michaelgathara commited on
Commit
36a2c49
·
verified ·
1 Parent(s): a711449

Upload folder using huggingface_hub

Browse files
README.md ADDED
@@ -0,0 +1,47 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ tags:
3
+ - image-classification
4
+ - pytorch
5
+ - huggingface
6
+ - vit
7
+ - emotion-recognition
8
+ datasets:
9
+ - raf-db
10
+ base_model: trpakov/vit-face-expression
11
+ library_name: transformers
12
+ ---
13
+
14
+ # ViT Face Expression (Fine-tuned on RAF-DB)
15
+
16
+ This model is a fine-tuned version of [trpakov/vit-face-expression](https://huggingface.co/trpakov/vit-face-expression) on the [RAF-DB (Real-world Affective Faces Database)](http://www.whdeng.cn/raf/model1.html).
17
+
18
+ ## Model Description
19
+ - **Architecture**: Vision Transformer (ViT)
20
+ - **Task**: Facial Emotion Recognition
21
+ - **Emotions**: Anger, Disgust, Fear, Happiness, Neutral, Sadness, Surprise
22
+
23
+ ## Dataset
24
+ RAF-DB is a large-scale facial expression database with around 30,000 facial images downloaded from the Internet. The images cover 7 basic emotions and 12 compound emotions. This model was fine-tuned on the 7 basic emotions subset.
25
+
26
+ ## Usage
27
+
28
+ ```python
29
+ from transformers import ViTImageProcessor, ViTForImageClassification
30
+ from PIL import Image
31
+ import requests
32
+
33
+ url = 'http://images.cocodataset.org/val2017/000000039769.jpg'
34
+ image = Image.open(requests.get(url, stream=True).raw)
35
+
36
+ repo_name = "michaelgathara/vit-face-raf-db"
37
+
38
+ processor = ViTImageProcessor.from_pretrained(repo_name)
39
+ model = ViTForImageClassification.from_pretrained(repo_name)
40
+
41
+ inputs = processor(images=image, return_tensors="pt")
42
+ outputs = model(**inputs)
43
+ logits = outputs.logits
44
+ # model predicts one of the 7 emotions
45
+ predicted_class_idx = logits.argmax(-1).item()
46
+ print("Predicted class:", model.config.id2label[predicted_class_idx])
47
+ ```
checkpoint-2047/config.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "ViTForImageClassification"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.0,
6
+ "dtype": "float32",
7
+ "encoder_stride": 16,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.0,
10
+ "hidden_size": 768,
11
+ "id2label": {
12
+ "0": "anger",
13
+ "1": "disgust",
14
+ "2": "fear",
15
+ "3": "happiness",
16
+ "4": "neutral",
17
+ "5": "sadness",
18
+ "6": "surprise"
19
+ },
20
+ "image_size": 224,
21
+ "initializer_range": 0.02,
22
+ "intermediate_size": 3072,
23
+ "label2id": {
24
+ "anger": "0",
25
+ "disgust": "1",
26
+ "fear": "2",
27
+ "happiness": "3",
28
+ "neutral": "4",
29
+ "sadness": "5",
30
+ "surprise": "6"
31
+ },
32
+ "layer_norm_eps": 1e-12,
33
+ "model_type": "vit",
34
+ "num_attention_heads": 12,
35
+ "num_channels": 3,
36
+ "num_hidden_layers": 12,
37
+ "patch_size": 16,
38
+ "pooler_act": "tanh",
39
+ "pooler_output_size": 768,
40
+ "problem_type": "single_label_classification",
41
+ "qkv_bias": true,
42
+ "transformers_version": "4.57.6"
43
+ }
checkpoint-2047/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a0280a8546c3a92190fa43a0b6fb9a34ef22e936b44700f3041bc080f5a2ab0b
3
+ size 343239356
checkpoint-2047/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6a4c3498856c42d1437bb1ebf4702e8e3d3a812d1b0de1774962dca41a6ef607
3
+ size 686594443
checkpoint-2047/preprocessor_config.json ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_convert_rgb": null,
3
+ "do_normalize": true,
4
+ "do_rescale": true,
5
+ "do_resize": true,
6
+ "image_mean": [
7
+ 0.5,
8
+ 0.5,
9
+ 0.5
10
+ ],
11
+ "image_processor_type": "ViTImageProcessor",
12
+ "image_std": [
13
+ 0.5,
14
+ 0.5,
15
+ 0.5
16
+ ],
17
+ "resample": 2,
18
+ "rescale_factor": 0.00392156862745098,
19
+ "size": {
20
+ "height": 224,
21
+ "width": 224
22
+ }
23
+ }
checkpoint-2047/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ba5ab596857b081dbf5384ce205bf3ca94cda6b4037d3ec9ef7e2781cc9ae4ef
3
+ size 14455
checkpoint-2047/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d7d750f394a4b6c7f0baf15ac5be235921ab13fb84967e386d29277ec180de81
3
+ size 1465
checkpoint-2047/trainer_state.json ADDED
@@ -0,0 +1,1471 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 2047,
3
+ "best_metric": 0.8527472527472527,
4
+ "best_model_checkpoint": "models/raf_db_finetuned\\checkpoint-2047",
5
+ "epoch": 1.0,
6
+ "eval_steps": 500,
7
+ "global_step": 2047,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.004885197850512946,
14
+ "grad_norm": 17.944799423217773,
15
+ "learning_rate": 1.991206643869077e-05,
16
+ "loss": 2.0291,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 0.009770395701025891,
21
+ "grad_norm": 9.91746711730957,
22
+ "learning_rate": 1.981436248168051e-05,
23
+ "loss": 1.2123,
24
+ "step": 20
25
+ },
26
+ {
27
+ "epoch": 0.014655593551538837,
28
+ "grad_norm": 15.047444343566895,
29
+ "learning_rate": 1.971665852467025e-05,
30
+ "loss": 1.0855,
31
+ "step": 30
32
+ },
33
+ {
34
+ "epoch": 0.019540791402051783,
35
+ "grad_norm": 23.127321243286133,
36
+ "learning_rate": 1.961895456765999e-05,
37
+ "loss": 1.1072,
38
+ "step": 40
39
+ },
40
+ {
41
+ "epoch": 0.024425989252564728,
42
+ "grad_norm": 11.329324722290039,
43
+ "learning_rate": 1.9521250610649734e-05,
44
+ "loss": 0.9336,
45
+ "step": 50
46
+ },
47
+ {
48
+ "epoch": 0.029311187103077674,
49
+ "grad_norm": 5.342128753662109,
50
+ "learning_rate": 1.9423546653639473e-05,
51
+ "loss": 0.7335,
52
+ "step": 60
53
+ },
54
+ {
55
+ "epoch": 0.03419638495359062,
56
+ "grad_norm": 12.106943130493164,
57
+ "learning_rate": 1.9325842696629215e-05,
58
+ "loss": 0.5957,
59
+ "step": 70
60
+ },
61
+ {
62
+ "epoch": 0.039081582804103565,
63
+ "grad_norm": 2.4607601165771484,
64
+ "learning_rate": 1.9228138739618955e-05,
65
+ "loss": 0.5493,
66
+ "step": 80
67
+ },
68
+ {
69
+ "epoch": 0.04396678065461651,
70
+ "grad_norm": 8.041631698608398,
71
+ "learning_rate": 1.9130434782608697e-05,
72
+ "loss": 0.9075,
73
+ "step": 90
74
+ },
75
+ {
76
+ "epoch": 0.048851978505129456,
77
+ "grad_norm": 10.222543716430664,
78
+ "learning_rate": 1.9032730825598437e-05,
79
+ "loss": 0.6926,
80
+ "step": 100
81
+ },
82
+ {
83
+ "epoch": 0.0537371763556424,
84
+ "grad_norm": 10.51362133026123,
85
+ "learning_rate": 1.893502686858818e-05,
86
+ "loss": 0.8827,
87
+ "step": 110
88
+ },
89
+ {
90
+ "epoch": 0.05862237420615535,
91
+ "grad_norm": 6.351449489593506,
92
+ "learning_rate": 1.883732291157792e-05,
93
+ "loss": 0.6727,
94
+ "step": 120
95
+ },
96
+ {
97
+ "epoch": 0.0635075720566683,
98
+ "grad_norm": 13.230606079101562,
99
+ "learning_rate": 1.873961895456766e-05,
100
+ "loss": 0.7156,
101
+ "step": 130
102
+ },
103
+ {
104
+ "epoch": 0.06839276990718124,
105
+ "grad_norm": 10.970815658569336,
106
+ "learning_rate": 1.86419149975574e-05,
107
+ "loss": 0.4933,
108
+ "step": 140
109
+ },
110
+ {
111
+ "epoch": 0.07327796775769418,
112
+ "grad_norm": 0.8731197118759155,
113
+ "learning_rate": 1.8544211040547143e-05,
114
+ "loss": 0.5562,
115
+ "step": 150
116
+ },
117
+ {
118
+ "epoch": 0.07816316560820713,
119
+ "grad_norm": 9.979445457458496,
120
+ "learning_rate": 1.8446507083536886e-05,
121
+ "loss": 0.5514,
122
+ "step": 160
123
+ },
124
+ {
125
+ "epoch": 0.08304836345872008,
126
+ "grad_norm": 10.89950180053711,
127
+ "learning_rate": 1.8348803126526625e-05,
128
+ "loss": 0.6483,
129
+ "step": 170
130
+ },
131
+ {
132
+ "epoch": 0.08793356130923302,
133
+ "grad_norm": 1.3648935556411743,
134
+ "learning_rate": 1.8251099169516368e-05,
135
+ "loss": 0.609,
136
+ "step": 180
137
+ },
138
+ {
139
+ "epoch": 0.09281875915974597,
140
+ "grad_norm": 2.5604639053344727,
141
+ "learning_rate": 1.8153395212506107e-05,
142
+ "loss": 0.8282,
143
+ "step": 190
144
+ },
145
+ {
146
+ "epoch": 0.09770395701025891,
147
+ "grad_norm": 6.0546956062316895,
148
+ "learning_rate": 1.805569125549585e-05,
149
+ "loss": 0.4972,
150
+ "step": 200
151
+ },
152
+ {
153
+ "epoch": 0.10258915486077186,
154
+ "grad_norm": 9.024284362792969,
155
+ "learning_rate": 1.7957987298485592e-05,
156
+ "loss": 0.7344,
157
+ "step": 210
158
+ },
159
+ {
160
+ "epoch": 0.1074743527112848,
161
+ "grad_norm": 15.3169527053833,
162
+ "learning_rate": 1.786028334147533e-05,
163
+ "loss": 0.4587,
164
+ "step": 220
165
+ },
166
+ {
167
+ "epoch": 0.11235955056179775,
168
+ "grad_norm": 6.840464115142822,
169
+ "learning_rate": 1.7762579384465074e-05,
170
+ "loss": 0.3876,
171
+ "step": 230
172
+ },
173
+ {
174
+ "epoch": 0.1172447484123107,
175
+ "grad_norm": 4.563595294952393,
176
+ "learning_rate": 1.7664875427454813e-05,
177
+ "loss": 0.5993,
178
+ "step": 240
179
+ },
180
+ {
181
+ "epoch": 0.12212994626282364,
182
+ "grad_norm": 5.909449100494385,
183
+ "learning_rate": 1.7567171470444556e-05,
184
+ "loss": 0.6124,
185
+ "step": 250
186
+ },
187
+ {
188
+ "epoch": 0.1270151441133366,
189
+ "grad_norm": 10.69510269165039,
190
+ "learning_rate": 1.7469467513434295e-05,
191
+ "loss": 0.5355,
192
+ "step": 260
193
+ },
194
+ {
195
+ "epoch": 0.13190034196384953,
196
+ "grad_norm": 9.001906394958496,
197
+ "learning_rate": 1.7371763556424038e-05,
198
+ "loss": 0.607,
199
+ "step": 270
200
+ },
201
+ {
202
+ "epoch": 0.13678553981436248,
203
+ "grad_norm": 14.748282432556152,
204
+ "learning_rate": 1.7274059599413777e-05,
205
+ "loss": 0.4316,
206
+ "step": 280
207
+ },
208
+ {
209
+ "epoch": 0.14167073766487542,
210
+ "grad_norm": 5.700569152832031,
211
+ "learning_rate": 1.717635564240352e-05,
212
+ "loss": 0.4765,
213
+ "step": 290
214
+ },
215
+ {
216
+ "epoch": 0.14655593551538837,
217
+ "grad_norm": 4.104219913482666,
218
+ "learning_rate": 1.707865168539326e-05,
219
+ "loss": 0.5415,
220
+ "step": 300
221
+ },
222
+ {
223
+ "epoch": 0.15144113336590131,
224
+ "grad_norm": 7.291189670562744,
225
+ "learning_rate": 1.6980947728383002e-05,
226
+ "loss": 0.6046,
227
+ "step": 310
228
+ },
229
+ {
230
+ "epoch": 0.15632633121641426,
231
+ "grad_norm": 6.037685394287109,
232
+ "learning_rate": 1.688324377137274e-05,
233
+ "loss": 0.5612,
234
+ "step": 320
235
+ },
236
+ {
237
+ "epoch": 0.1612115290669272,
238
+ "grad_norm": 13.00805377960205,
239
+ "learning_rate": 1.6785539814362484e-05,
240
+ "loss": 0.5643,
241
+ "step": 330
242
+ },
243
+ {
244
+ "epoch": 0.16609672691744015,
245
+ "grad_norm": 12.02161979675293,
246
+ "learning_rate": 1.6687835857352223e-05,
247
+ "loss": 0.53,
248
+ "step": 340
249
+ },
250
+ {
251
+ "epoch": 0.1709819247679531,
252
+ "grad_norm": 9.597453117370605,
253
+ "learning_rate": 1.6590131900341966e-05,
254
+ "loss": 0.5697,
255
+ "step": 350
256
+ },
257
+ {
258
+ "epoch": 0.17586712261846604,
259
+ "grad_norm": 1.7890377044677734,
260
+ "learning_rate": 1.6492427943331705e-05,
261
+ "loss": 0.4628,
262
+ "step": 360
263
+ },
264
+ {
265
+ "epoch": 0.180752320468979,
266
+ "grad_norm": 6.448623180389404,
267
+ "learning_rate": 1.6394723986321448e-05,
268
+ "loss": 0.5923,
269
+ "step": 370
270
+ },
271
+ {
272
+ "epoch": 0.18563751831949193,
273
+ "grad_norm": 10.735410690307617,
274
+ "learning_rate": 1.6297020029311187e-05,
275
+ "loss": 0.4705,
276
+ "step": 380
277
+ },
278
+ {
279
+ "epoch": 0.19052271617000488,
280
+ "grad_norm": 6.801768779754639,
281
+ "learning_rate": 1.619931607230093e-05,
282
+ "loss": 0.466,
283
+ "step": 390
284
+ },
285
+ {
286
+ "epoch": 0.19540791402051783,
287
+ "grad_norm": 6.989867687225342,
288
+ "learning_rate": 1.610161211529067e-05,
289
+ "loss": 0.4903,
290
+ "step": 400
291
+ },
292
+ {
293
+ "epoch": 0.20029311187103077,
294
+ "grad_norm": 2.0656211376190186,
295
+ "learning_rate": 1.600390815828041e-05,
296
+ "loss": 0.3788,
297
+ "step": 410
298
+ },
299
+ {
300
+ "epoch": 0.20517830972154372,
301
+ "grad_norm": 4.312288761138916,
302
+ "learning_rate": 1.590620420127015e-05,
303
+ "loss": 0.4398,
304
+ "step": 420
305
+ },
306
+ {
307
+ "epoch": 0.21006350757205666,
308
+ "grad_norm": 11.896406173706055,
309
+ "learning_rate": 1.5808500244259893e-05,
310
+ "loss": 0.503,
311
+ "step": 430
312
+ },
313
+ {
314
+ "epoch": 0.2149487054225696,
315
+ "grad_norm": 8.159215927124023,
316
+ "learning_rate": 1.5710796287249636e-05,
317
+ "loss": 0.3731,
318
+ "step": 440
319
+ },
320
+ {
321
+ "epoch": 0.21983390327308255,
322
+ "grad_norm": 9.295239448547363,
323
+ "learning_rate": 1.5613092330239375e-05,
324
+ "loss": 0.8158,
325
+ "step": 450
326
+ },
327
+ {
328
+ "epoch": 0.2247191011235955,
329
+ "grad_norm": 10.293549537658691,
330
+ "learning_rate": 1.5515388373229118e-05,
331
+ "loss": 0.4771,
332
+ "step": 460
333
+ },
334
+ {
335
+ "epoch": 0.22960429897410845,
336
+ "grad_norm": 4.682013988494873,
337
+ "learning_rate": 1.5417684416218857e-05,
338
+ "loss": 0.5787,
339
+ "step": 470
340
+ },
341
+ {
342
+ "epoch": 0.2344894968246214,
343
+ "grad_norm": 7.708519458770752,
344
+ "learning_rate": 1.53199804592086e-05,
345
+ "loss": 0.3852,
346
+ "step": 480
347
+ },
348
+ {
349
+ "epoch": 0.23937469467513434,
350
+ "grad_norm": 8.97301959991455,
351
+ "learning_rate": 1.522227650219834e-05,
352
+ "loss": 0.4236,
353
+ "step": 490
354
+ },
355
+ {
356
+ "epoch": 0.24425989252564728,
357
+ "grad_norm": 4.338143825531006,
358
+ "learning_rate": 1.512457254518808e-05,
359
+ "loss": 0.5029,
360
+ "step": 500
361
+ },
362
+ {
363
+ "epoch": 0.24914509037616023,
364
+ "grad_norm": 6.023448467254639,
365
+ "learning_rate": 1.5026868588177823e-05,
366
+ "loss": 0.3324,
367
+ "step": 510
368
+ },
369
+ {
370
+ "epoch": 0.2540302882266732,
371
+ "grad_norm": 18.636865615844727,
372
+ "learning_rate": 1.4929164631167564e-05,
373
+ "loss": 0.5345,
374
+ "step": 520
375
+ },
376
+ {
377
+ "epoch": 0.2589154860771861,
378
+ "grad_norm": 14.49988079071045,
379
+ "learning_rate": 1.4831460674157305e-05,
380
+ "loss": 0.3685,
381
+ "step": 530
382
+ },
383
+ {
384
+ "epoch": 0.26380068392769906,
385
+ "grad_norm": 10.043220520019531,
386
+ "learning_rate": 1.4733756717147046e-05,
387
+ "loss": 0.4565,
388
+ "step": 540
389
+ },
390
+ {
391
+ "epoch": 0.268685881778212,
392
+ "grad_norm": 7.664310932159424,
393
+ "learning_rate": 1.4636052760136787e-05,
394
+ "loss": 0.6481,
395
+ "step": 550
396
+ },
397
+ {
398
+ "epoch": 0.27357107962872496,
399
+ "grad_norm": 10.388060569763184,
400
+ "learning_rate": 1.4538348803126528e-05,
401
+ "loss": 0.6793,
402
+ "step": 560
403
+ },
404
+ {
405
+ "epoch": 0.2784562774792379,
406
+ "grad_norm": 16.167070388793945,
407
+ "learning_rate": 1.444064484611627e-05,
408
+ "loss": 0.4101,
409
+ "step": 570
410
+ },
411
+ {
412
+ "epoch": 0.28334147532975085,
413
+ "grad_norm": 11.468669891357422,
414
+ "learning_rate": 1.434294088910601e-05,
415
+ "loss": 0.5556,
416
+ "step": 580
417
+ },
418
+ {
419
+ "epoch": 0.2882266731802638,
420
+ "grad_norm": 10.949549674987793,
421
+ "learning_rate": 1.4245236932095752e-05,
422
+ "loss": 0.4404,
423
+ "step": 590
424
+ },
425
+ {
426
+ "epoch": 0.29311187103077674,
427
+ "grad_norm": 5.8457489013671875,
428
+ "learning_rate": 1.4147532975085491e-05,
429
+ "loss": 0.5041,
430
+ "step": 600
431
+ },
432
+ {
433
+ "epoch": 0.2979970688812897,
434
+ "grad_norm": 7.100272178649902,
435
+ "learning_rate": 1.4049829018075234e-05,
436
+ "loss": 0.4452,
437
+ "step": 610
438
+ },
439
+ {
440
+ "epoch": 0.30288226673180263,
441
+ "grad_norm": 8.908060073852539,
442
+ "learning_rate": 1.3952125061064973e-05,
443
+ "loss": 0.4337,
444
+ "step": 620
445
+ },
446
+ {
447
+ "epoch": 0.3077674645823156,
448
+ "grad_norm": 15.235097885131836,
449
+ "learning_rate": 1.3854421104054716e-05,
450
+ "loss": 0.5576,
451
+ "step": 630
452
+ },
453
+ {
454
+ "epoch": 0.3126526624328285,
455
+ "grad_norm": 0.3120326101779938,
456
+ "learning_rate": 1.3756717147044455e-05,
457
+ "loss": 0.2936,
458
+ "step": 640
459
+ },
460
+ {
461
+ "epoch": 0.31753786028334147,
462
+ "grad_norm": 9.999382019042969,
463
+ "learning_rate": 1.3659013190034198e-05,
464
+ "loss": 0.4919,
465
+ "step": 650
466
+ },
467
+ {
468
+ "epoch": 0.3224230581338544,
469
+ "grad_norm": 5.313811302185059,
470
+ "learning_rate": 1.3561309233023939e-05,
471
+ "loss": 0.6082,
472
+ "step": 660
473
+ },
474
+ {
475
+ "epoch": 0.32730825598436736,
476
+ "grad_norm": 7.280602931976318,
477
+ "learning_rate": 1.346360527601368e-05,
478
+ "loss": 0.4541,
479
+ "step": 670
480
+ },
481
+ {
482
+ "epoch": 0.3321934538348803,
483
+ "grad_norm": 11.211206436157227,
484
+ "learning_rate": 1.336590131900342e-05,
485
+ "loss": 0.3229,
486
+ "step": 680
487
+ },
488
+ {
489
+ "epoch": 0.33707865168539325,
490
+ "grad_norm": 1.7503819465637207,
491
+ "learning_rate": 1.3268197361993162e-05,
492
+ "loss": 0.3966,
493
+ "step": 690
494
+ },
495
+ {
496
+ "epoch": 0.3419638495359062,
497
+ "grad_norm": 3.4260499477386475,
498
+ "learning_rate": 1.3170493404982903e-05,
499
+ "loss": 0.291,
500
+ "step": 700
501
+ },
502
+ {
503
+ "epoch": 0.34684904738641914,
504
+ "grad_norm": 3.466951370239258,
505
+ "learning_rate": 1.3072789447972645e-05,
506
+ "loss": 0.2094,
507
+ "step": 710
508
+ },
509
+ {
510
+ "epoch": 0.3517342452369321,
511
+ "grad_norm": 10.148577690124512,
512
+ "learning_rate": 1.2975085490962385e-05,
513
+ "loss": 0.4038,
514
+ "step": 720
515
+ },
516
+ {
517
+ "epoch": 0.35661944308744503,
518
+ "grad_norm": 9.461824417114258,
519
+ "learning_rate": 1.2877381533952127e-05,
520
+ "loss": 0.5108,
521
+ "step": 730
522
+ },
523
+ {
524
+ "epoch": 0.361504640937958,
525
+ "grad_norm": 9.039957046508789,
526
+ "learning_rate": 1.2779677576941867e-05,
527
+ "loss": 0.4701,
528
+ "step": 740
529
+ },
530
+ {
531
+ "epoch": 0.3663898387884709,
532
+ "grad_norm": 7.5800065994262695,
533
+ "learning_rate": 1.268197361993161e-05,
534
+ "loss": 0.4359,
535
+ "step": 750
536
+ },
537
+ {
538
+ "epoch": 0.37127503663898387,
539
+ "grad_norm": 4.4555344581604,
540
+ "learning_rate": 1.2584269662921348e-05,
541
+ "loss": 0.2933,
542
+ "step": 760
543
+ },
544
+ {
545
+ "epoch": 0.3761602344894968,
546
+ "grad_norm": 10.27487850189209,
547
+ "learning_rate": 1.2486565705911091e-05,
548
+ "loss": 0.248,
549
+ "step": 770
550
+ },
551
+ {
552
+ "epoch": 0.38104543234000976,
553
+ "grad_norm": 13.357872009277344,
554
+ "learning_rate": 1.238886174890083e-05,
555
+ "loss": 0.3262,
556
+ "step": 780
557
+ },
558
+ {
559
+ "epoch": 0.3859306301905227,
560
+ "grad_norm": 0.9710779190063477,
561
+ "learning_rate": 1.2291157791890573e-05,
562
+ "loss": 0.4878,
563
+ "step": 790
564
+ },
565
+ {
566
+ "epoch": 0.39081582804103565,
567
+ "grad_norm": 8.897228240966797,
568
+ "learning_rate": 1.2193453834880314e-05,
569
+ "loss": 0.3705,
570
+ "step": 800
571
+ },
572
+ {
573
+ "epoch": 0.3957010258915486,
574
+ "grad_norm": 9.922685623168945,
575
+ "learning_rate": 1.2095749877870055e-05,
576
+ "loss": 0.6951,
577
+ "step": 810
578
+ },
579
+ {
580
+ "epoch": 0.40058622374206154,
581
+ "grad_norm": 12.713271141052246,
582
+ "learning_rate": 1.1998045920859796e-05,
583
+ "loss": 0.4512,
584
+ "step": 820
585
+ },
586
+ {
587
+ "epoch": 0.4054714215925745,
588
+ "grad_norm": 7.013885021209717,
589
+ "learning_rate": 1.1900341963849537e-05,
590
+ "loss": 0.4413,
591
+ "step": 830
592
+ },
593
+ {
594
+ "epoch": 0.41035661944308743,
595
+ "grad_norm": 17.14900016784668,
596
+ "learning_rate": 1.1802638006839278e-05,
597
+ "loss": 0.4688,
598
+ "step": 840
599
+ },
600
+ {
601
+ "epoch": 0.4152418172936004,
602
+ "grad_norm": 0.08244088292121887,
603
+ "learning_rate": 1.170493404982902e-05,
604
+ "loss": 0.3977,
605
+ "step": 850
606
+ },
607
+ {
608
+ "epoch": 0.4201270151441133,
609
+ "grad_norm": 6.348947048187256,
610
+ "learning_rate": 1.160723009281876e-05,
611
+ "loss": 0.5694,
612
+ "step": 860
613
+ },
614
+ {
615
+ "epoch": 0.42501221299462627,
616
+ "grad_norm": 0.7870075702667236,
617
+ "learning_rate": 1.1509526135808502e-05,
618
+ "loss": 0.4388,
619
+ "step": 870
620
+ },
621
+ {
622
+ "epoch": 0.4298974108451392,
623
+ "grad_norm": 9.187114715576172,
624
+ "learning_rate": 1.1411822178798242e-05,
625
+ "loss": 0.3909,
626
+ "step": 880
627
+ },
628
+ {
629
+ "epoch": 0.43478260869565216,
630
+ "grad_norm": 1.1734113693237305,
631
+ "learning_rate": 1.1314118221787984e-05,
632
+ "loss": 0.4053,
633
+ "step": 890
634
+ },
635
+ {
636
+ "epoch": 0.4396678065461651,
637
+ "grad_norm": 0.7321255803108215,
638
+ "learning_rate": 1.1216414264777724e-05,
639
+ "loss": 0.4325,
640
+ "step": 900
641
+ },
642
+ {
643
+ "epoch": 0.44455300439667805,
644
+ "grad_norm": 13.496896743774414,
645
+ "learning_rate": 1.1118710307767466e-05,
646
+ "loss": 0.4472,
647
+ "step": 910
648
+ },
649
+ {
650
+ "epoch": 0.449438202247191,
651
+ "grad_norm": 7.10970401763916,
652
+ "learning_rate": 1.1021006350757206e-05,
653
+ "loss": 0.5013,
654
+ "step": 920
655
+ },
656
+ {
657
+ "epoch": 0.45432340009770394,
658
+ "grad_norm": 17.575641632080078,
659
+ "learning_rate": 1.0923302393746948e-05,
660
+ "loss": 0.4886,
661
+ "step": 930
662
+ },
663
+ {
664
+ "epoch": 0.4592085979482169,
665
+ "grad_norm": 10.095125198364258,
666
+ "learning_rate": 1.0825598436736689e-05,
667
+ "loss": 0.4136,
668
+ "step": 940
669
+ },
670
+ {
671
+ "epoch": 0.46409379579872984,
672
+ "grad_norm": 6.8130292892456055,
673
+ "learning_rate": 1.072789447972643e-05,
674
+ "loss": 0.6457,
675
+ "step": 950
676
+ },
677
+ {
678
+ "epoch": 0.4689789936492428,
679
+ "grad_norm": 8.038426399230957,
680
+ "learning_rate": 1.0630190522716171e-05,
681
+ "loss": 0.3581,
682
+ "step": 960
683
+ },
684
+ {
685
+ "epoch": 0.4738641914997557,
686
+ "grad_norm": 3.269139528274536,
687
+ "learning_rate": 1.0532486565705912e-05,
688
+ "loss": 0.4969,
689
+ "step": 970
690
+ },
691
+ {
692
+ "epoch": 0.4787493893502687,
693
+ "grad_norm": 5.605432033538818,
694
+ "learning_rate": 1.0434782608695653e-05,
695
+ "loss": 0.3964,
696
+ "step": 980
697
+ },
698
+ {
699
+ "epoch": 0.4836345872007816,
700
+ "grad_norm": 6.879159927368164,
701
+ "learning_rate": 1.0337078651685396e-05,
702
+ "loss": 0.3005,
703
+ "step": 990
704
+ },
705
+ {
706
+ "epoch": 0.48851978505129456,
707
+ "grad_norm": 0.27517643570899963,
708
+ "learning_rate": 1.0239374694675135e-05,
709
+ "loss": 0.4044,
710
+ "step": 1000
711
+ },
712
+ {
713
+ "epoch": 0.4934049829018075,
714
+ "grad_norm": 18.99689483642578,
715
+ "learning_rate": 1.0141670737664878e-05,
716
+ "loss": 0.5072,
717
+ "step": 1010
718
+ },
719
+ {
720
+ "epoch": 0.49829018075232046,
721
+ "grad_norm": 6.253431797027588,
722
+ "learning_rate": 1.0043966780654617e-05,
723
+ "loss": 0.4955,
724
+ "step": 1020
725
+ },
726
+ {
727
+ "epoch": 0.5031753786028335,
728
+ "grad_norm": 5.144866466522217,
729
+ "learning_rate": 9.946262823644358e-06,
730
+ "loss": 0.2145,
731
+ "step": 1030
732
+ },
733
+ {
734
+ "epoch": 0.5080605764533463,
735
+ "grad_norm": 4.817426681518555,
736
+ "learning_rate": 9.848558866634099e-06,
737
+ "loss": 0.4397,
738
+ "step": 1040
739
+ },
740
+ {
741
+ "epoch": 0.5129457743038593,
742
+ "grad_norm": 11.465901374816895,
743
+ "learning_rate": 9.75085490962384e-06,
744
+ "loss": 0.3449,
745
+ "step": 1050
746
+ },
747
+ {
748
+ "epoch": 0.5178309721543722,
749
+ "grad_norm": 9.008246421813965,
750
+ "learning_rate": 9.65315095261358e-06,
751
+ "loss": 0.2979,
752
+ "step": 1060
753
+ },
754
+ {
755
+ "epoch": 0.5227161700048852,
756
+ "grad_norm": 4.4735636711120605,
757
+ "learning_rate": 9.555446995603323e-06,
758
+ "loss": 0.4377,
759
+ "step": 1070
760
+ },
761
+ {
762
+ "epoch": 0.5276013678553981,
763
+ "grad_norm": 2.9690377712249756,
764
+ "learning_rate": 9.457743038593064e-06,
765
+ "loss": 0.4898,
766
+ "step": 1080
767
+ },
768
+ {
769
+ "epoch": 0.5324865657059111,
770
+ "grad_norm": 4.649147033691406,
771
+ "learning_rate": 9.360039081582805e-06,
772
+ "loss": 0.5745,
773
+ "step": 1090
774
+ },
775
+ {
776
+ "epoch": 0.537371763556424,
777
+ "grad_norm": 1.9182021617889404,
778
+ "learning_rate": 9.262335124572546e-06,
779
+ "loss": 0.4778,
780
+ "step": 1100
781
+ },
782
+ {
783
+ "epoch": 0.542256961406937,
784
+ "grad_norm": 5.130582809448242,
785
+ "learning_rate": 9.164631167562287e-06,
786
+ "loss": 0.3995,
787
+ "step": 1110
788
+ },
789
+ {
790
+ "epoch": 0.5471421592574499,
791
+ "grad_norm": 1.418589472770691,
792
+ "learning_rate": 9.066927210552028e-06,
793
+ "loss": 0.3468,
794
+ "step": 1120
795
+ },
796
+ {
797
+ "epoch": 0.5520273571079629,
798
+ "grad_norm": 5.989200115203857,
799
+ "learning_rate": 8.969223253541769e-06,
800
+ "loss": 0.2423,
801
+ "step": 1130
802
+ },
803
+ {
804
+ "epoch": 0.5569125549584758,
805
+ "grad_norm": 12.977813720703125,
806
+ "learning_rate": 8.87151929653151e-06,
807
+ "loss": 0.4263,
808
+ "step": 1140
809
+ },
810
+ {
811
+ "epoch": 0.5617977528089888,
812
+ "grad_norm": 5.596517562866211,
813
+ "learning_rate": 8.773815339521251e-06,
814
+ "loss": 0.5422,
815
+ "step": 1150
816
+ },
817
+ {
818
+ "epoch": 0.5666829506595017,
819
+ "grad_norm": 9.800990104675293,
820
+ "learning_rate": 8.676111382510992e-06,
821
+ "loss": 0.4936,
822
+ "step": 1160
823
+ },
824
+ {
825
+ "epoch": 0.5715681485100147,
826
+ "grad_norm": 6.063791275024414,
827
+ "learning_rate": 8.578407425500733e-06,
828
+ "loss": 0.2772,
829
+ "step": 1170
830
+ },
831
+ {
832
+ "epoch": 0.5764533463605276,
833
+ "grad_norm": 6.097079753875732,
834
+ "learning_rate": 8.480703468490474e-06,
835
+ "loss": 0.2633,
836
+ "step": 1180
837
+ },
838
+ {
839
+ "epoch": 0.5813385442110406,
840
+ "grad_norm": 7.569230556488037,
841
+ "learning_rate": 8.382999511480215e-06,
842
+ "loss": 0.6212,
843
+ "step": 1190
844
+ },
845
+ {
846
+ "epoch": 0.5862237420615535,
847
+ "grad_norm": 1.887565016746521,
848
+ "learning_rate": 8.285295554469956e-06,
849
+ "loss": 0.4124,
850
+ "step": 1200
851
+ },
852
+ {
853
+ "epoch": 0.5911089399120665,
854
+ "grad_norm": 2.559924840927124,
855
+ "learning_rate": 8.187591597459697e-06,
856
+ "loss": 0.4965,
857
+ "step": 1210
858
+ },
859
+ {
860
+ "epoch": 0.5959941377625794,
861
+ "grad_norm": 1.9086635112762451,
862
+ "learning_rate": 8.08988764044944e-06,
863
+ "loss": 0.3139,
864
+ "step": 1220
865
+ },
866
+ {
867
+ "epoch": 0.6008793356130924,
868
+ "grad_norm": 2.0462982654571533,
869
+ "learning_rate": 7.99218368343918e-06,
870
+ "loss": 0.3031,
871
+ "step": 1230
872
+ },
873
+ {
874
+ "epoch": 0.6057645334636053,
875
+ "grad_norm": 3.3513333797454834,
876
+ "learning_rate": 7.894479726428921e-06,
877
+ "loss": 0.3425,
878
+ "step": 1240
879
+ },
880
+ {
881
+ "epoch": 0.6106497313141183,
882
+ "grad_norm": 9.786959648132324,
883
+ "learning_rate": 7.796775769418662e-06,
884
+ "loss": 0.5955,
885
+ "step": 1250
886
+ },
887
+ {
888
+ "epoch": 0.6155349291646312,
889
+ "grad_norm": 6.594499588012695,
890
+ "learning_rate": 7.699071812408403e-06,
891
+ "loss": 0.3468,
892
+ "step": 1260
893
+ },
894
+ {
895
+ "epoch": 0.6204201270151442,
896
+ "grad_norm": 14.10224437713623,
897
+ "learning_rate": 7.601367855398144e-06,
898
+ "loss": 0.5912,
899
+ "step": 1270
900
+ },
901
+ {
902
+ "epoch": 0.625305324865657,
903
+ "grad_norm": 9.290619850158691,
904
+ "learning_rate": 7.503663898387885e-06,
905
+ "loss": 0.6282,
906
+ "step": 1280
907
+ },
908
+ {
909
+ "epoch": 0.63019052271617,
910
+ "grad_norm": 11.20014762878418,
911
+ "learning_rate": 7.405959941377626e-06,
912
+ "loss": 0.2973,
913
+ "step": 1290
914
+ },
915
+ {
916
+ "epoch": 0.6350757205666829,
917
+ "grad_norm": 7.336958408355713,
918
+ "learning_rate": 7.308255984367367e-06,
919
+ "loss": 0.4738,
920
+ "step": 1300
921
+ },
922
+ {
923
+ "epoch": 0.6399609184171959,
924
+ "grad_norm": 0.04702078178524971,
925
+ "learning_rate": 7.210552027357108e-06,
926
+ "loss": 0.5161,
927
+ "step": 1310
928
+ },
929
+ {
930
+ "epoch": 0.6448461162677088,
931
+ "grad_norm": 7.214766502380371,
932
+ "learning_rate": 7.11284807034685e-06,
933
+ "loss": 0.4346,
934
+ "step": 1320
935
+ },
936
+ {
937
+ "epoch": 0.6497313141182218,
938
+ "grad_norm": 9.381763458251953,
939
+ "learning_rate": 7.015144113336591e-06,
940
+ "loss": 0.3613,
941
+ "step": 1330
942
+ },
943
+ {
944
+ "epoch": 0.6546165119687347,
945
+ "grad_norm": 6.74038028717041,
946
+ "learning_rate": 6.917440156326332e-06,
947
+ "loss": 0.4634,
948
+ "step": 1340
949
+ },
950
+ {
951
+ "epoch": 0.6595017098192477,
952
+ "grad_norm": 8.606428146362305,
953
+ "learning_rate": 6.819736199316073e-06,
954
+ "loss": 0.6249,
955
+ "step": 1350
956
+ },
957
+ {
958
+ "epoch": 0.6643869076697606,
959
+ "grad_norm": 14.990072250366211,
960
+ "learning_rate": 6.722032242305814e-06,
961
+ "loss": 0.517,
962
+ "step": 1360
963
+ },
964
+ {
965
+ "epoch": 0.6692721055202736,
966
+ "grad_norm": 5.394313335418701,
967
+ "learning_rate": 6.624328285295555e-06,
968
+ "loss": 0.3539,
969
+ "step": 1370
970
+ },
971
+ {
972
+ "epoch": 0.6741573033707865,
973
+ "grad_norm": 4.493410587310791,
974
+ "learning_rate": 6.526624328285296e-06,
975
+ "loss": 0.3574,
976
+ "step": 1380
977
+ },
978
+ {
979
+ "epoch": 0.6790425012212995,
980
+ "grad_norm": 12.217620849609375,
981
+ "learning_rate": 6.428920371275037e-06,
982
+ "loss": 0.5452,
983
+ "step": 1390
984
+ },
985
+ {
986
+ "epoch": 0.6839276990718124,
987
+ "grad_norm": 0.7943627834320068,
988
+ "learning_rate": 6.331216414264778e-06,
989
+ "loss": 0.2409,
990
+ "step": 1400
991
+ },
992
+ {
993
+ "epoch": 0.6888128969223254,
994
+ "grad_norm": 10.535755157470703,
995
+ "learning_rate": 6.233512457254519e-06,
996
+ "loss": 0.5708,
997
+ "step": 1410
998
+ },
999
+ {
1000
+ "epoch": 0.6936980947728383,
1001
+ "grad_norm": 7.569476127624512,
1002
+ "learning_rate": 6.13580850024426e-06,
1003
+ "loss": 0.2425,
1004
+ "step": 1420
1005
+ },
1006
+ {
1007
+ "epoch": 0.6985832926233513,
1008
+ "grad_norm": 7.93254280090332,
1009
+ "learning_rate": 6.038104543234001e-06,
1010
+ "loss": 0.3463,
1011
+ "step": 1430
1012
+ },
1013
+ {
1014
+ "epoch": 0.7034684904738642,
1015
+ "grad_norm": 2.2692394256591797,
1016
+ "learning_rate": 5.940400586223742e-06,
1017
+ "loss": 0.4627,
1018
+ "step": 1440
1019
+ },
1020
+ {
1021
+ "epoch": 0.7083536883243772,
1022
+ "grad_norm": 2.044731616973877,
1023
+ "learning_rate": 5.842696629213483e-06,
1024
+ "loss": 0.2857,
1025
+ "step": 1450
1026
+ },
1027
+ {
1028
+ "epoch": 0.7132388861748901,
1029
+ "grad_norm": 9.001659393310547,
1030
+ "learning_rate": 5.744992672203225e-06,
1031
+ "loss": 0.3712,
1032
+ "step": 1460
1033
+ },
1034
+ {
1035
+ "epoch": 0.7181240840254031,
1036
+ "grad_norm": 8.663535118103027,
1037
+ "learning_rate": 5.647288715192966e-06,
1038
+ "loss": 0.4056,
1039
+ "step": 1470
1040
+ },
1041
+ {
1042
+ "epoch": 0.723009281875916,
1043
+ "grad_norm": 2.3330843448638916,
1044
+ "learning_rate": 5.549584758182707e-06,
1045
+ "loss": 0.328,
1046
+ "step": 1480
1047
+ },
1048
+ {
1049
+ "epoch": 0.727894479726429,
1050
+ "grad_norm": 5.642626762390137,
1051
+ "learning_rate": 5.451880801172448e-06,
1052
+ "loss": 0.5109,
1053
+ "step": 1490
1054
+ },
1055
+ {
1056
+ "epoch": 0.7327796775769418,
1057
+ "grad_norm": 15.361739158630371,
1058
+ "learning_rate": 5.354176844162189e-06,
1059
+ "loss": 0.4808,
1060
+ "step": 1500
1061
+ },
1062
+ {
1063
+ "epoch": 0.7376648754274548,
1064
+ "grad_norm": 10.109480857849121,
1065
+ "learning_rate": 5.25647288715193e-06,
1066
+ "loss": 0.5841,
1067
+ "step": 1510
1068
+ },
1069
+ {
1070
+ "epoch": 0.7425500732779677,
1071
+ "grad_norm": 3.49552845954895,
1072
+ "learning_rate": 5.158768930141671e-06,
1073
+ "loss": 0.4657,
1074
+ "step": 1520
1075
+ },
1076
+ {
1077
+ "epoch": 0.7474352711284807,
1078
+ "grad_norm": 2.950178861618042,
1079
+ "learning_rate": 5.061064973131412e-06,
1080
+ "loss": 0.3558,
1081
+ "step": 1530
1082
+ },
1083
+ {
1084
+ "epoch": 0.7523204689789936,
1085
+ "grad_norm": 2.48479962348938,
1086
+ "learning_rate": 4.9633610161211535e-06,
1087
+ "loss": 0.5736,
1088
+ "step": 1540
1089
+ },
1090
+ {
1091
+ "epoch": 0.7572056668295066,
1092
+ "grad_norm": 2.948216676712036,
1093
+ "learning_rate": 4.8656570591108944e-06,
1094
+ "loss": 0.5026,
1095
+ "step": 1550
1096
+ },
1097
+ {
1098
+ "epoch": 0.7620908646800195,
1099
+ "grad_norm": 10.953573226928711,
1100
+ "learning_rate": 4.767953102100635e-06,
1101
+ "loss": 0.3647,
1102
+ "step": 1560
1103
+ },
1104
+ {
1105
+ "epoch": 0.7669760625305325,
1106
+ "grad_norm": 0.7188931703567505,
1107
+ "learning_rate": 4.670249145090376e-06,
1108
+ "loss": 0.4338,
1109
+ "step": 1570
1110
+ },
1111
+ {
1112
+ "epoch": 0.7718612603810454,
1113
+ "grad_norm": 5.085393905639648,
1114
+ "learning_rate": 4.572545188080117e-06,
1115
+ "loss": 0.3118,
1116
+ "step": 1580
1117
+ },
1118
+ {
1119
+ "epoch": 0.7767464582315584,
1120
+ "grad_norm": 0.31807640194892883,
1121
+ "learning_rate": 4.474841231069858e-06,
1122
+ "loss": 0.4693,
1123
+ "step": 1590
1124
+ },
1125
+ {
1126
+ "epoch": 0.7816316560820713,
1127
+ "grad_norm": 4.8351287841796875,
1128
+ "learning_rate": 4.377137274059599e-06,
1129
+ "loss": 0.5351,
1130
+ "step": 1600
1131
+ },
1132
+ {
1133
+ "epoch": 0.7865168539325843,
1134
+ "grad_norm": 4.054388999938965,
1135
+ "learning_rate": 4.279433317049341e-06,
1136
+ "loss": 0.4113,
1137
+ "step": 1610
1138
+ },
1139
+ {
1140
+ "epoch": 0.7914020517830972,
1141
+ "grad_norm": 9.017539024353027,
1142
+ "learning_rate": 4.181729360039082e-06,
1143
+ "loss": 0.3507,
1144
+ "step": 1620
1145
+ },
1146
+ {
1147
+ "epoch": 0.7962872496336102,
1148
+ "grad_norm": 0.9139829874038696,
1149
+ "learning_rate": 4.084025403028823e-06,
1150
+ "loss": 0.5123,
1151
+ "step": 1630
1152
+ },
1153
+ {
1154
+ "epoch": 0.8011724474841231,
1155
+ "grad_norm": 5.668177127838135,
1156
+ "learning_rate": 3.986321446018564e-06,
1157
+ "loss": 0.4119,
1158
+ "step": 1640
1159
+ },
1160
+ {
1161
+ "epoch": 0.8060576453346361,
1162
+ "grad_norm": 3.6785178184509277,
1163
+ "learning_rate": 3.888617489008305e-06,
1164
+ "loss": 0.3159,
1165
+ "step": 1650
1166
+ },
1167
+ {
1168
+ "epoch": 0.810942843185149,
1169
+ "grad_norm": 1.998555302619934,
1170
+ "learning_rate": 3.7909135319980463e-06,
1171
+ "loss": 0.3291,
1172
+ "step": 1660
1173
+ },
1174
+ {
1175
+ "epoch": 0.815828041035662,
1176
+ "grad_norm": 1.5727540254592896,
1177
+ "learning_rate": 3.6932095749877872e-06,
1178
+ "loss": 0.3416,
1179
+ "step": 1670
1180
+ },
1181
+ {
1182
+ "epoch": 0.8207132388861749,
1183
+ "grad_norm": 11.895509719848633,
1184
+ "learning_rate": 3.595505617977528e-06,
1185
+ "loss": 0.3537,
1186
+ "step": 1680
1187
+ },
1188
+ {
1189
+ "epoch": 0.8255984367366879,
1190
+ "grad_norm": 5.824265480041504,
1191
+ "learning_rate": 3.4978016609672696e-06,
1192
+ "loss": 0.2603,
1193
+ "step": 1690
1194
+ },
1195
+ {
1196
+ "epoch": 0.8304836345872008,
1197
+ "grad_norm": 12.911538124084473,
1198
+ "learning_rate": 3.4000977039570105e-06,
1199
+ "loss": 0.5172,
1200
+ "step": 1700
1201
+ },
1202
+ {
1203
+ "epoch": 0.8353688324377138,
1204
+ "grad_norm": 6.1299614906311035,
1205
+ "learning_rate": 3.3023937469467515e-06,
1206
+ "loss": 0.3283,
1207
+ "step": 1710
1208
+ },
1209
+ {
1210
+ "epoch": 0.8402540302882266,
1211
+ "grad_norm": 8.103076934814453,
1212
+ "learning_rate": 3.2046897899364925e-06,
1213
+ "loss": 0.2746,
1214
+ "step": 1720
1215
+ },
1216
+ {
1217
+ "epoch": 0.8451392281387397,
1218
+ "grad_norm": 10.692161560058594,
1219
+ "learning_rate": 3.106985832926234e-06,
1220
+ "loss": 0.6933,
1221
+ "step": 1730
1222
+ },
1223
+ {
1224
+ "epoch": 0.8500244259892525,
1225
+ "grad_norm": 5.779560089111328,
1226
+ "learning_rate": 3.009281875915975e-06,
1227
+ "loss": 0.3408,
1228
+ "step": 1740
1229
+ },
1230
+ {
1231
+ "epoch": 0.8549096238397655,
1232
+ "grad_norm": 6.992332935333252,
1233
+ "learning_rate": 2.9115779189057158e-06,
1234
+ "loss": 0.4603,
1235
+ "step": 1750
1236
+ },
1237
+ {
1238
+ "epoch": 0.8597948216902784,
1239
+ "grad_norm": 7.0532121658325195,
1240
+ "learning_rate": 2.8138739618954567e-06,
1241
+ "loss": 0.304,
1242
+ "step": 1760
1243
+ },
1244
+ {
1245
+ "epoch": 0.8646800195407914,
1246
+ "grad_norm": 14.865501403808594,
1247
+ "learning_rate": 2.716170004885198e-06,
1248
+ "loss": 0.6035,
1249
+ "step": 1770
1250
+ },
1251
+ {
1252
+ "epoch": 0.8695652173913043,
1253
+ "grad_norm": 8.351088523864746,
1254
+ "learning_rate": 2.618466047874939e-06,
1255
+ "loss": 0.4957,
1256
+ "step": 1780
1257
+ },
1258
+ {
1259
+ "epoch": 0.8744504152418173,
1260
+ "grad_norm": 8.012289047241211,
1261
+ "learning_rate": 2.52076209086468e-06,
1262
+ "loss": 0.4995,
1263
+ "step": 1790
1264
+ },
1265
+ {
1266
+ "epoch": 0.8793356130923302,
1267
+ "grad_norm": 10.582439422607422,
1268
+ "learning_rate": 2.4230581338544214e-06,
1269
+ "loss": 0.3412,
1270
+ "step": 1800
1271
+ },
1272
+ {
1273
+ "epoch": 0.8842208109428432,
1274
+ "grad_norm": 7.599929332733154,
1275
+ "learning_rate": 2.3253541768441624e-06,
1276
+ "loss": 0.2842,
1277
+ "step": 1810
1278
+ },
1279
+ {
1280
+ "epoch": 0.8891060087933561,
1281
+ "grad_norm": 9.53080940246582,
1282
+ "learning_rate": 2.2276502198339033e-06,
1283
+ "loss": 0.4446,
1284
+ "step": 1820
1285
+ },
1286
+ {
1287
+ "epoch": 0.8939912066438691,
1288
+ "grad_norm": 6.277515888214111,
1289
+ "learning_rate": 2.1299462628236443e-06,
1290
+ "loss": 0.5967,
1291
+ "step": 1830
1292
+ },
1293
+ {
1294
+ "epoch": 0.898876404494382,
1295
+ "grad_norm": 0.5566537380218506,
1296
+ "learning_rate": 2.0322423058133857e-06,
1297
+ "loss": 0.3504,
1298
+ "step": 1840
1299
+ },
1300
+ {
1301
+ "epoch": 0.903761602344895,
1302
+ "grad_norm": 8.149109840393066,
1303
+ "learning_rate": 1.9345383488031266e-06,
1304
+ "loss": 0.4962,
1305
+ "step": 1850
1306
+ },
1307
+ {
1308
+ "epoch": 0.9086468001954079,
1309
+ "grad_norm": 16.717893600463867,
1310
+ "learning_rate": 1.8368343917928678e-06,
1311
+ "loss": 0.543,
1312
+ "step": 1860
1313
+ },
1314
+ {
1315
+ "epoch": 0.9135319980459209,
1316
+ "grad_norm": 6.368210315704346,
1317
+ "learning_rate": 1.7391304347826088e-06,
1318
+ "loss": 0.3421,
1319
+ "step": 1870
1320
+ },
1321
+ {
1322
+ "epoch": 0.9184171958964338,
1323
+ "grad_norm": 7.611034393310547,
1324
+ "learning_rate": 1.64142647777235e-06,
1325
+ "loss": 0.2211,
1326
+ "step": 1880
1327
+ },
1328
+ {
1329
+ "epoch": 0.9233023937469468,
1330
+ "grad_norm": 6.015079498291016,
1331
+ "learning_rate": 1.5437225207620909e-06,
1332
+ "loss": 0.4114,
1333
+ "step": 1890
1334
+ },
1335
+ {
1336
+ "epoch": 0.9281875915974597,
1337
+ "grad_norm": 5.758971214294434,
1338
+ "learning_rate": 1.446018563751832e-06,
1339
+ "loss": 0.2586,
1340
+ "step": 1900
1341
+ },
1342
+ {
1343
+ "epoch": 0.9330727894479727,
1344
+ "grad_norm": 10.397595405578613,
1345
+ "learning_rate": 1.348314606741573e-06,
1346
+ "loss": 0.3601,
1347
+ "step": 1910
1348
+ },
1349
+ {
1350
+ "epoch": 0.9379579872984856,
1351
+ "grad_norm": 6.008655548095703,
1352
+ "learning_rate": 1.2506106497313142e-06,
1353
+ "loss": 0.3286,
1354
+ "step": 1920
1355
+ },
1356
+ {
1357
+ "epoch": 0.9428431851489986,
1358
+ "grad_norm": 10.383055686950684,
1359
+ "learning_rate": 1.1529066927210554e-06,
1360
+ "loss": 0.3969,
1361
+ "step": 1930
1362
+ },
1363
+ {
1364
+ "epoch": 0.9477283829995115,
1365
+ "grad_norm": 5.051438331604004,
1366
+ "learning_rate": 1.0552027357107963e-06,
1367
+ "loss": 0.2447,
1368
+ "step": 1940
1369
+ },
1370
+ {
1371
+ "epoch": 0.9526135808500245,
1372
+ "grad_norm": 9.03067684173584,
1373
+ "learning_rate": 9.574987787005375e-07,
1374
+ "loss": 0.2699,
1375
+ "step": 1950
1376
+ },
1377
+ {
1378
+ "epoch": 0.9574987787005373,
1379
+ "grad_norm": 5.482412338256836,
1380
+ "learning_rate": 8.597948216902785e-07,
1381
+ "loss": 0.4556,
1382
+ "step": 1960
1383
+ },
1384
+ {
1385
+ "epoch": 0.9623839765510503,
1386
+ "grad_norm": 9.010185241699219,
1387
+ "learning_rate": 7.620908646800196e-07,
1388
+ "loss": 0.3432,
1389
+ "step": 1970
1390
+ },
1391
+ {
1392
+ "epoch": 0.9672691744015632,
1393
+ "grad_norm": 7.145694255828857,
1394
+ "learning_rate": 6.643869076697607e-07,
1395
+ "loss": 0.2654,
1396
+ "step": 1980
1397
+ },
1398
+ {
1399
+ "epoch": 0.9721543722520762,
1400
+ "grad_norm": 6.059538841247559,
1401
+ "learning_rate": 5.666829506595018e-07,
1402
+ "loss": 0.2687,
1403
+ "step": 1990
1404
+ },
1405
+ {
1406
+ "epoch": 0.9770395701025891,
1407
+ "grad_norm": 3.838576555252075,
1408
+ "learning_rate": 4.689789936492428e-07,
1409
+ "loss": 0.3676,
1410
+ "step": 2000
1411
+ },
1412
+ {
1413
+ "epoch": 0.9819247679531021,
1414
+ "grad_norm": 6.402182102203369,
1415
+ "learning_rate": 3.712750366389839e-07,
1416
+ "loss": 0.415,
1417
+ "step": 2010
1418
+ },
1419
+ {
1420
+ "epoch": 0.986809965803615,
1421
+ "grad_norm": 1.021118402481079,
1422
+ "learning_rate": 2.7357107962872495e-07,
1423
+ "loss": 0.307,
1424
+ "step": 2020
1425
+ },
1426
+ {
1427
+ "epoch": 0.991695163654128,
1428
+ "grad_norm": 8.60698413848877,
1429
+ "learning_rate": 1.7586712261846604e-07,
1430
+ "loss": 0.5131,
1431
+ "step": 2030
1432
+ },
1433
+ {
1434
+ "epoch": 0.9965803615046409,
1435
+ "grad_norm": 7.274533271789551,
1436
+ "learning_rate": 7.816316560820714e-08,
1437
+ "loss": 0.3784,
1438
+ "step": 2040
1439
+ },
1440
+ {
1441
+ "epoch": 1.0,
1442
+ "eval_accuracy": 0.8527472527472527,
1443
+ "eval_loss": 0.4573296904563904,
1444
+ "eval_runtime": 505.9062,
1445
+ "eval_samples_per_second": 8.094,
1446
+ "eval_steps_per_second": 1.012,
1447
+ "step": 2047
1448
+ }
1449
+ ],
1450
+ "logging_steps": 10,
1451
+ "max_steps": 2047,
1452
+ "num_input_tokens_seen": 0,
1453
+ "num_train_epochs": 1,
1454
+ "save_steps": 500,
1455
+ "stateful_callbacks": {
1456
+ "TrainerControl": {
1457
+ "args": {
1458
+ "should_epoch_stop": false,
1459
+ "should_evaluate": false,
1460
+ "should_log": false,
1461
+ "should_save": true,
1462
+ "should_training_stop": true
1463
+ },
1464
+ "attributes": {}
1465
+ }
1466
+ },
1467
+ "total_flos": 1.2690656905519596e+18,
1468
+ "train_batch_size": 8,
1469
+ "trial_name": null,
1470
+ "trial_params": null
1471
+ }
checkpoint-2047/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:66708c48cae981cf94bc4aeb4bfd5275154c3f88e241dffc030d8fd493a80290
3
+ size 5777
config.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "ViTForImageClassification"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.0,
6
+ "dtype": "float32",
7
+ "encoder_stride": 16,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.0,
10
+ "hidden_size": 768,
11
+ "id2label": {
12
+ "0": "anger",
13
+ "1": "disgust",
14
+ "2": "fear",
15
+ "3": "happiness",
16
+ "4": "neutral",
17
+ "5": "sadness",
18
+ "6": "surprise"
19
+ },
20
+ "image_size": 224,
21
+ "initializer_range": 0.02,
22
+ "intermediate_size": 3072,
23
+ "label2id": {
24
+ "anger": "0",
25
+ "disgust": "1",
26
+ "fear": "2",
27
+ "happiness": "3",
28
+ "neutral": "4",
29
+ "sadness": "5",
30
+ "surprise": "6"
31
+ },
32
+ "layer_norm_eps": 1e-12,
33
+ "model_type": "vit",
34
+ "num_attention_heads": 12,
35
+ "num_channels": 3,
36
+ "num_hidden_layers": 12,
37
+ "patch_size": 16,
38
+ "pooler_act": "tanh",
39
+ "pooler_output_size": 768,
40
+ "problem_type": "single_label_classification",
41
+ "qkv_bias": true,
42
+ "transformers_version": "4.57.6"
43
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a0280a8546c3a92190fa43a0b6fb9a34ef22e936b44700f3041bc080f5a2ab0b
3
+ size 343239356
preprocessor_config.json ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_convert_rgb": null,
3
+ "do_normalize": true,
4
+ "do_rescale": true,
5
+ "do_resize": true,
6
+ "image_mean": [
7
+ 0.5,
8
+ 0.5,
9
+ 0.5
10
+ ],
11
+ "image_processor_type": "ViTImageProcessor",
12
+ "image_std": [
13
+ 0.5,
14
+ 0.5,
15
+ 0.5
16
+ ],
17
+ "resample": 2,
18
+ "rescale_factor": 0.00392156862745098,
19
+ "size": {
20
+ "height": 224,
21
+ "width": 224
22
+ }
23
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:66708c48cae981cf94bc4aeb4bfd5275154c3f88e241dffc030d8fd493a80290
3
+ size 5777