": 7000,
|
@@ -10,212 +10,212 @@
|
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 0.05332764505119454,
|
13 |
-
"grad_norm":
|
14 |
"learning_rate": 4.911120591581342e-07,
|
15 |
-
"loss": 0.
|
16 |
"step": 500
|
17 |
},
|
18 |
{
|
19 |
"epoch": 0.05332764505119454,
|
20 |
-
"eval_loss": 1.
|
21 |
-
"eval_runtime":
|
22 |
-
"eval_samples_per_second": 15.
|
23 |
-
"eval_steps_per_second": 1.
|
24 |
"step": 500
|
25 |
},
|
26 |
{
|
27 |
"epoch": 0.10665529010238908,
|
28 |
-
"grad_norm":
|
29 |
"learning_rate": 4.822241183162685e-07,
|
30 |
-
"loss": 0.
|
31 |
"step": 1000
|
32 |
},
|
33 |
{
|
34 |
"epoch": 0.10665529010238908,
|
35 |
-
"eval_loss": 0.
|
36 |
-
"eval_runtime":
|
37 |
-
"eval_samples_per_second": 15.
|
38 |
-
"eval_steps_per_second": 1.
|
39 |
"step": 1000
|
40 |
},
|
41 |
{
|
42 |
"epoch": 0.1599829351535836,
|
43 |
-
"grad_norm":
|
44 |
"learning_rate": 4.733361774744027e-07,
|
45 |
-
"loss": 0.
|
46 |
"step": 1500
|
47 |
},
|
48 |
{
|
49 |
"epoch": 0.1599829351535836,
|
50 |
-
"eval_loss": 0.
|
51 |
-
"eval_runtime":
|
52 |
-
"eval_samples_per_second": 15.
|
53 |
-
"eval_steps_per_second": 1.
|
54 |
"step": 1500
|
55 |
},
|
56 |
{
|
57 |
"epoch": 0.21331058020477817,
|
58 |
-
"grad_norm": 0.
|
59 |
"learning_rate": 4.6444823663253695e-07,
|
60 |
-
"loss": 0.
|
61 |
"step": 2000
|
62 |
},
|
63 |
{
|
64 |
"epoch": 0.21331058020477817,
|
65 |
-
"eval_loss": 0.
|
66 |
-
"eval_runtime": 63.
|
67 |
-
"eval_samples_per_second": 15.
|
68 |
-
"eval_steps_per_second": 1.
|
69 |
"step": 2000
|
70 |
},
|
71 |
{
|
72 |
"epoch": 0.2666382252559727,
|
73 |
-
"grad_norm":
|
74 |
"learning_rate": 4.5556029579067116e-07,
|
75 |
-
"loss": 0.
|
76 |
"step": 2500
|
77 |
},
|
78 |
{
|
79 |
"epoch": 0.2666382252559727,
|
80 |
-
"eval_loss": 0.
|
81 |
-
"eval_runtime":
|
82 |
-
"eval_samples_per_second": 15.
|
83 |
-
"eval_steps_per_second": 1.
|
84 |
"step": 2500
|
85 |
},
|
86 |
{
|
87 |
"epoch": 0.3199658703071672,
|
88 |
-
"grad_norm": 9.
|
89 |
"learning_rate": 4.4667235494880547e-07,
|
90 |
-
"loss": 0.
|
91 |
"step": 3000
|
92 |
},
|
93 |
{
|
94 |
"epoch": 0.3199658703071672,
|
95 |
-
"eval_loss": 0.
|
96 |
-
"eval_runtime":
|
97 |
-
"eval_samples_per_second": 15.
|
98 |
-
"eval_steps_per_second": 1.
|
99 |
"step": 3000
|
100 |
},
|
101 |
{
|
102 |
"epoch": 0.37329351535836175,
|
103 |
-
"grad_norm": 0.
|
104 |
"learning_rate": 4.377844141069397e-07,
|
105 |
-
"loss": 0.
|
106 |
"step": 3500
|
107 |
},
|
108 |
{
|
109 |
"epoch": 0.37329351535836175,
|
110 |
-
"eval_loss": 0.
|
111 |
-
"eval_runtime":
|
112 |
-
"eval_samples_per_second": 15.
|
113 |
-
"eval_steps_per_second": 1.
|
114 |
"step": 3500
|
115 |
},
|
116 |
{
|
117 |
"epoch": 0.42662116040955633,
|
118 |
-
"grad_norm":
|
119 |
"learning_rate": 4.2889647326507393e-07,
|
120 |
-
"loss": 0.
|
121 |
"step": 4000
|
122 |
},
|
123 |
{
|
124 |
"epoch": 0.42662116040955633,
|
125 |
-
"eval_loss": 0.
|
126 |
-
"eval_runtime":
|
127 |
-
"eval_samples_per_second": 15.
|
128 |
-
"eval_steps_per_second": 1.
|
129 |
"step": 4000
|
130 |
},
|
131 |
{
|
132 |
"epoch": 0.47994880546075086,
|
133 |
-
"grad_norm":
|
134 |
"learning_rate": 4.2000853242320814e-07,
|
135 |
-
"loss": 0.
|
136 |
"step": 4500
|
137 |
},
|
138 |
{
|
139 |
"epoch": 0.47994880546075086,
|
140 |
-
"eval_loss": 0.
|
141 |
-
"eval_runtime":
|
142 |
-
"eval_samples_per_second": 15.
|
143 |
-
"eval_steps_per_second": 1.
|
144 |
"step": 4500
|
145 |
},
|
146 |
{
|
147 |
"epoch": 0.5332764505119454,
|
148 |
-
"grad_norm":
|
149 |
"learning_rate": 4.1112059158134245e-07,
|
150 |
-
"loss": 0.
|
151 |
"step": 5000
|
152 |
},
|
153 |
{
|
154 |
"epoch": 0.5332764505119454,
|
155 |
-
"eval_loss": 0.
|
156 |
-
"eval_runtime":
|
157 |
-
"eval_samples_per_second": 15.
|
158 |
-
"eval_steps_per_second": 1.
|
159 |
"step": 5000
|
160 |
},
|
161 |
{
|
162 |
"epoch": 0.58660409556314,
|
163 |
-
"grad_norm":
|
164 |
"learning_rate": 4.0223265073947665e-07,
|
165 |
-
"loss": 0.
|
166 |
"step": 5500
|
167 |
},
|
168 |
{
|
169 |
"epoch": 0.58660409556314,
|
170 |
-
"eval_loss": 0.
|
171 |
-
"eval_runtime":
|
172 |
-
"eval_samples_per_second": 15.
|
173 |
-
"eval_steps_per_second": 1.
|
174 |
"step": 5500
|
175 |
},
|
176 |
{
|
177 |
"epoch": 0.6399317406143344,
|
178 |
-
"grad_norm":
|
179 |
"learning_rate": 3.933447098976109e-07,
|
180 |
-
"loss": 0.
|
181 |
"step": 6000
|
182 |
},
|
183 |
{
|
184 |
"epoch": 0.6399317406143344,
|
185 |
-
"eval_loss": 0.
|
186 |
-
"eval_runtime": 64.
|
187 |
-
"eval_samples_per_second": 15.
|
188 |
-
"eval_steps_per_second": 1.
|
189 |
"step": 6000
|
190 |
},
|
191 |
{
|
192 |
"epoch": 0.693259385665529,
|
193 |
-
"grad_norm":
|
194 |
"learning_rate": 3.8445676905574517e-07,
|
195 |
-
"loss": 0.
|
196 |
"step": 6500
|
197 |
},
|
198 |
{
|
199 |
"epoch": 0.693259385665529,
|
200 |
-
"eval_loss": 0.
|
201 |
-
"eval_runtime":
|
202 |
-
"eval_samples_per_second": 15.
|
203 |
-
"eval_steps_per_second": 1.
|
204 |
"step": 6500
|
205 |
},
|
206 |
{
|
207 |
"epoch": 0.7465870307167235,
|
208 |
-
"grad_norm":
|
209 |
"learning_rate": 3.755688282138794e-07,
|
210 |
-
"loss": 0.
|
211 |
"step": 7000
|
212 |
},
|
213 |
{
|
214 |
"epoch": 0.7465870307167235,
|
215 |
-
"eval_loss": 0.
|
216 |
-
"eval_runtime": 62.
|
217 |
-
"eval_samples_per_second": 15.
|
218 |
-
"eval_steps_per_second": 1.
|
219 |
"step": 7000
|
220 |
}
|
221 |
],
|
|
|
1 |
{
|
2 |
+
"best_metric": 0.6227805614471436,
|
3 |
+
"best_model_checkpoint": "./output/clip-finetuned-csu-p14-336-e4l57-l/checkpoint-6500",
|
4 |
"epoch": 0.7465870307167235,
|
5 |
"eval_steps": 500,
|
6 |
"global_step": 7000,
|
|
|
10 |
"log_history": [
|
11 |
{
|
12 |
"epoch": 0.05332764505119454,
|
13 |
+
"grad_norm": 401.61187744140625,
|
14 |
"learning_rate": 4.911120591581342e-07,
|
15 |
+
"loss": 0.3808,
|
16 |
"step": 500
|
17 |
},
|
18 |
{
|
19 |
"epoch": 0.05332764505119454,
|
20 |
+
"eval_loss": 1.1107642650604248,
|
21 |
+
"eval_runtime": 62.4169,
|
22 |
+
"eval_samples_per_second": 15.813,
|
23 |
+
"eval_steps_per_second": 1.987,
|
24 |
"step": 500
|
25 |
},
|
26 |
{
|
27 |
"epoch": 0.10665529010238908,
|
28 |
+
"grad_norm": 42.57551193237305,
|
29 |
"learning_rate": 4.822241183162685e-07,
|
30 |
+
"loss": 0.2666,
|
31 |
"step": 1000
|
32 |
},
|
33 |
{
|
34 |
"epoch": 0.10665529010238908,
|
35 |
+
"eval_loss": 0.9644754528999329,
|
36 |
+
"eval_runtime": 63.4304,
|
37 |
+
"eval_samples_per_second": 15.56,
|
38 |
+
"eval_steps_per_second": 1.955,
|
39 |
"step": 1000
|
40 |
},
|
41 |
{
|
42 |
"epoch": 0.1599829351535836,
|
43 |
+
"grad_norm": 518.2286376953125,
|
44 |
"learning_rate": 4.733361774744027e-07,
|
45 |
+
"loss": 0.2085,
|
46 |
"step": 1500
|
47 |
},
|
48 |
{
|
49 |
"epoch": 0.1599829351535836,
|
50 |
+
"eval_loss": 0.9051361680030823,
|
51 |
+
"eval_runtime": 63.9009,
|
52 |
+
"eval_samples_per_second": 15.446,
|
53 |
+
"eval_steps_per_second": 1.941,
|
54 |
"step": 1500
|
55 |
},
|
56 |
{
|
57 |
"epoch": 0.21331058020477817,
|
58 |
+
"grad_norm": 0.0223398394882679,
|
59 |
"learning_rate": 4.6444823663253695e-07,
|
60 |
+
"loss": 0.1867,
|
61 |
"step": 2000
|
62 |
},
|
63 |
{
|
64 |
"epoch": 0.21331058020477817,
|
65 |
+
"eval_loss": 0.8611643314361572,
|
66 |
+
"eval_runtime": 63.9649,
|
67 |
+
"eval_samples_per_second": 15.43,
|
68 |
+
"eval_steps_per_second": 1.939,
|
69 |
"step": 2000
|
70 |
},
|
71 |
{
|
72 |
"epoch": 0.2666382252559727,
|
73 |
+
"grad_norm": 175.0380096435547,
|
74 |
"learning_rate": 4.5556029579067116e-07,
|
75 |
+
"loss": 0.2068,
|
76 |
"step": 2500
|
77 |
},
|
78 |
{
|
79 |
"epoch": 0.2666382252559727,
|
80 |
+
"eval_loss": 0.7947371006011963,
|
81 |
+
"eval_runtime": 64.0455,
|
82 |
+
"eval_samples_per_second": 15.411,
|
83 |
+
"eval_steps_per_second": 1.936,
|
84 |
"step": 2500
|
85 |
},
|
86 |
{
|
87 |
"epoch": 0.3199658703071672,
|
88 |
+
"grad_norm": 9.192651084788395e-09,
|
89 |
"learning_rate": 4.4667235494880547e-07,
|
90 |
+
"loss": 0.163,
|
91 |
"step": 3000
|
92 |
},
|
93 |
{
|
94 |
"epoch": 0.3199658703071672,
|
95 |
+
"eval_loss": 0.7677457332611084,
|
96 |
+
"eval_runtime": 64.1593,
|
97 |
+
"eval_samples_per_second": 15.384,
|
98 |
+
"eval_steps_per_second": 1.933,
|
99 |
"step": 3000
|
100 |
},
|
101 |
{
|
102 |
"epoch": 0.37329351535836175,
|
103 |
+
"grad_norm": 0.002516463166102767,
|
104 |
"learning_rate": 4.377844141069397e-07,
|
105 |
+
"loss": 0.1729,
|
106 |
"step": 3500
|
107 |
},
|
108 |
{
|
109 |
"epoch": 0.37329351535836175,
|
110 |
+
"eval_loss": 0.7635349035263062,
|
111 |
+
"eval_runtime": 62.8257,
|
112 |
+
"eval_samples_per_second": 15.71,
|
113 |
+
"eval_steps_per_second": 1.974,
|
114 |
"step": 3500
|
115 |
},
|
116 |
{
|
117 |
"epoch": 0.42662116040955633,
|
118 |
+
"grad_norm": 7.581352710723877,
|
119 |
"learning_rate": 4.2889647326507393e-07,
|
120 |
+
"loss": 0.1149,
|
121 |
"step": 4000
|
122 |
},
|
123 |
{
|
124 |
"epoch": 0.42662116040955633,
|
125 |
+
"eval_loss": 0.7458549737930298,
|
126 |
+
"eval_runtime": 64.2184,
|
127 |
+
"eval_samples_per_second": 15.369,
|
128 |
+
"eval_steps_per_second": 1.931,
|
129 |
"step": 4000
|
130 |
},
|
131 |
{
|
132 |
"epoch": 0.47994880546075086,
|
133 |
+
"grad_norm": 1.3885610103607178,
|
134 |
"learning_rate": 4.2000853242320814e-07,
|
135 |
+
"loss": 0.1264,
|
136 |
"step": 4500
|
137 |
},
|
138 |
{
|
139 |
"epoch": 0.47994880546075086,
|
140 |
+
"eval_loss": 0.7185410261154175,
|
141 |
+
"eval_runtime": 62.6998,
|
142 |
+
"eval_samples_per_second": 15.742,
|
143 |
+
"eval_steps_per_second": 1.978,
|
144 |
"step": 4500
|
145 |
},
|
146 |
{
|
147 |
"epoch": 0.5332764505119454,
|
148 |
+
"grad_norm": 4.22987541242037e-05,
|
149 |
"learning_rate": 4.1112059158134245e-07,
|
150 |
+
"loss": 0.1638,
|
151 |
"step": 5000
|
152 |
},
|
153 |
{
|
154 |
"epoch": 0.5332764505119454,
|
155 |
+
"eval_loss": 0.7118513584136963,
|
156 |
+
"eval_runtime": 63.9674,
|
157 |
+
"eval_samples_per_second": 15.43,
|
158 |
+
"eval_steps_per_second": 1.938,
|
159 |
"step": 5000
|
160 |
},
|
161 |
{
|
162 |
"epoch": 0.58660409556314,
|
163 |
+
"grad_norm": 3.7000099837314337e-06,
|
164 |
"learning_rate": 4.0223265073947665e-07,
|
165 |
+
"loss": 0.1287,
|
166 |
"step": 5500
|
167 |
},
|
168 |
{
|
169 |
"epoch": 0.58660409556314,
|
170 |
+
"eval_loss": 0.6782647967338562,
|
171 |
+
"eval_runtime": 62.7899,
|
172 |
+
"eval_samples_per_second": 15.719,
|
173 |
+
"eval_steps_per_second": 1.975,
|
174 |
"step": 5500
|
175 |
},
|
176 |
{
|
177 |
"epoch": 0.6399317406143344,
|
178 |
+
"grad_norm": 43.72269058227539,
|
179 |
"learning_rate": 3.933447098976109e-07,
|
180 |
+
"loss": 0.1324,
|
181 |
"step": 6000
|
182 |
},
|
183 |
{
|
184 |
"epoch": 0.6399317406143344,
|
185 |
+
"eval_loss": 0.6554597020149231,
|
186 |
+
"eval_runtime": 64.1964,
|
187 |
+
"eval_samples_per_second": 15.375,
|
188 |
+
"eval_steps_per_second": 1.932,
|
189 |
"step": 6000
|
190 |
},
|
191 |
{
|
192 |
"epoch": 0.693259385665529,
|
193 |
+
"grad_norm": 9.483234316576272e-05,
|
194 |
"learning_rate": 3.8445676905574517e-07,
|
195 |
+
"loss": 0.1051,
|
196 |
"step": 6500
|
197 |
},
|
198 |
{
|
199 |
"epoch": 0.693259385665529,
|
200 |
+
"eval_loss": 0.6227805614471436,
|
201 |
+
"eval_runtime": 63.0075,
|
202 |
+
"eval_samples_per_second": 15.665,
|
203 |
+
"eval_steps_per_second": 1.968,
|
204 |
"step": 6500
|
205 |
},
|
206 |
{
|
207 |
"epoch": 0.7465870307167235,
|
208 |
+
"grad_norm": 9.806918797039543e-07,
|
209 |
"learning_rate": 3.755688282138794e-07,
|
210 |
+
"loss": 0.1486,
|
211 |
"step": 7000
|
212 |
},
|
213 |
{
|
214 |
"epoch": 0.7465870307167235,
|
215 |
+
"eval_loss": 0.6244157552719116,
|
216 |
+
"eval_runtime": 62.7149,
|
217 |
+
"eval_samples_per_second": 15.738,
|
218 |
+
"eval_steps_per_second": 1.977,
|
219 |
"step": 7000
|
220 |
}
|
221 |
],
|
checkpoint-7000/training_args.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 4847
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:54c5f1c69a0ead76c168a8c74b80758f8f8eb77228d72de2bfc13a6b8cdc784b
|
3 |
size 4847
|