Hugging Face's logo

Shawon16
/
ViViT_lsa64_coR

Video Classification
Transformers
Safetensors
vivit
Generated from Trainer
Model card Files
xet
 Community
Shawon16 commited on
Commit
124ec6b
·
verified ·
1 Parent(s): 41eea0e

Model save

Browse files
Files changed (3) hide show
  1. README.md +72 -0
  2. model.safetensors +1 -1
  3. trainer_state.json +179 -0
README.md ADDED
@@ -0,0 +1,72 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ license: mit
4
+ base_model: google/vivit-b-16x2-kinetics400
5
+ tags:
6
+ - generated_from_trainer
7
+ metrics:
8
+ - accuracy
9
+ model-index:
10
+ - name: ViViT_lsa64_coR
11
+ results: []
12
+ ---
13
+
14
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
15
+ should probably proofread and complete it, then remove this comment. -->
16
+
17
+ # ViViT_lsa64_coR
18
+
19
+ This model is a fine-tuned version of [google/vivit-b-16x2-kinetics400](https://huggingface.co/google/vivit-b-16x2-kinetics400) on an unknown dataset.
20
+ It achieves the following results on the evaluation set:
21
+ - Loss: 0.0008
22
+ - Accuracy: 1.0
23
+
24
+ ## Model description
25
+
26
+ More information needed
27
+
28
+ ## Intended uses & limitations
29
+
30
+ More information needed
31
+
32
+ ## Training and evaluation data
33
+
34
+ More information needed
35
+
36
+ ## Training procedure
37
+
38
+ ### Training hyperparameters
39
+
40
+ The following hyperparameters were used during training:
41
+ - learning_rate: 5e-05
42
+ - train_batch_size: 2
43
+ - eval_batch_size: 2
44
+ - seed: 42
45
+ - gradient_accumulation_steps: 4
46
+ - total_train_batch_size: 8
47
+ - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
48
+ - lr_scheduler_type: linear
49
+ - lr_scheduler_warmup_ratio: 0.1
50
+ - training_steps: 2880
51
+ - mixed_precision_training: Native AMP
52
+
53
+ ### Training results
54
+
55
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy |
56
+ |:-------------:|:-----:|:----:|:---------------:|:--------:|
57
+ | 12.7049 | 0.1 | 288 | 1.3316 | 0.8438 |
58
+ | 1.4335 | 1.1 | 576 | 0.0854 | 0.9922 |
59
+ | 0.0869 | 2.1 | 864 | 0.0054 | 1.0 |
60
+ | 0.0225 | 3.1 | 1152 | 0.0021 | 1.0 |
61
+ | 0.0057 | 4.1 | 1440 | 0.0012 | 1.0 |
62
+ | 0.0038 | 5.1 | 1728 | 0.0010 | 1.0 |
63
+ | 0.0024 | 6.1 | 2016 | 0.0008 | 1.0 |
64
+ | 0.0016 | 7.1 | 2304 | 0.0008 | 1.0 |
65
+
66
+
67
+ ### Framework versions
68
+
69
+ - Transformers 4.46.1
70
+ - Pytorch 2.5.1+cu124
71
+ - Datasets 3.1.0
72
+ - Tokenizers 0.20.1
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f45f510109650f5c9936ed7ebefbaa225f6b8456357359a48d5654709c183a34
3
  size 354806104
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:27a4e9d01fdd3c437e1750dc143c04b7bf5249b8ab5924c6d4b841535a09017f
3
  size 354806104
trainer_state.json ADDED
@@ -0,0 +1,179 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 1.0,
3
+ "best_model_checkpoint": "/media/cse/HDD/Shawon/shawon/10 fold timesformer/ViViT_lsa64_coR/checkpoint-864",
4
+ "epoch": 7.1,
5
+ "eval_steps": 500,
6
+ "global_step": 2304,
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
+ "grad_norm": 34.6246223449707,
14
+ "learning_rate": 4.930555555555556e-05,
15
+ "loss": 12.7049,
16
+ "step": 288
17
+ },
18
+ {
19
+ "epoch": 0.1,
20
+ "eval_accuracy": 0.84375,
21
+ "eval_loss": 1.3316307067871094,
22
+ "eval_runtime": 289.0213,
23
+ "eval_samples_per_second": 0.886,
24
+ "eval_steps_per_second": 0.443,
25
+ "step": 288
26
+ },
27
+ {
28
+ "epoch": 1.1,
29
+ "grad_norm": 1.3600924015045166,
30
+ "learning_rate": 4.4521604938271604e-05,
31
+ "loss": 1.4335,
32
+ "step": 576
33
+ },
34
+ {
35
+ "epoch": 1.1,
36
+ "eval_accuracy": 0.9921875,
37
+ "eval_loss": 0.08541239798069,
38
+ "eval_runtime": 353.5336,
39
+ "eval_samples_per_second": 0.724,
40
+ "eval_steps_per_second": 0.362,
41
+ "step": 576
42
+ },
43
+ {
44
+ "epoch": 2.1,
45
+ "grad_norm": 0.12772603332996368,
46
+ "learning_rate": 3.8966049382716055e-05,
47
+ "loss": 0.0869,
48
+ "step": 864
49
+ },
50
+ {
51
+ "epoch": 2.1,
52
+ "eval_accuracy": 1.0,
53
+ "eval_loss": 0.00544479675590992,
54
+ "eval_runtime": 303.6919,
55
+ "eval_samples_per_second": 0.843,
56
+ "eval_steps_per_second": 0.421,
57
+ "step": 864
58
+ },
59
+ {
60
+ "epoch": 3.1,
61
+ "grad_norm": 0.10945271700620651,
62
+ "learning_rate": 3.341049382716049e-05,
63
+ "loss": 0.0225,
64
+ "step": 1152
65
+ },
66
+ {
67
+ "epoch": 3.1,
68
+ "eval_accuracy": 1.0,
69
+ "eval_loss": 0.002123473212122917,
70
+ "eval_runtime": 394.5624,
71
+ "eval_samples_per_second": 0.649,
72
+ "eval_steps_per_second": 0.324,
73
+ "step": 1152
74
+ },
75
+ {
76
+ "epoch": 4.1,
77
+ "grad_norm": 0.0422612763941288,
78
+ "learning_rate": 2.785493827160494e-05,
79
+ "loss": 0.0057,
80
+ "step": 1440
81
+ },
82
+ {
83
+ "epoch": 4.1,
84
+ "eval_accuracy": 1.0,
85
+ "eval_loss": 0.0011819098144769669,
86
+ "eval_runtime": 324.5781,
87
+ "eval_samples_per_second": 0.789,
88
+ "eval_steps_per_second": 0.394,
89
+ "step": 1440
90
+ },
91
+ {
92
+ "epoch": 5.1,
93
+ "grad_norm": 0.026100359857082367,
94
+ "learning_rate": 2.2299382716049384e-05,
95
+ "loss": 0.0038,
96
+ "step": 1728
97
+ },
98
+ {
99
+ "epoch": 5.1,
100
+ "eval_accuracy": 1.0,
101
+ "eval_loss": 0.0009726639837026596,
102
+ "eval_runtime": 354.0444,
103
+ "eval_samples_per_second": 0.723,
104
+ "eval_steps_per_second": 0.362,
105
+ "step": 1728
106
+ },
107
+ {
108
+ "epoch": 6.1,
109
+ "grad_norm": 0.02095157280564308,
110
+ "learning_rate": 1.674382716049383e-05,
111
+ "loss": 0.0024,
112
+ "step": 2016
113
+ },
114
+ {
115
+ "epoch": 6.1,
116
+ "eval_accuracy": 1.0,
117
+ "eval_loss": 0.0008304659277200699,
118
+ "eval_runtime": 359.9913,
119
+ "eval_samples_per_second": 0.711,
120
+ "eval_steps_per_second": 0.356,
121
+ "step": 2016
122
+ },
123
+ {
124
+ "epoch": 7.1,
125
+ "grad_norm": 0.028533462435007095,
126
+ "learning_rate": 1.1188271604938271e-05,
127
+ "loss": 0.0016,
128
+ "step": 2304
129
+ },
130
+ {
131
+ "epoch": 7.1,
132
+ "eval_accuracy": 1.0,
133
+ "eval_loss": 0.0007986431010067463,
134
+ "eval_runtime": 294.083,
135
+ "eval_samples_per_second": 0.871,
136
+ "eval_steps_per_second": 0.435,
137
+ "step": 2304
138
+ },
139
+ {
140
+ "epoch": 7.1,
141
+ "step": 2304,
142
+ "total_flos": 4.72490560063663e+19,
143
+ "train_loss": 1.7826555360419054,
144
+ "train_runtime": 26105.5788,
145
+ "train_samples_per_second": 0.883,
146
+ "train_steps_per_second": 0.11
147
+ }
148
+ ],
149
+ "logging_steps": 500,
150
+ "max_steps": 2880,
151
+ "num_input_tokens_seen": 0,
152
+ "num_train_epochs": 9223372036854775807,
153
+ "save_steps": 500,
154
+ "stateful_callbacks": {
155
+ "EarlyStoppingCallback": {
156
+ "args": {
157
+ "early_stopping_patience": 5,
158
+ "early_stopping_threshold": 0.0
159
+ },
160
+ "attributes": {
161
+ "early_stopping_patience_counter": 5
162
+ }
163
+ },
164
+ "TrainerControl": {
165
+ "args": {
166
+ "should_epoch_stop": false,
167
+ "should_evaluate": false,
168
+ "should_log": false,
169
+ "should_save": true,
170
+ "should_training_stop": true
171
+ },
172
+ "attributes": {}
173
+ }
174
+ },
175
+ "total_flos": 4.72490560063663e+19,
176
+ "train_batch_size": 2,
177
+ "trial_name": null,
178
+ "trial_params": null
179
+ }