AlekseyKorshuk
commited on
Commit
•
79fe28f
1
Parent(s):
cdc75a4
huggingartists
Browse files- README.md +3 -3
- config.json +2 -2
- evaluation.txt +1 -1
- flax_model.msgpack +1 -1
- optimizer.pt +1 -1
- pytorch_model.bin +1 -1
- rng_state.pth +1 -1
- scheduler.pt +1 -1
- tokenizer_config.json +1 -1
- trainer_state.json +113 -7
- training_args.bin +1 -1
README.md
CHANGED
@@ -45,15 +45,15 @@ from datasets import load_dataset
|
|
45 |
dataset = load_dataset("huggingartists/linkin-park")
|
46 |
```
|
47 |
|
48 |
-
[Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/
|
49 |
|
50 |
## Training procedure
|
51 |
|
52 |
The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on Linkin Park's lyrics.
|
53 |
|
54 |
-
Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/
|
55 |
|
56 |
-
At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/
|
57 |
|
58 |
## How to use
|
59 |
|
|
|
45 |
dataset = load_dataset("huggingartists/linkin-park")
|
46 |
```
|
47 |
|
48 |
+
[Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/1s6xt77z/artifacts), which is tracked with [W&B artifacts](https://docs.wandb.com/artifacts) at every step of the pipeline.
|
49 |
|
50 |
## Training procedure
|
51 |
|
52 |
The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on Linkin Park's lyrics.
|
53 |
|
54 |
+
Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/1i4je18p) for full transparency and reproducibility.
|
55 |
|
56 |
+
At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/1i4je18p/artifacts) is logged and versioned.
|
57 |
|
58 |
## How to use
|
59 |
|
config.json
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
{
|
2 |
-
"_name_or_path": "
|
3 |
"activation_function": "gelu_new",
|
4 |
"architectures": [
|
5 |
"GPT2LMHeadModel"
|
@@ -35,7 +35,7 @@
|
|
35 |
}
|
36 |
},
|
37 |
"torch_dtype": "float32",
|
38 |
-
"transformers_version": "4.10.
|
39 |
"use_cache": true,
|
40 |
"vocab_size": 50257
|
41 |
}
|
|
|
1 |
{
|
2 |
+
"_name_or_path": "huggingartists/linkin-park",
|
3 |
"activation_function": "gelu_new",
|
4 |
"architectures": [
|
5 |
"GPT2LMHeadModel"
|
|
|
35 |
}
|
36 |
},
|
37 |
"torch_dtype": "float32",
|
38 |
+
"transformers_version": "4.10.2",
|
39 |
"use_cache": true,
|
40 |
"vocab_size": 50257
|
41 |
}
|
evaluation.txt
CHANGED
@@ -1 +1 @@
|
|
1 |
-
{"eval_loss": 2.
|
|
|
1 |
+
{"eval_loss": 2.248004913330078, "eval_runtime": 4.6644, "eval_samples_per_second": 20.796, "eval_steps_per_second": 2.787, "epoch": 2.0}
|
flax_model.msgpack
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 497764120
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:225d151b89706812d44a3b7db2ec9b070074c0f7ba7e9d0ffe24145cc0990adb
|
3 |
size 497764120
|
optimizer.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 995603825
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:35bd95469cbf0ab080be66768e5242378073c1190b28bd15f2700cf295d526f6
|
3 |
size 995603825
|
pytorch_model.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 510403817
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8a3eb8db73abd6dfffd1f85d4cbb59cf5647bfbe01d3c02e21a5a72a96906a3a
|
3 |
size 510403817
|
rng_state.pth
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 14503
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:db7af3f0fec071276d1a4d8600845adbb6ac8e80806a422d4311d79890098c71
|
3 |
size 14503
|
scheduler.pt
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 623
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:03b8c59abd7807eb31fbb6b182b880936157d534c8d17e22038fc44702f9354e
|
3 |
size 623
|
tokenizer_config.json
CHANGED
@@ -1 +1 @@
|
|
1 |
-
{"unk_token": "<|endoftext|>", "bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "add_prefix_space": false, "model_max_length": 1024, "special_tokens_map_file": null, "name_or_path": "
|
|
|
1 |
+
{"unk_token": "<|endoftext|>", "bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "add_prefix_space": false, "model_max_length": 1024, "special_tokens_map_file": null, "name_or_path": "huggingartists/linkin-park", "tokenizer_class": "GPT2Tokenizer"}
|
trainer_state.json
CHANGED
@@ -1,8 +1,8 @@
|
|
1 |
{
|
2 |
-
"best_metric": 2.
|
3 |
-
"best_model_checkpoint": "output/linkin-park/checkpoint-
|
4 |
-
"epoch":
|
5 |
-
"global_step":
|
6 |
"is_hyper_param_search": false,
|
7 |
"is_local_process_zero": true,
|
8 |
"is_world_process_zero": true,
|
@@ -92,11 +92,117 @@
|
|
92 |
"eval_samples_per_second": 22.041,
|
93 |
"eval_steps_per_second": 2.782,
|
94 |
"step": 69
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
95 |
}
|
96 |
],
|
97 |
-
"max_steps":
|
98 |
-
"num_train_epochs":
|
99 |
-
"total_flos":
|
100 |
"trial_name": null,
|
101 |
"trial_params": null
|
102 |
}
|
|
|
1 |
{
|
2 |
+
"best_metric": 2.248004913330078,
|
3 |
+
"best_model_checkpoint": "output/linkin-park/checkpoint-140",
|
4 |
+
"epoch": 2.0,
|
5 |
+
"global_step": 140,
|
6 |
"is_hyper_param_search": false,
|
7 |
"is_local_process_zero": true,
|
8 |
"is_world_process_zero": true,
|
|
|
92 |
"eval_samples_per_second": 22.041,
|
93 |
"eval_steps_per_second": 2.782,
|
94 |
"step": 69
|
95 |
+
},
|
96 |
+
{
|
97 |
+
"epoch": 1.0,
|
98 |
+
"learning_rate": 0.0,
|
99 |
+
"loss": 1.9438,
|
100 |
+
"step": 70
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 1.0,
|
104 |
+
"eval_loss": 2.463841438293457,
|
105 |
+
"eval_runtime": 4.6348,
|
106 |
+
"eval_samples_per_second": 20.929,
|
107 |
+
"eval_steps_per_second": 2.805,
|
108 |
+
"step": 70
|
109 |
+
},
|
110 |
+
{
|
111 |
+
"epoch": 1.07,
|
112 |
+
"learning_rate": 1.7199452243268996e-06,
|
113 |
+
"loss": 2.4952,
|
114 |
+
"step": 75
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 1.14,
|
118 |
+
"learning_rate": 6.793535661894047e-06,
|
119 |
+
"loss": 2.5481,
|
120 |
+
"step": 80
|
121 |
+
},
|
122 |
+
{
|
123 |
+
"epoch": 1.21,
|
124 |
+
"learning_rate": 1.4966360302693133e-05,
|
125 |
+
"loss": 2.6492,
|
126 |
+
"step": 85
|
127 |
+
},
|
128 |
+
{
|
129 |
+
"epoch": 1.29,
|
130 |
+
"learning_rate": 2.582859959249087e-05,
|
131 |
+
"loss": 2.3078,
|
132 |
+
"step": 90
|
133 |
+
},
|
134 |
+
{
|
135 |
+
"epoch": 1.36,
|
136 |
+
"learning_rate": 3.88355754965355e-05,
|
137 |
+
"loss": 2.6034,
|
138 |
+
"step": 95
|
139 |
+
},
|
140 |
+
{
|
141 |
+
"epoch": 1.43,
|
142 |
+
"learning_rate": 5.333506393059682e-05,
|
143 |
+
"loss": 2.4777,
|
144 |
+
"step": 100
|
145 |
+
},
|
146 |
+
{
|
147 |
+
"epoch": 1.5,
|
148 |
+
"learning_rate": 6.859999999999999e-05,
|
149 |
+
"loss": 2.2842,
|
150 |
+
"step": 105
|
151 |
+
},
|
152 |
+
{
|
153 |
+
"epoch": 1.57,
|
154 |
+
"learning_rate": 8.386493606940316e-05,
|
155 |
+
"loss": 2.577,
|
156 |
+
"step": 110
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 1.64,
|
160 |
+
"learning_rate": 9.836442450346448e-05,
|
161 |
+
"loss": 2.3194,
|
162 |
+
"step": 115
|
163 |
+
},
|
164 |
+
{
|
165 |
+
"epoch": 1.71,
|
166 |
+
"learning_rate": 0.00011137140040750911,
|
167 |
+
"loss": 2.3364,
|
168 |
+
"step": 120
|
169 |
+
},
|
170 |
+
{
|
171 |
+
"epoch": 1.79,
|
172 |
+
"learning_rate": 0.00012223363969730684,
|
173 |
+
"loss": 2.3984,
|
174 |
+
"step": 125
|
175 |
+
},
|
176 |
+
{
|
177 |
+
"epoch": 1.86,
|
178 |
+
"learning_rate": 0.00013040646433810595,
|
179 |
+
"loss": 2.2814,
|
180 |
+
"step": 130
|
181 |
+
},
|
182 |
+
{
|
183 |
+
"epoch": 1.93,
|
184 |
+
"learning_rate": 0.0001354800547756731,
|
185 |
+
"loss": 2.2059,
|
186 |
+
"step": 135
|
187 |
+
},
|
188 |
+
{
|
189 |
+
"epoch": 2.0,
|
190 |
+
"learning_rate": 0.0001372,
|
191 |
+
"loss": 1.9945,
|
192 |
+
"step": 140
|
193 |
+
},
|
194 |
+
{
|
195 |
+
"epoch": 2.0,
|
196 |
+
"eval_loss": 2.248004913330078,
|
197 |
+
"eval_runtime": 4.6064,
|
198 |
+
"eval_samples_per_second": 21.058,
|
199 |
+
"eval_steps_per_second": 2.822,
|
200 |
+
"step": 140
|
201 |
}
|
202 |
],
|
203 |
+
"max_steps": 140,
|
204 |
+
"num_train_epochs": 2,
|
205 |
+
"total_flos": 144625139712000.0,
|
206 |
"trial_name": null,
|
207 |
"trial_params": null
|
208 |
}
|
training_args.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 2671
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3c394c879763a7ac8c43760c1791f5a9aee9dc1f21c84e74abc5a6b80af9588a
|
3 |
size 2671
|