Upload folder using huggingface_hub
Browse files- README.md +108 -0
- metadata.json +8 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +484 -0
- training_args.bin +3 -0
README.md
ADDED
@@ -0,0 +1,108 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
base_model: google/vit-base-patch16-224
|
4 |
+
tags:
|
5 |
+
- Image Regression
|
6 |
+
datasets:
|
7 |
+
- "tonyassi/tony__assi-ig-ds200"
|
8 |
+
metrics:
|
9 |
+
- accuracy
|
10 |
+
model-index:
|
11 |
+
- name: "tony__assi-ig-prediction200"
|
12 |
+
results: []
|
13 |
+
---
|
14 |
+
|
15 |
+
# tony__assi-ig-prediction200
|
16 |
+
## IG Prediction
|
17 |
+
|
18 |
+
This model was trained with [IGPrediction](https://github.com/TonyAssi/IGPrediction). It predicts how many likes an image will get.
|
19 |
+
|
20 |
+
```python
|
21 |
+
from IGPredict import predict_ig
|
22 |
+
predict_ig(repo_id='tonyassi/tony__assi-ig-prediction200',image_path='image.jpg')
|
23 |
+
```
|
24 |
+
|
25 |
+
---
|
26 |
+
|
27 |
+
## Dataset
|
28 |
+
Dataset: tonyassi/tony__assi-ig-ds200\
|
29 |
+
Value Column:\
|
30 |
+
Train Test Split: 0.2
|
31 |
+
|
32 |
+
---
|
33 |
+
|
34 |
+
## Training
|
35 |
+
Base Model: [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224)\
|
36 |
+
Epochs: 20\
|
37 |
+
Learning Rate: 0.0001
|
38 |
+
|
39 |
+
---
|
40 |
+
|
41 |
+
## Usage
|
42 |
+
|
43 |
+
### Download
|
44 |
+
```bash
|
45 |
+
git clone https://github.com/TonyAssi/IGPrediction.git
|
46 |
+
cd IGPrediction
|
47 |
+
```
|
48 |
+
|
49 |
+
### Installation
|
50 |
+
```bash
|
51 |
+
pip install -r requirements.txt
|
52 |
+
```
|
53 |
+
|
54 |
+
### Import
|
55 |
+
```python
|
56 |
+
from IGPredict import ig_download, upload_dataset, train_ig_model, upload_ig_model, predict_ig
|
57 |
+
```
|
58 |
+
|
59 |
+
### Download Instagram Images
|
60 |
+
- **username** Instagram username
|
61 |
+
- **num_images** maximum number of images to download
|
62 |
+
```python
|
63 |
+
ig_download(username='instagarm_username', num_images=100)
|
64 |
+
```
|
65 |
+
Instagram images will be downloaded to *'./images'* folder, each one named like so *"index-likes.jpg"*. E.g. *"3-17.jpg"* is the third image and has 17 likes.
|
66 |
+
|
67 |
+
### Upload Dataset
|
68 |
+
- **dataset_name** name of dataset to be uploaded
|
69 |
+
- **token** go [here](https://huggingface.co/settings/tokens) to create a new 🤗 token
|
70 |
+
```python
|
71 |
+
upload_dataset(dataset_name='tonyassi/tony__assi-ig-ds200', token='YOUR_HF_TOKEN')
|
72 |
+
```
|
73 |
+
Go to your 🤗 profile to find your uploaded dataset, it should look similar to [tonyassi/tony__assi-ig-ds](https://huggingface.co/datasets/tonyassi/tony__assi-ig-ds).
|
74 |
+
|
75 |
+
|
76 |
+
### Train Model
|
77 |
+
- **dataset_id** 🤗 dataset id
|
78 |
+
- **test_split** test split of the train/test split
|
79 |
+
- **num_train_epochs** training epochs
|
80 |
+
- **learning_rate** learning rate
|
81 |
+
```python
|
82 |
+
train_ig_model(dataset_id='tonyassi/tony__assi-ig-ds200',
|
83 |
+
test_split=0.2,
|
84 |
+
num_train_epochs=20,
|
85 |
+
learning_rate=0.0001)
|
86 |
+
|
87 |
+
```
|
88 |
+
The trainer will save the checkpoints in the 'results' folder. The model.safetensors are the trained weights you'll use for inference (predicton).
|
89 |
+
|
90 |
+
### Upload Model
|
91 |
+
This function will upload your model to the 🤗 Hub.
|
92 |
+
- **model_id** the name of the model id
|
93 |
+
- **token** go [here](https://huggingface.co/settings/tokens) to create a new 🤗 token
|
94 |
+
- **checkpoint_dir** checkpoint folder that will be uploaded
|
95 |
+
```python
|
96 |
+
upload_ig_model(model_id='tony__assi-ig-prediction200',
|
97 |
+
token='YOUR_HF_TOKEN',
|
98 |
+
checkpoint_dir='./results/checkpoint-940')
|
99 |
+
```
|
100 |
+
|
101 |
+
### Inference (Prediction)
|
102 |
+
- **repo_id** 🤗 repo id of the model
|
103 |
+
- **image_path** path to image
|
104 |
+
```python
|
105 |
+
predict_ig(repo_id='tonyassi/tony__assi-ig-prediction200',
|
106 |
+
image_path='image.jpg')
|
107 |
+
```
|
108 |
+
The first time this function is called it'll download the safetensor model. Subsequent function calls will run faster.
|
metadata.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"dataset_id": "tonyassi/tony__assi-ig-ds200",
|
3 |
+
"value_column_name": "likes",
|
4 |
+
"test_split": 0.2,
|
5 |
+
"num_train_epochs": 20,
|
6 |
+
"learning_rate": 0.0001,
|
7 |
+
"max_value": 177
|
8 |
+
}
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:11165000fc90d43fc659af621bd55bdac2bf8e48a855e0e3faafe53bd6f69919
|
3 |
+
size 345583444
|
optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2b2f195d5daaea171bd20a884a9738f494432d6acfa0bc1cf0390ff3f9d9f488
|
3 |
+
size 686557178
|
rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8098b3ea1099554a17a75b08e08b0801f551831d4783349401defb321e1c1c1d
|
3 |
+
size 13990
|
scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f3ad1e13a84e759bcc62d699a5143c5c275ee621d7d69ae8816d20e1d9121294
|
3 |
+
size 1064
|
trainer_state.json
ADDED
@@ -0,0 +1,484 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 20.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 400,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.5,
|
13 |
+
"grad_norm": 14.416919708251953,
|
14 |
+
"learning_rate": 9.75e-05,
|
15 |
+
"loss": 0.3808,
|
16 |
+
"step": 10
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 1.0,
|
20 |
+
"grad_norm": 14.942909240722656,
|
21 |
+
"learning_rate": 9.5e-05,
|
22 |
+
"loss": 0.1636,
|
23 |
+
"step": 20
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 1.0,
|
27 |
+
"eval_loss": 0.1972920000553131,
|
28 |
+
"eval_mse": 0.1972920149564743,
|
29 |
+
"eval_runtime": 1.8069,
|
30 |
+
"eval_samples_per_second": 21.584,
|
31 |
+
"eval_steps_per_second": 2.767,
|
32 |
+
"step": 20
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 1.5,
|
36 |
+
"grad_norm": 2.4087517261505127,
|
37 |
+
"learning_rate": 9.250000000000001e-05,
|
38 |
+
"loss": 0.2302,
|
39 |
+
"step": 30
|
40 |
+
},
|
41 |
+
{
|
42 |
+
"epoch": 2.0,
|
43 |
+
"grad_norm": 21.778547286987305,
|
44 |
+
"learning_rate": 9e-05,
|
45 |
+
"loss": 0.1331,
|
46 |
+
"step": 40
|
47 |
+
},
|
48 |
+
{
|
49 |
+
"epoch": 2.0,
|
50 |
+
"eval_loss": 0.046439673751592636,
|
51 |
+
"eval_mse": 0.046439677476882935,
|
52 |
+
"eval_runtime": 1.622,
|
53 |
+
"eval_samples_per_second": 24.044,
|
54 |
+
"eval_steps_per_second": 3.083,
|
55 |
+
"step": 40
|
56 |
+
},
|
57 |
+
{
|
58 |
+
"epoch": 2.5,
|
59 |
+
"grad_norm": 1.1269134283065796,
|
60 |
+
"learning_rate": 8.75e-05,
|
61 |
+
"loss": 0.05,
|
62 |
+
"step": 50
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 3.0,
|
66 |
+
"grad_norm": 4.305675983428955,
|
67 |
+
"learning_rate": 8.5e-05,
|
68 |
+
"loss": 0.0289,
|
69 |
+
"step": 60
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 3.0,
|
73 |
+
"eval_loss": 0.03575298935174942,
|
74 |
+
"eval_mse": 0.03575298190116882,
|
75 |
+
"eval_runtime": 1.6208,
|
76 |
+
"eval_samples_per_second": 24.063,
|
77 |
+
"eval_steps_per_second": 3.085,
|
78 |
+
"step": 60
|
79 |
+
},
|
80 |
+
{
|
81 |
+
"epoch": 3.5,
|
82 |
+
"grad_norm": 1.7014061212539673,
|
83 |
+
"learning_rate": 8.25e-05,
|
84 |
+
"loss": 0.0246,
|
85 |
+
"step": 70
|
86 |
+
},
|
87 |
+
{
|
88 |
+
"epoch": 4.0,
|
89 |
+
"grad_norm": 0.39174097776412964,
|
90 |
+
"learning_rate": 8e-05,
|
91 |
+
"loss": 0.0221,
|
92 |
+
"step": 80
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 4.0,
|
96 |
+
"eval_loss": 0.03326353803277016,
|
97 |
+
"eval_mse": 0.03326353803277016,
|
98 |
+
"eval_runtime": 1.6114,
|
99 |
+
"eval_samples_per_second": 24.203,
|
100 |
+
"eval_steps_per_second": 3.103,
|
101 |
+
"step": 80
|
102 |
+
},
|
103 |
+
{
|
104 |
+
"epoch": 4.5,
|
105 |
+
"grad_norm": 2.6145267486572266,
|
106 |
+
"learning_rate": 7.75e-05,
|
107 |
+
"loss": 0.021,
|
108 |
+
"step": 90
|
109 |
+
},
|
110 |
+
{
|
111 |
+
"epoch": 5.0,
|
112 |
+
"grad_norm": 13.847771644592285,
|
113 |
+
"learning_rate": 7.500000000000001e-05,
|
114 |
+
"loss": 0.0223,
|
115 |
+
"step": 100
|
116 |
+
},
|
117 |
+
{
|
118 |
+
"epoch": 5.0,
|
119 |
+
"eval_loss": 0.03398064896464348,
|
120 |
+
"eval_mse": 0.033980656415224075,
|
121 |
+
"eval_runtime": 1.6515,
|
122 |
+
"eval_samples_per_second": 23.615,
|
123 |
+
"eval_steps_per_second": 3.028,
|
124 |
+
"step": 100
|
125 |
+
},
|
126 |
+
{
|
127 |
+
"epoch": 5.5,
|
128 |
+
"grad_norm": 0.43660324811935425,
|
129 |
+
"learning_rate": 7.25e-05,
|
130 |
+
"loss": 0.0131,
|
131 |
+
"step": 110
|
132 |
+
},
|
133 |
+
{
|
134 |
+
"epoch": 6.0,
|
135 |
+
"grad_norm": 2.1774399280548096,
|
136 |
+
"learning_rate": 7e-05,
|
137 |
+
"loss": 0.0117,
|
138 |
+
"step": 120
|
139 |
+
},
|
140 |
+
{
|
141 |
+
"epoch": 6.0,
|
142 |
+
"eval_loss": 0.04457540437579155,
|
143 |
+
"eval_mse": 0.04457540065050125,
|
144 |
+
"eval_runtime": 1.6081,
|
145 |
+
"eval_samples_per_second": 24.252,
|
146 |
+
"eval_steps_per_second": 3.109,
|
147 |
+
"step": 120
|
148 |
+
},
|
149 |
+
{
|
150 |
+
"epoch": 6.5,
|
151 |
+
"grad_norm": 2.61739182472229,
|
152 |
+
"learning_rate": 6.750000000000001e-05,
|
153 |
+
"loss": 0.0168,
|
154 |
+
"step": 130
|
155 |
+
},
|
156 |
+
{
|
157 |
+
"epoch": 7.0,
|
158 |
+
"grad_norm": 2.7107529640197754,
|
159 |
+
"learning_rate": 6.500000000000001e-05,
|
160 |
+
"loss": 0.0107,
|
161 |
+
"step": 140
|
162 |
+
},
|
163 |
+
{
|
164 |
+
"epoch": 7.0,
|
165 |
+
"eval_loss": 0.03702976927161217,
|
166 |
+
"eval_mse": 0.037029776722192764,
|
167 |
+
"eval_runtime": 1.6027,
|
168 |
+
"eval_samples_per_second": 24.334,
|
169 |
+
"eval_steps_per_second": 3.12,
|
170 |
+
"step": 140
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 7.5,
|
174 |
+
"grad_norm": 4.807140350341797,
|
175 |
+
"learning_rate": 6.25e-05,
|
176 |
+
"loss": 0.0112,
|
177 |
+
"step": 150
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 8.0,
|
181 |
+
"grad_norm": 1.6699814796447754,
|
182 |
+
"learning_rate": 6e-05,
|
183 |
+
"loss": 0.0096,
|
184 |
+
"step": 160
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 8.0,
|
188 |
+
"eval_loss": 0.03073795698583126,
|
189 |
+
"eval_mse": 0.030737943947315216,
|
190 |
+
"eval_runtime": 1.678,
|
191 |
+
"eval_samples_per_second": 23.242,
|
192 |
+
"eval_steps_per_second": 2.98,
|
193 |
+
"step": 160
|
194 |
+
},
|
195 |
+
{
|
196 |
+
"epoch": 8.5,
|
197 |
+
"grad_norm": 5.444133281707764,
|
198 |
+
"learning_rate": 5.7499999999999995e-05,
|
199 |
+
"loss": 0.0099,
|
200 |
+
"step": 170
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 9.0,
|
204 |
+
"grad_norm": 1.5312561988830566,
|
205 |
+
"learning_rate": 5.500000000000001e-05,
|
206 |
+
"loss": 0.0142,
|
207 |
+
"step": 180
|
208 |
+
},
|
209 |
+
{
|
210 |
+
"epoch": 9.0,
|
211 |
+
"eval_loss": 0.03504549711942673,
|
212 |
+
"eval_mse": 0.03504551202058792,
|
213 |
+
"eval_runtime": 1.6103,
|
214 |
+
"eval_samples_per_second": 24.218,
|
215 |
+
"eval_steps_per_second": 3.105,
|
216 |
+
"step": 180
|
217 |
+
},
|
218 |
+
{
|
219 |
+
"epoch": 9.5,
|
220 |
+
"grad_norm": 1.527550220489502,
|
221 |
+
"learning_rate": 5.25e-05,
|
222 |
+
"loss": 0.0051,
|
223 |
+
"step": 190
|
224 |
+
},
|
225 |
+
{
|
226 |
+
"epoch": 10.0,
|
227 |
+
"grad_norm": 1.0232219696044922,
|
228 |
+
"learning_rate": 5e-05,
|
229 |
+
"loss": 0.0069,
|
230 |
+
"step": 200
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 10.0,
|
234 |
+
"eval_loss": 0.032399099320173264,
|
235 |
+
"eval_mse": 0.03239908814430237,
|
236 |
+
"eval_runtime": 1.61,
|
237 |
+
"eval_samples_per_second": 24.224,
|
238 |
+
"eval_steps_per_second": 3.106,
|
239 |
+
"step": 200
|
240 |
+
},
|
241 |
+
{
|
242 |
+
"epoch": 10.5,
|
243 |
+
"grad_norm": 1.1013288497924805,
|
244 |
+
"learning_rate": 4.75e-05,
|
245 |
+
"loss": 0.0034,
|
246 |
+
"step": 210
|
247 |
+
},
|
248 |
+
{
|
249 |
+
"epoch": 11.0,
|
250 |
+
"grad_norm": 0.35051777958869934,
|
251 |
+
"learning_rate": 4.5e-05,
|
252 |
+
"loss": 0.0028,
|
253 |
+
"step": 220
|
254 |
+
},
|
255 |
+
{
|
256 |
+
"epoch": 11.0,
|
257 |
+
"eval_loss": 0.02933628484606743,
|
258 |
+
"eval_mse": 0.029336294159293175,
|
259 |
+
"eval_runtime": 1.7012,
|
260 |
+
"eval_samples_per_second": 22.925,
|
261 |
+
"eval_steps_per_second": 2.939,
|
262 |
+
"step": 220
|
263 |
+
},
|
264 |
+
{
|
265 |
+
"epoch": 11.5,
|
266 |
+
"grad_norm": 1.1170843839645386,
|
267 |
+
"learning_rate": 4.25e-05,
|
268 |
+
"loss": 0.0019,
|
269 |
+
"step": 230
|
270 |
+
},
|
271 |
+
{
|
272 |
+
"epoch": 12.0,
|
273 |
+
"grad_norm": 1.3299288749694824,
|
274 |
+
"learning_rate": 4e-05,
|
275 |
+
"loss": 0.0044,
|
276 |
+
"step": 240
|
277 |
+
},
|
278 |
+
{
|
279 |
+
"epoch": 12.0,
|
280 |
+
"eval_loss": 0.028278259560465813,
|
281 |
+
"eval_mse": 0.028278270736336708,
|
282 |
+
"eval_runtime": 1.5914,
|
283 |
+
"eval_samples_per_second": 24.506,
|
284 |
+
"eval_steps_per_second": 3.142,
|
285 |
+
"step": 240
|
286 |
+
},
|
287 |
+
{
|
288 |
+
"epoch": 12.5,
|
289 |
+
"grad_norm": 1.6604584455490112,
|
290 |
+
"learning_rate": 3.7500000000000003e-05,
|
291 |
+
"loss": 0.002,
|
292 |
+
"step": 250
|
293 |
+
},
|
294 |
+
{
|
295 |
+
"epoch": 13.0,
|
296 |
+
"grad_norm": 1.2441127300262451,
|
297 |
+
"learning_rate": 3.5e-05,
|
298 |
+
"loss": 0.0011,
|
299 |
+
"step": 260
|
300 |
+
},
|
301 |
+
{
|
302 |
+
"epoch": 13.0,
|
303 |
+
"eval_loss": 0.029920559376478195,
|
304 |
+
"eval_mse": 0.029920564964413643,
|
305 |
+
"eval_runtime": 1.6282,
|
306 |
+
"eval_samples_per_second": 23.953,
|
307 |
+
"eval_steps_per_second": 3.071,
|
308 |
+
"step": 260
|
309 |
+
},
|
310 |
+
{
|
311 |
+
"epoch": 13.5,
|
312 |
+
"grad_norm": 0.7714802026748657,
|
313 |
+
"learning_rate": 3.2500000000000004e-05,
|
314 |
+
"loss": 0.0008,
|
315 |
+
"step": 270
|
316 |
+
},
|
317 |
+
{
|
318 |
+
"epoch": 14.0,
|
319 |
+
"grad_norm": 0.5498138070106506,
|
320 |
+
"learning_rate": 3e-05,
|
321 |
+
"loss": 0.0005,
|
322 |
+
"step": 280
|
323 |
+
},
|
324 |
+
{
|
325 |
+
"epoch": 14.0,
|
326 |
+
"eval_loss": 0.027942122891545296,
|
327 |
+
"eval_mse": 0.027942117303609848,
|
328 |
+
"eval_runtime": 1.5994,
|
329 |
+
"eval_samples_per_second": 24.384,
|
330 |
+
"eval_steps_per_second": 3.126,
|
331 |
+
"step": 280
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 14.5,
|
335 |
+
"grad_norm": 0.5462870001792908,
|
336 |
+
"learning_rate": 2.7500000000000004e-05,
|
337 |
+
"loss": 0.0006,
|
338 |
+
"step": 290
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 15.0,
|
342 |
+
"grad_norm": 0.32672354578971863,
|
343 |
+
"learning_rate": 2.5e-05,
|
344 |
+
"loss": 0.0005,
|
345 |
+
"step": 300
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 15.0,
|
349 |
+
"eval_loss": 0.029117202386260033,
|
350 |
+
"eval_mse": 0.02911720797419548,
|
351 |
+
"eval_runtime": 1.595,
|
352 |
+
"eval_samples_per_second": 24.451,
|
353 |
+
"eval_steps_per_second": 3.135,
|
354 |
+
"step": 300
|
355 |
+
},
|
356 |
+
{
|
357 |
+
"epoch": 15.5,
|
358 |
+
"grad_norm": 0.7088171243667603,
|
359 |
+
"learning_rate": 2.25e-05,
|
360 |
+
"loss": 0.0012,
|
361 |
+
"step": 310
|
362 |
+
},
|
363 |
+
{
|
364 |
+
"epoch": 16.0,
|
365 |
+
"grad_norm": 0.3224898874759674,
|
366 |
+
"learning_rate": 2e-05,
|
367 |
+
"loss": 0.0011,
|
368 |
+
"step": 320
|
369 |
+
},
|
370 |
+
{
|
371 |
+
"epoch": 16.0,
|
372 |
+
"eval_loss": 0.028802577406167984,
|
373 |
+
"eval_mse": 0.028802569955587387,
|
374 |
+
"eval_runtime": 1.6242,
|
375 |
+
"eval_samples_per_second": 24.012,
|
376 |
+
"eval_steps_per_second": 3.078,
|
377 |
+
"step": 320
|
378 |
+
},
|
379 |
+
{
|
380 |
+
"epoch": 16.5,
|
381 |
+
"grad_norm": 0.2536928951740265,
|
382 |
+
"learning_rate": 1.75e-05,
|
383 |
+
"loss": 0.0002,
|
384 |
+
"step": 330
|
385 |
+
},
|
386 |
+
{
|
387 |
+
"epoch": 17.0,
|
388 |
+
"grad_norm": 0.2693057060241699,
|
389 |
+
"learning_rate": 1.5e-05,
|
390 |
+
"loss": 0.0003,
|
391 |
+
"step": 340
|
392 |
+
},
|
393 |
+
{
|
394 |
+
"epoch": 17.0,
|
395 |
+
"eval_loss": 0.028974896296858788,
|
396 |
+
"eval_mse": 0.02897489443421364,
|
397 |
+
"eval_runtime": 1.5937,
|
398 |
+
"eval_samples_per_second": 24.472,
|
399 |
+
"eval_steps_per_second": 3.137,
|
400 |
+
"step": 340
|
401 |
+
},
|
402 |
+
{
|
403 |
+
"epoch": 17.5,
|
404 |
+
"grad_norm": 0.22231905162334442,
|
405 |
+
"learning_rate": 1.25e-05,
|
406 |
+
"loss": 0.0001,
|
407 |
+
"step": 350
|
408 |
+
},
|
409 |
+
{
|
410 |
+
"epoch": 18.0,
|
411 |
+
"grad_norm": 0.44173210859298706,
|
412 |
+
"learning_rate": 1e-05,
|
413 |
+
"loss": 0.0001,
|
414 |
+
"step": 360
|
415 |
+
},
|
416 |
+
{
|
417 |
+
"epoch": 18.0,
|
418 |
+
"eval_loss": 0.029911378398537636,
|
419 |
+
"eval_mse": 0.029911383986473083,
|
420 |
+
"eval_runtime": 1.5913,
|
421 |
+
"eval_samples_per_second": 24.509,
|
422 |
+
"eval_steps_per_second": 3.142,
|
423 |
+
"step": 360
|
424 |
+
},
|
425 |
+
{
|
426 |
+
"epoch": 18.5,
|
427 |
+
"grad_norm": 0.2958744764328003,
|
428 |
+
"learning_rate": 7.5e-06,
|
429 |
+
"loss": 0.0001,
|
430 |
+
"step": 370
|
431 |
+
},
|
432 |
+
{
|
433 |
+
"epoch": 19.0,
|
434 |
+
"grad_norm": 0.41316938400268555,
|
435 |
+
"learning_rate": 5e-06,
|
436 |
+
"loss": 0.0001,
|
437 |
+
"step": 380
|
438 |
+
},
|
439 |
+
{
|
440 |
+
"epoch": 19.0,
|
441 |
+
"eval_loss": 0.029724078252911568,
|
442 |
+
"eval_mse": 0.029724083840847015,
|
443 |
+
"eval_runtime": 1.5907,
|
444 |
+
"eval_samples_per_second": 24.518,
|
445 |
+
"eval_steps_per_second": 3.143,
|
446 |
+
"step": 380
|
447 |
+
},
|
448 |
+
{
|
449 |
+
"epoch": 19.5,
|
450 |
+
"grad_norm": 0.0391409695148468,
|
451 |
+
"learning_rate": 2.5e-06,
|
452 |
+
"loss": 0.0,
|
453 |
+
"step": 390
|
454 |
+
},
|
455 |
+
{
|
456 |
+
"epoch": 20.0,
|
457 |
+
"grad_norm": 0.021498844027519226,
|
458 |
+
"learning_rate": 0.0,
|
459 |
+
"loss": 0.0,
|
460 |
+
"step": 400
|
461 |
+
}
|
462 |
+
],
|
463 |
+
"logging_steps": 10,
|
464 |
+
"max_steps": 400,
|
465 |
+
"num_input_tokens_seen": 0,
|
466 |
+
"num_train_epochs": 20,
|
467 |
+
"save_steps": 10,
|
468 |
+
"stateful_callbacks": {
|
469 |
+
"TrainerControl": {
|
470 |
+
"args": {
|
471 |
+
"should_epoch_stop": false,
|
472 |
+
"should_evaluate": false,
|
473 |
+
"should_log": false,
|
474 |
+
"should_save": true,
|
475 |
+
"should_training_stop": true
|
476 |
+
},
|
477 |
+
"attributes": {}
|
478 |
+
}
|
479 |
+
},
|
480 |
+
"total_flos": 0.0,
|
481 |
+
"train_batch_size": 8,
|
482 |
+
"trial_name": null,
|
483 |
+
"trial_params": null
|
484 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a19a6489ee5ffde1427142bcbf54dbb7e81a85205c07442cc69c6f9847f8b45e
|
3 |
+
size 5048
|