vuongnhathien
commited on
Commit
•
4539311
1
Parent(s):
7692d15
End of training
Browse files- README.md +3 -3
- all_results.json +8 -0
- runs/May24_14-22-57_bad680ddfb8e/events.out.tfevents.1716567266.bad680ddfb8e.23.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +309 -0
README.md
CHANGED
@@ -22,7 +22,7 @@ model-index:
|
|
22 |
metrics:
|
23 |
- name: Accuracy
|
24 |
type: accuracy
|
25 |
-
value: 0.
|
26 |
---
|
27 |
|
28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
@@ -32,8 +32,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
32 |
|
33 |
This model is a fine-tuned version of [facebook/convnextv2-nano-22k-384](https://huggingface.co/facebook/convnextv2-nano-22k-384) on the imagefolder dataset.
|
34 |
It achieves the following results on the evaluation set:
|
35 |
-
- Loss: 0.
|
36 |
-
- Accuracy: 0.
|
37 |
|
38 |
## Model description
|
39 |
|
|
|
22 |
metrics:
|
23 |
- name: Accuracy
|
24 |
type: accuracy
|
25 |
+
value: 0.9279761904761905
|
26 |
---
|
27 |
|
28 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
32 |
|
33 |
This model is a fine-tuned version of [facebook/convnextv2-nano-22k-384](https://huggingface.co/facebook/convnextv2-nano-22k-384) on the imagefolder dataset.
|
34 |
It achieves the following results on the evaluation set:
|
35 |
+
- Loss: 0.2872
|
36 |
+
- Accuracy: 0.9280
|
37 |
|
38 |
## Model description
|
39 |
|
all_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 10.0,
|
3 |
+
"total_flos": 7.000491898906214e+18,
|
4 |
+
"train_loss": 0.35536065743186257,
|
5 |
+
"train_runtime": 6476.8205,
|
6 |
+
"train_samples_per_second": 27.144,
|
7 |
+
"train_steps_per_second": 0.425
|
8 |
+
}
|
runs/May24_14-22-57_bad680ddfb8e/events.out.tfevents.1716567266.bad680ddfb8e.23.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:103d21656fbfb23c9e46982de13aed9cb0155fb7d9eb8594b9cf7489d0cb88a4
|
3 |
+
size 411
|
train_results.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 10.0,
|
3 |
+
"total_flos": 7.000491898906214e+18,
|
4 |
+
"train_loss": 0.35536065743186257,
|
5 |
+
"train_runtime": 6476.8205,
|
6 |
+
"train_samples_per_second": 27.144,
|
7 |
+
"train_steps_per_second": 0.425
|
8 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,309 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.29478973150253296,
|
3 |
+
"best_model_checkpoint": "./convnext-nano-3e-4-augment/checkpoint-2750",
|
4 |
+
"epoch": 10.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 2750,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.36,
|
13 |
+
"grad_norm": 17.436264038085938,
|
14 |
+
"learning_rate": 0.00029902226030228247,
|
15 |
+
"loss": 1.4532,
|
16 |
+
"step": 100
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.73,
|
20 |
+
"grad_norm": 8.700443267822266,
|
21 |
+
"learning_rate": 0.00029610178754135,
|
22 |
+
"loss": 0.8119,
|
23 |
+
"step": 200
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 1.0,
|
27 |
+
"eval_accuracy": 0.827037773359841,
|
28 |
+
"eval_loss": 0.5636539459228516,
|
29 |
+
"eval_runtime": 61.1609,
|
30 |
+
"eval_samples_per_second": 41.121,
|
31 |
+
"eval_steps_per_second": 0.654,
|
32 |
+
"step": 275
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 1.09,
|
36 |
+
"grad_norm": 19.5541934967041,
|
37 |
+
"learning_rate": 0.0002912766545459287,
|
38 |
+
"loss": 0.7009,
|
39 |
+
"step": 300
|
40 |
+
},
|
41 |
+
{
|
42 |
+
"epoch": 1.45,
|
43 |
+
"grad_norm": 5.992691516876221,
|
44 |
+
"learning_rate": 0.0002846097643037037,
|
45 |
+
"loss": 0.5634,
|
46 |
+
"step": 400
|
47 |
+
},
|
48 |
+
{
|
49 |
+
"epoch": 1.82,
|
50 |
+
"grad_norm": 10.371554374694824,
|
51 |
+
"learning_rate": 0.0002761880299246772,
|
52 |
+
"loss": 0.5829,
|
53 |
+
"step": 500
|
54 |
+
},
|
55 |
+
{
|
56 |
+
"epoch": 2.0,
|
57 |
+
"eval_accuracy": 0.8497017892644135,
|
58 |
+
"eval_loss": 0.5015920400619507,
|
59 |
+
"eval_runtime": 60.5306,
|
60 |
+
"eval_samples_per_second": 41.549,
|
61 |
+
"eval_steps_per_second": 0.661,
|
62 |
+
"step": 550
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 2.18,
|
66 |
+
"grad_norm": 6.42030143737793,
|
67 |
+
"learning_rate": 0.00026612124159586237,
|
68 |
+
"loss": 0.5118,
|
69 |
+
"step": 600
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 2.55,
|
73 |
+
"grad_norm": 5.376116752624512,
|
74 |
+
"learning_rate": 0.00025454063529829405,
|
75 |
+
"loss": 0.4771,
|
76 |
+
"step": 700
|
77 |
+
},
|
78 |
+
{
|
79 |
+
"epoch": 2.91,
|
80 |
+
"grad_norm": 9.462407112121582,
|
81 |
+
"learning_rate": 0.00024159718194531572,
|
82 |
+
"loss": 0.4623,
|
83 |
+
"step": 800
|
84 |
+
},
|
85 |
+
{
|
86 |
+
"epoch": 3.0,
|
87 |
+
"eval_accuracy": 0.8755467196819086,
|
88 |
+
"eval_loss": 0.44941774010658264,
|
89 |
+
"eval_runtime": 60.9271,
|
90 |
+
"eval_samples_per_second": 41.279,
|
91 |
+
"eval_steps_per_second": 0.657,
|
92 |
+
"step": 825
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 3.27,
|
96 |
+
"grad_norm": 6.973038196563721,
|
97 |
+
"learning_rate": 0.00022745961924584428,
|
98 |
+
"loss": 0.3929,
|
99 |
+
"step": 900
|
100 |
+
},
|
101 |
+
{
|
102 |
+
"epoch": 3.64,
|
103 |
+
"grad_norm": 6.476861953735352,
|
104 |
+
"learning_rate": 0.00021231225195028297,
|
105 |
+
"loss": 0.3711,
|
106 |
+
"step": 1000
|
107 |
+
},
|
108 |
+
{
|
109 |
+
"epoch": 4.0,
|
110 |
+
"grad_norm": 6.22141170501709,
|
111 |
+
"learning_rate": 0.0001963525491562421,
|
112 |
+
"loss": 0.359,
|
113 |
+
"step": 1100
|
114 |
+
},
|
115 |
+
{
|
116 |
+
"epoch": 4.0,
|
117 |
+
"eval_accuracy": 0.8886679920477137,
|
118 |
+
"eval_loss": 0.3809404671192169,
|
119 |
+
"eval_runtime": 61.1213,
|
120 |
+
"eval_samples_per_second": 41.148,
|
121 |
+
"eval_steps_per_second": 0.654,
|
122 |
+
"step": 1100
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 4.36,
|
126 |
+
"grad_norm": 7.65138578414917,
|
127 |
+
"learning_rate": 0.0001797885699968618,
|
128 |
+
"loss": 0.3315,
|
129 |
+
"step": 1200
|
130 |
+
},
|
131 |
+
{
|
132 |
+
"epoch": 4.73,
|
133 |
+
"grad_norm": 4.203142166137695,
|
134 |
+
"learning_rate": 0.00016283625127182596,
|
135 |
+
"loss": 0.2881,
|
136 |
+
"step": 1300
|
137 |
+
},
|
138 |
+
{
|
139 |
+
"epoch": 5.0,
|
140 |
+
"eval_accuracy": 0.8998011928429424,
|
141 |
+
"eval_loss": 0.3741509020328522,
|
142 |
+
"eval_runtime": 61.4609,
|
143 |
+
"eval_samples_per_second": 40.92,
|
144 |
+
"eval_steps_per_second": 0.651,
|
145 |
+
"step": 1375
|
146 |
+
},
|
147 |
+
{
|
148 |
+
"epoch": 5.09,
|
149 |
+
"grad_norm": 6.004030227661133,
|
150 |
+
"learning_rate": 0.00014571659238094556,
|
151 |
+
"loss": 0.286,
|
152 |
+
"step": 1400
|
153 |
+
},
|
154 |
+
{
|
155 |
+
"epoch": 5.45,
|
156 |
+
"grad_norm": 6.018474578857422,
|
157 |
+
"learning_rate": 0.00012865277425900724,
|
158 |
+
"loss": 0.2292,
|
159 |
+
"step": 1500
|
160 |
+
},
|
161 |
+
{
|
162 |
+
"epoch": 5.82,
|
163 |
+
"grad_norm": 4.614959716796875,
|
164 |
+
"learning_rate": 0.00011186724987097698,
|
165 |
+
"loss": 0.2302,
|
166 |
+
"step": 1600
|
167 |
+
},
|
168 |
+
{
|
169 |
+
"epoch": 6.0,
|
170 |
+
"eval_accuracy": 0.9113320079522863,
|
171 |
+
"eval_loss": 0.3401913046836853,
|
172 |
+
"eval_runtime": 61.1216,
|
173 |
+
"eval_samples_per_second": 41.147,
|
174 |
+
"eval_steps_per_second": 0.654,
|
175 |
+
"step": 1650
|
176 |
+
},
|
177 |
+
{
|
178 |
+
"epoch": 6.18,
|
179 |
+
"grad_norm": 4.227890968322754,
|
180 |
+
"learning_rate": 9.557884419740386e-05,
|
181 |
+
"loss": 0.2225,
|
182 |
+
"step": 1700
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 6.55,
|
186 |
+
"grad_norm": 4.184410572052002,
|
187 |
+
"learning_rate": 7.999990151614894e-05,
|
188 |
+
"loss": 0.1851,
|
189 |
+
"step": 1800
|
190 |
+
},
|
191 |
+
{
|
192 |
+
"epoch": 6.91,
|
193 |
+
"grad_norm": 6.824435234069824,
|
194 |
+
"learning_rate": 6.533351716998465e-05,
|
195 |
+
"loss": 0.1827,
|
196 |
+
"step": 1900
|
197 |
+
},
|
198 |
+
{
|
199 |
+
"epoch": 7.0,
|
200 |
+
"eval_accuracy": 0.9121272365805169,
|
201 |
+
"eval_loss": 0.31500810384750366,
|
202 |
+
"eval_runtime": 60.793,
|
203 |
+
"eval_samples_per_second": 41.37,
|
204 |
+
"eval_steps_per_second": 0.658,
|
205 |
+
"step": 1925
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 7.27,
|
209 |
+
"grad_norm": 3.374882221221924,
|
210 |
+
"learning_rate": 5.1770889908207245e-05,
|
211 |
+
"loss": 0.1551,
|
212 |
+
"step": 2000
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 7.64,
|
216 |
+
"grad_norm": 4.39635705947876,
|
217 |
+
"learning_rate": 3.948882931853924e-05,
|
218 |
+
"loss": 0.1564,
|
219 |
+
"step": 2100
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 8.0,
|
223 |
+
"grad_norm": 2.6348471641540527,
|
224 |
+
"learning_rate": 2.8647450843757897e-05,
|
225 |
+
"loss": 0.1466,
|
226 |
+
"step": 2200
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 8.0,
|
230 |
+
"eval_accuracy": 0.9228628230616303,
|
231 |
+
"eval_loss": 0.30119746923446655,
|
232 |
+
"eval_runtime": 61.3308,
|
233 |
+
"eval_samples_per_second": 41.007,
|
234 |
+
"eval_steps_per_second": 0.652,
|
235 |
+
"step": 2200
|
236 |
+
},
|
237 |
+
{
|
238 |
+
"epoch": 8.36,
|
239 |
+
"grad_norm": 5.662447929382324,
|
240 |
+
"learning_rate": 1.9388088432033443e-05,
|
241 |
+
"loss": 0.1236,
|
242 |
+
"step": 2300
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 8.73,
|
246 |
+
"grad_norm": 0.9912707209587097,
|
247 |
+
"learning_rate": 1.1831452032772498e-05,
|
248 |
+
"loss": 0.1223,
|
249 |
+
"step": 2400
|
250 |
+
},
|
251 |
+
{
|
252 |
+
"epoch": 9.0,
|
253 |
+
"eval_accuracy": 0.9248508946322067,
|
254 |
+
"eval_loss": 0.2995615005493164,
|
255 |
+
"eval_runtime": 60.6772,
|
256 |
+
"eval_samples_per_second": 41.449,
|
257 |
+
"eval_steps_per_second": 0.659,
|
258 |
+
"step": 2475
|
259 |
+
},
|
260 |
+
{
|
261 |
+
"epoch": 9.09,
|
262 |
+
"grad_norm": 2.5069386959075928,
|
263 |
+
"learning_rate": 6.076053957825411e-06,
|
264 |
+
"loss": 0.1283,
|
265 |
+
"step": 2500
|
266 |
+
},
|
267 |
+
{
|
268 |
+
"epoch": 9.45,
|
269 |
+
"grad_norm": 7.067609786987305,
|
270 |
+
"learning_rate": 2.1969246228460523e-06,
|
271 |
+
"loss": 0.1142,
|
272 |
+
"step": 2600
|
273 |
+
},
|
274 |
+
{
|
275 |
+
"epoch": 9.82,
|
276 |
+
"grad_norm": 4.694984436035156,
|
277 |
+
"learning_rate": 2.4463441107965276e-07,
|
278 |
+
"loss": 0.1332,
|
279 |
+
"step": 2700
|
280 |
+
},
|
281 |
+
{
|
282 |
+
"epoch": 10.0,
|
283 |
+
"eval_accuracy": 0.9248508946322067,
|
284 |
+
"eval_loss": 0.29478973150253296,
|
285 |
+
"eval_runtime": 61.4481,
|
286 |
+
"eval_samples_per_second": 40.929,
|
287 |
+
"eval_steps_per_second": 0.651,
|
288 |
+
"step": 2750
|
289 |
+
},
|
290 |
+
{
|
291 |
+
"epoch": 10.0,
|
292 |
+
"step": 2750,
|
293 |
+
"total_flos": 7.000491898906214e+18,
|
294 |
+
"train_loss": 0.35536065743186257,
|
295 |
+
"train_runtime": 6476.8205,
|
296 |
+
"train_samples_per_second": 27.144,
|
297 |
+
"train_steps_per_second": 0.425
|
298 |
+
}
|
299 |
+
],
|
300 |
+
"logging_steps": 100,
|
301 |
+
"max_steps": 2750,
|
302 |
+
"num_input_tokens_seen": 0,
|
303 |
+
"num_train_epochs": 10,
|
304 |
+
"save_steps": 500,
|
305 |
+
"total_flos": 7.000491898906214e+18,
|
306 |
+
"train_batch_size": 64,
|
307 |
+
"trial_name": null,
|
308 |
+
"trial_params": null
|
309 |
+
}
|