Commit
•
758a5a8
1
Parent(s):
c787ad4
Training in progress, step 500
Browse files- .gitattributes +1 -0
- .gitignore +1 -0
- config.json +285 -0
- merges.txt +0 -0
- preprocessor_config.json +9 -0
- pytorch_model.bin +3 -0
- run.sh +1 -1
- run_xtreme_s.py +27 -2
- runs/May03_12-39-14_sanchit--v100/1651584502.8814862/events.out.tfevents.1651584502.sanchit--v100.38894.1 +3 -0
- runs/May03_12-39-14_sanchit--v100/events.out.tfevents.1651584502.sanchit--v100.38894.0 +3 -0
- runs/May03_15-33-55_sanchit--v100/1651592261.409201/events.out.tfevents.1651592261.sanchit--v100.40140.1 +3 -0
- runs/May03_15-33-55_sanchit--v100/events.out.tfevents.1651592261.sanchit--v100.40140.0 +3 -0
- runs/May03_15-58-13_sanchit--v100/1651596054.9030492/events.out.tfevents.1651596054.sanchit--v100.40903.1 +3 -0
- runs/May03_15-58-13_sanchit--v100/events.out.tfevents.1651596054.sanchit--v100.40903.0 +3 -0
- runs/May03_17-15-22_sanchit--v100/1651598399.7088904/events.out.tfevents.1651598399.sanchit--v100.42111.1 +3 -0
- runs/May03_17-15-22_sanchit--v100/events.out.tfevents.1651598399.sanchit--v100.42111.0 +3 -0
- special_tokens_map.json +1 -0
- sweep.yaml +2 -1
- tokenizer.json +0 -0
- tokenizer_config.json +1 -0
- training_args.bin +3 -0
- vocab.json +0 -0
- wandb/debug-cli.log +155 -0
- wandb/debug-internal.log +1 -0
- wandb/debug.log +1 -0
- wandb/latest-run +1 -0
- wandb/run-20220503_171959-a6039xud/files/config.yaml +0 -0
- wandb/run-20220503_171959-a6039xud/files/output.log +0 -0
- wandb/run-20220503_171959-a6039xud/files/requirements.txt +287 -0
- wandb/run-20220503_171959-a6039xud/files/wandb-metadata.json +57 -0
- wandb/run-20220503_171959-a6039xud/files/wandb-summary.json +0 -0
- wandb/run-20220503_171959-a6039xud/logs/debug-internal.log +0 -0
- wandb/run-20220503_171959-a6039xud/logs/debug.log +28 -0
- wandb/run-20220503_171959-a6039xud/run-a6039xud.wandb +3 -0
- wandb/sweep-39ci3gkf/config-a6039xud.yaml +44 -0
- wandb/sweep-y3ak427l/config-irggvkgd.yaml +44 -0
- wandb/sweep-y3ak427l/config-ldsojzle.yaml +44 -0
- wandb/sweep-y3ak427l/config-qv3vjr6j.yaml +44 -0
- wandb/sweep-y3ak427l/config-vz5ppd75.yaml +44 -0
- wandb/sweep-y3ak427l/config-xur584bd.yaml +44 -0
.gitattributes
CHANGED
@@ -26,3 +26,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
26 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
27 |
*.zstandard filter=lfs diff=lfs merge=lfs -text
|
28 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
26 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
27 |
*.zstandard filter=lfs diff=lfs merge=lfs -text
|
28 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
29 |
+
wandb/run-20220503_171959-a6039xud/run-a6039xud.wandb filter=lfs diff=lfs merge=lfs -text
|
.gitignore
ADDED
@@ -0,0 +1 @@
|
|
|
1 |
+
checkpoint-*/
|
config.json
ADDED
@@ -0,0 +1,285 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "./",
|
3 |
+
"architectures": [
|
4 |
+
"SpeechEncoderDecoderModel"
|
5 |
+
],
|
6 |
+
"decoder": {
|
7 |
+
"_name_or_path": "facebook/bart-large",
|
8 |
+
"activation_dropout": 0.1,
|
9 |
+
"activation_function": "gelu",
|
10 |
+
"add_bias_logits": false,
|
11 |
+
"add_cross_attention": true,
|
12 |
+
"add_final_layer_norm": false,
|
13 |
+
"architectures": [
|
14 |
+
"BartModel"
|
15 |
+
],
|
16 |
+
"attention_dropout": 0.1,
|
17 |
+
"bad_words_ids": null,
|
18 |
+
"bos_token_id": 0,
|
19 |
+
"chunk_size_feed_forward": 0,
|
20 |
+
"classif_dropout": 0.1,
|
21 |
+
"classifier_dropout": 0.0,
|
22 |
+
"cross_attention_hidden_size": null,
|
23 |
+
"d_model": 1024,
|
24 |
+
"decoder_attention_heads": 16,
|
25 |
+
"decoder_ffn_dim": 4096,
|
26 |
+
"decoder_layerdrop": 0.0,
|
27 |
+
"decoder_layers": 12,
|
28 |
+
"decoder_start_token_id": 2,
|
29 |
+
"diversity_penalty": 0.0,
|
30 |
+
"do_sample": false,
|
31 |
+
"dropout": 0.1,
|
32 |
+
"early_stopping": true,
|
33 |
+
"encoder_attention_heads": 16,
|
34 |
+
"encoder_ffn_dim": 4096,
|
35 |
+
"encoder_layerdrop": 0.0,
|
36 |
+
"encoder_layers": 12,
|
37 |
+
"encoder_no_repeat_ngram_size": 0,
|
38 |
+
"eos_token_id": 2,
|
39 |
+
"exponential_decay_length_penalty": null,
|
40 |
+
"finetuning_task": null,
|
41 |
+
"forced_bos_token_id": 0,
|
42 |
+
"forced_eos_token_id": 2,
|
43 |
+
"gradient_checkpointing": false,
|
44 |
+
"id2label": {
|
45 |
+
"0": "LABEL_0",
|
46 |
+
"1": "LABEL_1",
|
47 |
+
"2": "LABEL_2"
|
48 |
+
},
|
49 |
+
"init_std": 0.02,
|
50 |
+
"is_decoder": true,
|
51 |
+
"is_encoder_decoder": false,
|
52 |
+
"label2id": {
|
53 |
+
"LABEL_0": 0,
|
54 |
+
"LABEL_1": 1,
|
55 |
+
"LABEL_2": 2
|
56 |
+
},
|
57 |
+
"length_penalty": 1.0,
|
58 |
+
"max_length": 20,
|
59 |
+
"max_position_embeddings": 1024,
|
60 |
+
"min_length": 0,
|
61 |
+
"model_type": "bart",
|
62 |
+
"no_repeat_ngram_size": 3,
|
63 |
+
"normalize_before": false,
|
64 |
+
"num_beam_groups": 1,
|
65 |
+
"num_beams": 4,
|
66 |
+
"num_hidden_layers": 12,
|
67 |
+
"num_return_sequences": 1,
|
68 |
+
"output_attentions": false,
|
69 |
+
"output_hidden_states": false,
|
70 |
+
"output_scores": false,
|
71 |
+
"pad_token_id": 1,
|
72 |
+
"prefix": null,
|
73 |
+
"problem_type": null,
|
74 |
+
"pruned_heads": {},
|
75 |
+
"remove_invalid_values": false,
|
76 |
+
"repetition_penalty": 1.0,
|
77 |
+
"return_dict": true,
|
78 |
+
"return_dict_in_generate": false,
|
79 |
+
"scale_embedding": false,
|
80 |
+
"sep_token_id": null,
|
81 |
+
"task_specific_params": {
|
82 |
+
"summarization": {
|
83 |
+
"length_penalty": 1.0,
|
84 |
+
"max_length": 128,
|
85 |
+
"min_length": 12,
|
86 |
+
"num_beams": 4
|
87 |
+
},
|
88 |
+
"summarization_cnn": {
|
89 |
+
"length_penalty": 2.0,
|
90 |
+
"max_length": 142,
|
91 |
+
"min_length": 56,
|
92 |
+
"num_beams": 4
|
93 |
+
},
|
94 |
+
"summarization_xsum": {
|
95 |
+
"length_penalty": 1.0,
|
96 |
+
"max_length": 62,
|
97 |
+
"min_length": 11,
|
98 |
+
"num_beams": 6
|
99 |
+
}
|
100 |
+
},
|
101 |
+
"temperature": 1.0,
|
102 |
+
"tie_encoder_decoder": false,
|
103 |
+
"tie_word_embeddings": true,
|
104 |
+
"tokenizer_class": null,
|
105 |
+
"top_k": 50,
|
106 |
+
"top_p": 1.0,
|
107 |
+
"torch_dtype": null,
|
108 |
+
"torchscript": false,
|
109 |
+
"transformers_version": "4.19.0.dev0",
|
110 |
+
"typical_p": 1.0,
|
111 |
+
"use_bfloat16": false,
|
112 |
+
"use_cache": true,
|
113 |
+
"vocab_size": 50265
|
114 |
+
},
|
115 |
+
"decoder_start_token_id": 0,
|
116 |
+
"encoder": {
|
117 |
+
"_name_or_path": "facebook/wav2vec2-xls-r-300m",
|
118 |
+
"activation_dropout": 0.0,
|
119 |
+
"adapter_kernel_size": 3,
|
120 |
+
"adapter_stride": 2,
|
121 |
+
"add_adapter": true,
|
122 |
+
"add_cross_attention": false,
|
123 |
+
"apply_spec_augment": true,
|
124 |
+
"architectures": [
|
125 |
+
"Wav2Vec2ForPreTraining"
|
126 |
+
],
|
127 |
+
"attention_dropout": 0.1,
|
128 |
+
"bad_words_ids": null,
|
129 |
+
"bos_token_id": 1,
|
130 |
+
"chunk_size_feed_forward": 0,
|
131 |
+
"classifier_proj_size": 256,
|
132 |
+
"codevector_dim": 768,
|
133 |
+
"contrastive_logits_temperature": 0.1,
|
134 |
+
"conv_bias": true,
|
135 |
+
"conv_dim": [
|
136 |
+
512,
|
137 |
+
512,
|
138 |
+
512,
|
139 |
+
512,
|
140 |
+
512,
|
141 |
+
512,
|
142 |
+
512
|
143 |
+
],
|
144 |
+
"conv_kernel": [
|
145 |
+
10,
|
146 |
+
3,
|
147 |
+
3,
|
148 |
+
3,
|
149 |
+
3,
|
150 |
+
2,
|
151 |
+
2
|
152 |
+
],
|
153 |
+
"conv_stride": [
|
154 |
+
5,
|
155 |
+
2,
|
156 |
+
2,
|
157 |
+
2,
|
158 |
+
2,
|
159 |
+
2,
|
160 |
+
2
|
161 |
+
],
|
162 |
+
"cross_attention_hidden_size": null,
|
163 |
+
"ctc_loss_reduction": "sum",
|
164 |
+
"ctc_zero_infinity": false,
|
165 |
+
"decoder_start_token_id": null,
|
166 |
+
"diversity_loss_weight": 0.1,
|
167 |
+
"diversity_penalty": 0.0,
|
168 |
+
"do_sample": false,
|
169 |
+
"do_stable_layer_norm": true,
|
170 |
+
"early_stopping": false,
|
171 |
+
"encoder_no_repeat_ngram_size": 0,
|
172 |
+
"eos_token_id": 2,
|
173 |
+
"exponential_decay_length_penalty": null,
|
174 |
+
"feat_extract_activation": "gelu",
|
175 |
+
"feat_extract_dropout": 0.0,
|
176 |
+
"feat_extract_norm": "layer",
|
177 |
+
"feat_proj_dropout": 0.0,
|
178 |
+
"feat_quantizer_dropout": 0.0,
|
179 |
+
"final_dropout": 0.0,
|
180 |
+
"finetuning_task": null,
|
181 |
+
"forced_bos_token_id": null,
|
182 |
+
"forced_eos_token_id": null,
|
183 |
+
"gradient_checkpointing": false,
|
184 |
+
"hidden_act": "gelu",
|
185 |
+
"hidden_dropout": 0.06862889720223829,
|
186 |
+
"hidden_size": 1024,
|
187 |
+
"id2label": {
|
188 |
+
"0": "LABEL_0",
|
189 |
+
"1": "LABEL_1"
|
190 |
+
},
|
191 |
+
"initializer_range": 0.02,
|
192 |
+
"intermediate_size": 4096,
|
193 |
+
"is_decoder": false,
|
194 |
+
"is_encoder_decoder": false,
|
195 |
+
"label2id": {
|
196 |
+
"LABEL_0": 0,
|
197 |
+
"LABEL_1": 1
|
198 |
+
},
|
199 |
+
"layer_norm_eps": 1e-05,
|
200 |
+
"layerdrop": 0.0,
|
201 |
+
"length_penalty": 1.0,
|
202 |
+
"mask_feature_length": 10,
|
203 |
+
"mask_feature_min_masks": 0,
|
204 |
+
"mask_feature_prob": 0.0,
|
205 |
+
"mask_time_length": 10,
|
206 |
+
"mask_time_min_masks": 2,
|
207 |
+
"mask_time_prob": 0.1,
|
208 |
+
"max_length": 20,
|
209 |
+
"min_length": 0,
|
210 |
+
"model_type": "wav2vec2",
|
211 |
+
"no_repeat_ngram_size": 0,
|
212 |
+
"num_adapter_layers": 3,
|
213 |
+
"num_attention_heads": 16,
|
214 |
+
"num_beam_groups": 1,
|
215 |
+
"num_beams": 1,
|
216 |
+
"num_codevector_groups": 2,
|
217 |
+
"num_codevectors_per_group": 320,
|
218 |
+
"num_conv_pos_embedding_groups": 16,
|
219 |
+
"num_conv_pos_embeddings": 128,
|
220 |
+
"num_feat_extract_layers": 7,
|
221 |
+
"num_hidden_layers": 24,
|
222 |
+
"num_negatives": 100,
|
223 |
+
"num_return_sequences": 1,
|
224 |
+
"output_attentions": false,
|
225 |
+
"output_hidden_size": 1024,
|
226 |
+
"output_hidden_states": false,
|
227 |
+
"output_scores": false,
|
228 |
+
"pad_token_id": 0,
|
229 |
+
"prefix": null,
|
230 |
+
"problem_type": null,
|
231 |
+
"proj_codevector_dim": 768,
|
232 |
+
"pruned_heads": {},
|
233 |
+
"remove_invalid_values": false,
|
234 |
+
"repetition_penalty": 1.0,
|
235 |
+
"return_dict": true,
|
236 |
+
"return_dict_in_generate": false,
|
237 |
+
"sep_token_id": null,
|
238 |
+
"task_specific_params": null,
|
239 |
+
"tdnn_dilation": [
|
240 |
+
1,
|
241 |
+
2,
|
242 |
+
3,
|
243 |
+
1,
|
244 |
+
1
|
245 |
+
],
|
246 |
+
"tdnn_dim": [
|
247 |
+
512,
|
248 |
+
512,
|
249 |
+
512,
|
250 |
+
512,
|
251 |
+
1500
|
252 |
+
],
|
253 |
+
"tdnn_kernel": [
|
254 |
+
5,
|
255 |
+
3,
|
256 |
+
3,
|
257 |
+
1,
|
258 |
+
1
|
259 |
+
],
|
260 |
+
"temperature": 1.0,
|
261 |
+
"tie_encoder_decoder": false,
|
262 |
+
"tie_word_embeddings": true,
|
263 |
+
"tokenizer_class": null,
|
264 |
+
"top_k": 50,
|
265 |
+
"top_p": 1.0,
|
266 |
+
"torch_dtype": "float32",
|
267 |
+
"torchscript": false,
|
268 |
+
"transformers_version": "4.19.0.dev0",
|
269 |
+
"typical_p": 1.0,
|
270 |
+
"use_bfloat16": false,
|
271 |
+
"use_weighted_layer_sum": false,
|
272 |
+
"vocab_size": 32,
|
273 |
+
"xvector_output_dim": 512
|
274 |
+
},
|
275 |
+
"eos_token_id": 2,
|
276 |
+
"is_encoder_decoder": true,
|
277 |
+
"max_length": 40,
|
278 |
+
"model_type": "speech-encoder-decoder",
|
279 |
+
"pad_token_id": 1,
|
280 |
+
"processor_class": "Wav2Vec2Processor",
|
281 |
+
"tie_word_embeddings": false,
|
282 |
+
"torch_dtype": "float32",
|
283 |
+
"transformers_version": null,
|
284 |
+
"use_cache": false
|
285 |
+
}
|
merges.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
preprocessor_config.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"do_normalize": true,
|
3 |
+
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
|
4 |
+
"feature_size": 1,
|
5 |
+
"padding_side": "right",
|
6 |
+
"padding_value": 0,
|
7 |
+
"return_attention_mask": true,
|
8 |
+
"sampling_rate": 16000
|
9 |
+
}
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:008e06044c5250309092dc773bd06898ae5bf91e5bddf24e216307b3e34cbee0
|
3 |
+
size 2353867057
|
run.sh
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
#!/usr/bin/env bash
|
2 |
-
CUDA_VISIBLE_DEVICES=
|
3 |
--model_name_or_path="./" \
|
4 |
--task="covost2" \
|
5 |
--language="fr.en" \
|
1 |
#!/usr/bin/env bash
|
2 |
+
CUDA_VISIBLE_DEVICES=0 python run_xtreme_s.py \
|
3 |
--model_name_or_path="./" \
|
4 |
--task="covost2" \
|
5 |
--language="fr.en" \
|
run_xtreme_s.py
CHANGED
@@ -41,6 +41,7 @@ from transformers import (
|
|
41 |
HfArgumentParser,
|
42 |
Seq2SeqTrainer,
|
43 |
Seq2SeqTrainingArguments,
|
|
|
44 |
Trainer,
|
45 |
set_seed,
|
46 |
)
|
@@ -547,9 +548,9 @@ def main():
|
|
547 |
|
548 |
def remove_special_characters(batch):
|
549 |
if chars_to_ignore_regex is not None:
|
550 |
-
batch["target_text"] = re.sub(chars_to_ignore_regex, "", batch[target_column_name]).lower()
|
551 |
else:
|
552 |
-
batch["target_text"] = batch[target_column_name].lower()
|
553 |
return batch
|
554 |
|
555 |
if is_text_target:
|
@@ -565,6 +566,30 @@ def main():
|
|
565 |
unk_token = data_args.unk_token
|
566 |
pad_token = data_args.pad_token
|
567 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
568 |
# 3. Next, let's load the config as we might need it to create
|
569 |
# the tokenizer
|
570 |
config = AutoConfig.from_pretrained(
|
41 |
HfArgumentParser,
|
42 |
Seq2SeqTrainer,
|
43 |
Seq2SeqTrainingArguments,
|
44 |
+
SpeechEncoderDecoderModel,
|
45 |
Trainer,
|
46 |
set_seed,
|
47 |
)
|
548 |
|
549 |
def remove_special_characters(batch):
|
550 |
if chars_to_ignore_regex is not None:
|
551 |
+
batch["target_text"] = re.sub(chars_to_ignore_regex, "", batch[target_column_name]).lower()
|
552 |
else:
|
553 |
+
batch["target_text"] = batch[target_column_name].lower()
|
554 |
return batch
|
555 |
|
556 |
if is_text_target:
|
566 |
unk_token = data_args.unk_token
|
567 |
pad_token = data_args.pad_token
|
568 |
|
569 |
+
|
570 |
+
encoder_id = "facebook/wav2vec2-xls-r-300m"
|
571 |
+
decoder_id = "facebook/bart-large"
|
572 |
+
|
573 |
+
model = SpeechEncoderDecoderModel.from_encoder_decoder_pretrained(encoder_id, decoder_id, encoder_add_adapter=True)
|
574 |
+
model.config.encoder.feat_proj_dropout = 0.0
|
575 |
+
model.config.encoder.final_dropout = 0.0
|
576 |
+
model.config.encoder.mask_time_prob = 0.1
|
577 |
+
model.config.decoder_start_token_id = model.decoder.config.bos_token_id
|
578 |
+
model.config.pad_token_id = model.decoder.config.pad_token_id
|
579 |
+
model.config.eos_token_id = model.decoder.config.eos_token_id
|
580 |
+
model.config.max_length = 40
|
581 |
+
model.config.num_beams = 1
|
582 |
+
model.config.encoder.layerdrop = 0.0
|
583 |
+
model.config.use_cache = False
|
584 |
+
model.config.processor_class = "Wav2Vec2Processor"
|
585 |
+
|
586 |
+
model.save_pretrained(model_args.model_name_or_path)
|
587 |
+
|
588 |
+
feature_etxractor = AutoFeatureExtractor.from_pretrained(encoder_id)
|
589 |
+
feature_etxractor.save_pretrained(model_args.model_name_or_path)
|
590 |
+
tokenizer = AutoTokenizer.from_pretrained(decoder_id)
|
591 |
+
tokenizer.save_pretrained(model_args.model_name_or_path)
|
592 |
+
|
593 |
# 3. Next, let's load the config as we might need it to create
|
594 |
# the tokenizer
|
595 |
config = AutoConfig.from_pretrained(
|
runs/May03_12-39-14_sanchit--v100/1651584502.8814862/events.out.tfevents.1651584502.sanchit--v100.38894.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1a5e50bb326f116678bb0d2af94ee25e92d82ce3268121a68b288dd930e13470
|
3 |
+
size 5184
|
runs/May03_12-39-14_sanchit--v100/events.out.tfevents.1651584502.sanchit--v100.38894.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:65a28e37ffdeeb562c734e7a4815296918b93b818763a73a70ab20d29627105c
|
3 |
+
size 10592
|
runs/May03_15-33-55_sanchit--v100/1651592261.409201/events.out.tfevents.1651592261.sanchit--v100.40140.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7c3651206b236f0c31a3ff08bbeb2416a7fde03f54ccb583a7df7f8256f3fd29
|
3 |
+
size 5184
|
runs/May03_15-33-55_sanchit--v100/events.out.tfevents.1651592261.sanchit--v100.40140.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c965accf9ea2be03062bfecc4c45cdb54eb6b4b444ea37a2c9533c8db4fed044
|
3 |
+
size 87941
|
runs/May03_15-58-13_sanchit--v100/1651596054.9030492/events.out.tfevents.1651596054.sanchit--v100.40903.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:76e39c8c39b9be3c4531fd72d066671f7365b550e8079fee72b6060a36be59f6
|
3 |
+
size 5184
|
runs/May03_15-58-13_sanchit--v100/events.out.tfevents.1651596054.sanchit--v100.40903.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:410298c8949435d86a040fb00c0e0a7bde2f0445544abeadadae587ba569bfba
|
3 |
+
size 73654
|
runs/May03_17-15-22_sanchit--v100/1651598399.7088904/events.out.tfevents.1651598399.sanchit--v100.42111.1
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:61b94cd769015be4bbc81b78b8074c963d947a90c91e7cf517abaad8c3ff3078
|
3 |
+
size 5184
|
runs/May03_17-15-22_sanchit--v100/events.out.tfevents.1651598399.sanchit--v100.42111.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:031a66a2aa473ef2b64f357fd6f9159fb836bed0b5c5036f7eb9c55c533a7ad7
|
3 |
+
size 88290
|
special_tokens_map.json
ADDED
@@ -0,0 +1 @@
|
|
|
1 |
+
{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": false}}
|
sweep.yaml
CHANGED
@@ -4,6 +4,7 @@ command:
|
|
4 |
- --overwrite_output_dir
|
5 |
- --freeze_feature_encoder
|
6 |
- --gradient_checkpointing
|
|
|
7 |
- --fp16
|
8 |
- --group_by_length
|
9 |
- --do_train
|
@@ -26,7 +27,7 @@ parameters:
|
|
26 |
eval_split_name:
|
27 |
value: test
|
28 |
output_dir:
|
29 |
-
value: ./
|
30 |
num_train_epochs:
|
31 |
value: 3
|
32 |
per_device_train_batch_size:
|
4 |
- --overwrite_output_dir
|
5 |
- --freeze_feature_encoder
|
6 |
- --gradient_checkpointing
|
7 |
+
- --predict_with_generate
|
8 |
- --fp16
|
9 |
- --group_by_length
|
10 |
- --do_train
|
27 |
eval_split_name:
|
28 |
value: test
|
29 |
output_dir:
|
30 |
+
value: ./
|
31 |
num_train_epochs:
|
32 |
value: 3
|
33 |
per_device_train_batch_size:
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
tokenizer_config.json
ADDED
@@ -0,0 +1 @@
|
|
|
1 |
+
{"errors": "replace", "bos_token": "<s>", "eos_token": "</s>", "sep_token": "</s>", "cls_token": "<s>", "unk_token": "<unk>", "pad_token": "<pad>", "mask_token": "<mask>", "add_prefix_space": false, "trim_offsets": true, "model_max_length": 1024, "special_tokens_map_file": null, "name_or_path": "./", "tokenizer_class": "BartTokenizer"}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:94e05d5641aa58db52aee32293e970801889668d13a72d205ffa6fe324861d4b
|
3 |
+
size 3247
|
vocab.json
ADDED
The diff for this file is too large to render.
See raw diff
|
wandb/debug-cli.log
ADDED
@@ -0,0 +1,155 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
2022-05-03 17:09:46 INFO Running runs: []
|
2 |
+
2022-05-03 17:09:46 INFO Agent received command: run
|
3 |
+
2022-05-03 17:09:46 INFO Agent starting run with config:
|
4 |
+
eval_split_name: test
|
5 |
+
eval_steps: 500
|
6 |
+
evaluation_strategy: steps
|
7 |
+
generation_max_length: 40
|
8 |
+
generation_num_beams: 1
|
9 |
+
gradient_accumulation_steps: 8
|
10 |
+
greater_is_better: True
|
11 |
+
hidden_dropout: 0.036619638921206475
|
12 |
+
language: fr.en
|
13 |
+
learning_rate: 0.00024391819705381628
|
14 |
+
logging_steps: 1
|
15 |
+
max_duration_in_seconds: 20
|
16 |
+
metric_for_best_model: bleu
|
17 |
+
model_name_or_path: ./
|
18 |
+
num_train_epochs: 3
|
19 |
+
output_dir: ./output_dir
|
20 |
+
per_device_eval_batch_size: 4
|
21 |
+
per_device_train_batch_size: 4
|
22 |
+
save_steps: 500
|
23 |
+
task: covost2
|
24 |
+
warmup_steps: 500
|
25 |
+
2022-05-03 17:09:46 INFO About to run command: python3 run_xtreme_s.py --overwrite_output_dir --freeze_feature_encoder --gradient_checkpointing --predict_with_generate --fp16 --group_by_length --do_train --do_eval --load_best_model_at_end --push_to_hub --use_auth_token --eval_split_name=test --eval_steps=500 --evaluation_strategy=steps --generation_max_length=40 --generation_num_beams=1 --gradient_accumulation_steps=8 --greater_is_better=True --hidden_dropout=0.036619638921206475 --language=fr.en --learning_rate=0.00024391819705381628 --logging_steps=1 --max_duration_in_seconds=20 --metric_for_best_model=bleu --model_name_or_path=./ --num_train_epochs=3 --output_dir=./output_dir --per_device_eval_batch_size=4 --per_device_train_batch_size=4 --save_steps=500 --task=covost2 --warmup_steps=500
|
26 |
+
2022-05-03 17:09:51 INFO Running runs: ['vz5ppd75']
|
27 |
+
2022-05-03 17:10:26 INFO Cleaning up finished run: vz5ppd75
|
28 |
+
2022-05-03 17:10:28 INFO Agent received command: run
|
29 |
+
2022-05-03 17:10:28 INFO Agent starting run with config:
|
30 |
+
eval_split_name: test
|
31 |
+
eval_steps: 500
|
32 |
+
evaluation_strategy: steps
|
33 |
+
generation_max_length: 40
|
34 |
+
generation_num_beams: 1
|
35 |
+
gradient_accumulation_steps: 8
|
36 |
+
greater_is_better: True
|
37 |
+
hidden_dropout: 0.1875094322808032
|
38 |
+
language: fr.en
|
39 |
+
learning_rate: 0.00024438201183496223
|
40 |
+
logging_steps: 1
|
41 |
+
max_duration_in_seconds: 20
|
42 |
+
metric_for_best_model: bleu
|
43 |
+
model_name_or_path: ./
|
44 |
+
num_train_epochs: 3
|
45 |
+
output_dir: ./output_dir
|
46 |
+
per_device_eval_batch_size: 4
|
47 |
+
per_device_train_batch_size: 4
|
48 |
+
save_steps: 500
|
49 |
+
task: covost2
|
50 |
+
warmup_steps: 500
|
51 |
+
2022-05-03 17:10:36 INFO Running runs: []
|
52 |
+
2022-05-03 17:10:36 INFO Agent received command: run
|
53 |
+
2022-05-03 17:10:36 INFO Agent starting run with config:
|
54 |
+
eval_split_name: test
|
55 |
+
eval_steps: 500
|
56 |
+
evaluation_strategy: steps
|
57 |
+
generation_max_length: 40
|
58 |
+
generation_num_beams: 1
|
59 |
+
gradient_accumulation_steps: 8
|
60 |
+
greater_is_better: True
|
61 |
+
hidden_dropout: 0.055722391000930585
|
62 |
+
language: fr.en
|
63 |
+
learning_rate: 0.0006457481677728278
|
64 |
+
logging_steps: 1
|
65 |
+
max_duration_in_seconds: 20
|
66 |
+
metric_for_best_model: bleu
|
67 |
+
model_name_or_path: ./
|
68 |
+
num_train_epochs: 3
|
69 |
+
output_dir: ./output_dir
|
70 |
+
per_device_eval_batch_size: 4
|
71 |
+
per_device_train_batch_size: 4
|
72 |
+
save_steps: 500
|
73 |
+
task: covost2
|
74 |
+
warmup_steps: 500
|
75 |
+
2022-05-03 17:10:36 INFO About to run command: python3 run_xtreme_s.py --overwrite_output_dir --freeze_feature_encoder --gradient_checkpointing --predict_with_generate --fp16 --group_by_length --do_train --do_eval --load_best_model_at_end --push_to_hub --use_auth_token --eval_split_name=test --eval_steps=500 --evaluation_strategy=steps --generation_max_length=40 --generation_num_beams=1 --gradient_accumulation_steps=8 --greater_is_better=True --hidden_dropout=0.055722391000930585 --language=fr.en --learning_rate=0.0006457481677728278 --logging_steps=1 --max_duration_in_seconds=20 --metric_for_best_model=bleu --model_name_or_path=./ --num_train_epochs=3 --output_dir=./output_dir --per_device_eval_batch_size=4 --per_device_train_batch_size=4 --save_steps=500 --task=covost2 --warmup_steps=500
|
76 |
+
2022-05-03 17:10:41 INFO Running runs: ['ldsojzle']
|
77 |
+
2022-05-03 17:11:07 INFO Cleaning up finished run: ldsojzle
|
78 |
+
2022-05-03 17:11:07 INFO Agent received command: run
|
79 |
+
2022-05-03 17:11:07 INFO Agent starting run with config:
|
80 |
+
eval_split_name: test
|
81 |
+
eval_steps: 500
|
82 |
+
evaluation_strategy: steps
|
83 |
+
generation_max_length: 40
|
84 |
+
generation_num_beams: 1
|
85 |
+
gradient_accumulation_steps: 8
|
86 |
+
greater_is_better: True
|
87 |
+
hidden_dropout: 0.056807662149569525
|
88 |
+
language: fr.en
|
89 |
+
learning_rate: 0.0005558468401613797
|
90 |
+
logging_steps: 1
|
91 |
+
max_duration_in_seconds: 20
|
92 |
+
metric_for_best_model: bleu
|
93 |
+
model_name_or_path: ./
|
94 |
+
num_train_epochs: 3
|
95 |
+
output_dir: ./output_dir
|
96 |
+
per_device_eval_batch_size: 4
|
97 |
+
per_device_train_batch_size: 4
|
98 |
+
save_steps: 500
|
99 |
+
task: covost2
|
100 |
+
warmup_steps: 500
|
101 |
+
2022-05-03 17:11:07 INFO About to run command: python3 run_xtreme_s.py --overwrite_output_dir --freeze_feature_encoder --gradient_checkpointing --predict_with_generate --fp16 --group_by_length --do_train --do_eval --load_best_model_at_end --push_to_hub --use_auth_token --eval_split_name=test --eval_steps=500 --evaluation_strategy=steps --generation_max_length=40 --generation_num_beams=1 --gradient_accumulation_steps=8 --greater_is_better=True --hidden_dropout=0.056807662149569525 --language=fr.en --learning_rate=0.0005558468401613797 --logging_steps=1 --max_duration_in_seconds=20 --metric_for_best_model=bleu --model_name_or_path=./ --num_train_epochs=3 --output_dir=./output_dir --per_device_eval_batch_size=4 --per_device_train_batch_size=4 --save_steps=500 --task=covost2 --warmup_steps=500
|
102 |
+
2022-05-03 17:11:12 INFO Running runs: ['qv3vjr6j']
|
103 |
+
2022-05-03 17:10:28 INFO About to run command: python3 run_xtreme_s.py --overwrite_output_dir --freeze_feature_encoder --gradient_checkpointing --predict_with_generate --fp16 --group_by_length --do_train --do_eval --load_best_model_at_end --push_to_hub --use_auth_token --eval_split_name=test --eval_steps=500 --evaluation_strategy=steps --generation_max_length=40 --generation_num_beams=1 --gradient_accumulation_steps=8 --greater_is_better=True --hidden_dropout=0.1875094322808032 --language=fr.en --learning_rate=0.00024438201183496223 --logging_steps=1 --max_duration_in_seconds=20 --metric_for_best_model=bleu --model_name_or_path=./ --num_train_epochs=3 --output_dir=./output_dir --per_device_eval_batch_size=4 --per_device_train_batch_size=4 --save_steps=500 --task=covost2 --warmup_steps=500
|
104 |
+
2022-05-03 17:11:29 INFO Running runs: ['irggvkgd']
|
105 |
+
2022-05-03 17:11:37 INFO Cleaning up finished run: qv3vjr6j
|
106 |
+
2022-05-03 17:11:37 INFO Agent received command: run
|
107 |
+
2022-05-03 17:11:37 INFO Agent starting run with config:
|
108 |
+
eval_split_name: test
|
109 |
+
eval_steps: 500
|
110 |
+
evaluation_strategy: steps
|
111 |
+
generation_max_length: 40
|
112 |
+
generation_num_beams: 1
|
113 |
+
gradient_accumulation_steps: 8
|
114 |
+
greater_is_better: True
|
115 |
+
hidden_dropout: 0.03413483050532159
|
116 |
+
language: fr.en
|
117 |
+
learning_rate: 0.00022086866790135088
|
118 |
+
logging_steps: 1
|
119 |
+
max_duration_in_seconds: 20
|
120 |
+
metric_for_best_model: bleu
|
121 |
+
model_name_or_path: ./
|
122 |
+
num_train_epochs: 3
|
123 |
+
output_dir: ./output_dir
|
124 |
+
per_device_eval_batch_size: 4
|
125 |
+
per_device_train_batch_size: 4
|
126 |
+
save_steps: 500
|
127 |
+
task: covost2
|
128 |
+
warmup_steps: 500
|
129 |
+
2022-05-03 17:11:37 INFO About to run command: python3 run_xtreme_s.py --overwrite_output_dir --freeze_feature_encoder --gradient_checkpointing --predict_with_generate --fp16 --group_by_length --do_train --do_eval --load_best_model_at_end --push_to_hub --use_auth_token --eval_split_name=test --eval_steps=500 --evaluation_strategy=steps --generation_max_length=40 --generation_num_beams=1 --gradient_accumulation_steps=8 --greater_is_better=True --hidden_dropout=0.03413483050532159 --language=fr.en --learning_rate=0.00022086866790135088 --logging_steps=1 --max_duration_in_seconds=20 --metric_for_best_model=bleu --model_name_or_path=./ --num_train_epochs=3 --output_dir=./output_dir --per_device_eval_batch_size=4 --per_device_train_batch_size=4 --save_steps=500 --task=covost2 --warmup_steps=500
|
130 |
+
2022-05-03 17:15:19 INFO Running runs: []
|
131 |
+
2022-05-03 17:15:19 INFO Agent received command: run
|
132 |
+
2022-05-03 17:15:19 INFO Agent starting run with config:
|
133 |
+
eval_split_name: test
|
134 |
+
eval_steps: 500
|
135 |
+
evaluation_strategy: steps
|
136 |
+
generation_max_length: 40
|
137 |
+
generation_num_beams: 1
|
138 |
+
gradient_accumulation_steps: 8
|
139 |
+
greater_is_better: True
|
140 |
+
hidden_dropout: 0.06862889720223829
|
141 |
+
language: fr.en
|
142 |
+
learning_rate: 0.0004848089062550082
|
143 |
+
logging_steps: 1
|
144 |
+
max_duration_in_seconds: 20
|
145 |
+
metric_for_best_model: bleu
|
146 |
+
model_name_or_path: ./
|
147 |
+
num_train_epochs: 3
|
148 |
+
output_dir: ./
|
149 |
+
per_device_eval_batch_size: 4
|
150 |
+
per_device_train_batch_size: 4
|
151 |
+
save_steps: 500
|
152 |
+
task: covost2
|
153 |
+
warmup_steps: 500
|
154 |
+
2022-05-03 17:15:19 INFO About to run command: python3 run_xtreme_s.py --overwrite_output_dir --freeze_feature_encoder --gradient_checkpointing --predict_with_generate --fp16 --group_by_length --do_train --do_eval --load_best_model_at_end --push_to_hub --use_auth_token --eval_split_name=test --eval_steps=500 --evaluation_strategy=steps --generation_max_length=40 --generation_num_beams=1 --gradient_accumulation_steps=8 --greater_is_better=True --hidden_dropout=0.06862889720223829 --language=fr.en --learning_rate=0.0004848089062550082 --logging_steps=1 --max_duration_in_seconds=20 --metric_for_best_model=bleu --model_name_or_path=./ --num_train_epochs=3 --output_dir=./ --per_device_eval_batch_size=4 --per_device_train_batch_size=4 --save_steps=500 --task=covost2 --warmup_steps=500
|
155 |
+
2022-05-03 17:15:24 INFO Running runs: ['a6039xud']
|
wandb/debug-internal.log
ADDED
@@ -0,0 +1 @@
|
|
|
1 |
+
run-20220503_171959-a6039xud/logs/debug-internal.log
|
wandb/debug.log
ADDED
@@ -0,0 +1 @@
|
|
|
1 |
+
run-20220503_171959-a6039xud/logs/debug.log
|
wandb/latest-run
ADDED
@@ -0,0 +1 @@
|
|
|
1 |
+
run-20220503_171959-a6039xud
|
wandb/run-20220503_171959-a6039xud/files/config.yaml
ADDED
The diff for this file is too large to render.
See raw diff
|
wandb/run-20220503_171959-a6039xud/files/output.log
ADDED
The diff for this file is too large to render.
See raw diff
|
wandb/run-20220503_171959-a6039xud/files/requirements.txt
ADDED
@@ -0,0 +1,287 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
absl-py==1.0.0
|
2 |
+
aiohttp==3.8.1
|
3 |
+
aiosignal==1.2.0
|
4 |
+
alembic==1.7.7
|
5 |
+
anyio==3.5.0
|
6 |
+
appdirs==1.4.4
|
7 |
+
apscheduler==3.9.1
|
8 |
+
argon2-cffi-bindings==21.2.0
|
9 |
+
argon2-cffi==21.3.0
|
10 |
+
arrow==1.2.2
|
11 |
+
asttokens==2.0.5
|
12 |
+
astunparse==1.6.3
|
13 |
+
async-timeout==4.0.2
|
14 |
+
attrs==21.4.0
|
15 |
+
audioread==2.1.9
|
16 |
+
autopage==0.5.0
|
17 |
+
babel==2.9.1
|
18 |
+
backcall==0.2.0
|
19 |
+
backoff==1.11.1
|
20 |
+
binaryornot==0.4.4
|
21 |
+
bitsandbytes-cuda113==0.26.0
|
22 |
+
black==22.1.0
|
23 |
+
bleach==4.1.0
|
24 |
+
boto3==1.16.34
|
25 |
+
botocore==1.19.63
|
26 |
+
brotli==1.0.9
|
27 |
+
cachetools==5.0.0
|
28 |
+
certifi==2021.10.8
|
29 |
+
cffi==1.15.0
|
30 |
+
chardet==4.0.0
|
31 |
+
charset-normalizer==2.0.11
|
32 |
+
chex==0.1.0
|
33 |
+
click==8.0.3
|
34 |
+
cliff==3.10.1
|
35 |
+
clldutils==3.10.1
|
36 |
+
cmaes==0.8.2
|
37 |
+
cmd2==2.4.0
|
38 |
+
codecarbon==1.2.0
|
39 |
+
colorlog==6.6.0
|
40 |
+
cookiecutter==1.7.3
|
41 |
+
cryptography==36.0.2
|
42 |
+
csvw==1.11.0
|
43 |
+
cycler==0.11.0
|
44 |
+
dash-bootstrap-components==1.1.0
|
45 |
+
dash-core-components==2.0.0
|
46 |
+
dash-html-components==2.0.0
|
47 |
+
dash-table==5.0.0
|
48 |
+
dash==2.3.1
|
49 |
+
datasets==2.1.1.dev0
|
50 |
+
debugpy==1.5.1
|
51 |
+
decorator==5.1.1
|
52 |
+
defusedxml==0.7.1
|
53 |
+
deprecated==1.2.13
|
54 |
+
dill==0.3.4
|
55 |
+
dlinfo==1.2.1
|
56 |
+
dm-tree==0.1.6
|
57 |
+
docker-pycreds==0.4.0
|
58 |
+
docker==4.4.4
|
59 |
+
entrypoints==0.4
|
60 |
+
execnet==1.9.0
|
61 |
+
executing==0.8.2
|
62 |
+
faiss-cpu==1.7.2
|
63 |
+
filelock==3.4.2
|
64 |
+
fire==0.4.0
|
65 |
+
flake8==4.0.1
|
66 |
+
flask-compress==1.11
|
67 |
+
flask==2.1.1
|
68 |
+
flatbuffers==1.12
|
69 |
+
flax==0.4.0
|
70 |
+
fonttools==4.29.1
|
71 |
+
frozenlist==1.3.0
|
72 |
+
fsspec==2022.1.0
|
73 |
+
fugashi==1.1.2
|
74 |
+
gast==0.5.3
|
75 |
+
gitdb==4.0.9
|
76 |
+
gitpython==3.1.18
|
77 |
+
google-auth-oauthlib==0.4.6
|
78 |
+
google-auth==2.6.0
|
79 |
+
google-pasta==0.2.0
|
80 |
+
greenlet==1.1.2
|
81 |
+
grpcio==1.43.0
|
82 |
+
h5py==3.6.0
|
83 |
+
hf-doc-builder==0.2.0
|
84 |
+
huggingface-hub==0.4.0
|
85 |
+
hypothesis==6.36.1
|
86 |
+
idna==3.3
|
87 |
+
importlib-metadata==4.10.1
|
88 |
+
iniconfig==1.1.1
|
89 |
+
ipadic==1.0.0
|
90 |
+
ipdb==0.13.9
|
91 |
+
ipykernel==6.8.0
|
92 |
+
ipython-genutils==0.2.0
|
93 |
+
ipython==8.0.1
|
94 |
+
ipywidgets==7.6.5
|
95 |
+
isodate==0.6.1
|
96 |
+
isort==5.10.1
|
97 |
+
itsdangerous==2.1.2
|
98 |
+
jax==0.2.28
|
99 |
+
jaxlib==0.1.76+cuda11.cudnn82
|
100 |
+
jedi==0.18.1
|
101 |
+
jinja2-time==0.2.0
|
102 |
+
jinja2==3.0.3
|
103 |
+
jiwer==2.3.0
|
104 |
+
jmespath==0.10.0
|
105 |
+
joblib==1.1.0
|
106 |
+
json5==0.9.6
|
107 |
+
jsonschema==4.4.0
|
108 |
+
jupyter-client==7.1.2
|
109 |
+
jupyter-console==6.4.0
|
110 |
+
jupyter-core==4.9.1
|
111 |
+
jupyter-server==1.13.5
|
112 |
+
jupyter==1.0.0
|
113 |
+
jupyterlab-pygments==0.1.2
|
114 |
+
jupyterlab-server==2.10.3
|
115 |
+
jupyterlab-widgets==1.0.2
|
116 |
+
jupyterlab==3.2.9
|
117 |
+
keras-preprocessing==1.1.2
|
118 |
+
keras==2.8.0
|
119 |
+
kiwisolver==1.3.2
|
120 |
+
kubernetes==12.0.1
|
121 |
+
libclang==13.0.0
|
122 |
+
librosa==0.8.1
|
123 |
+
llvmlite==0.38.0
|
124 |
+
mako==1.2.0
|
125 |
+
markdown==3.3.6
|
126 |
+
markupsafe==2.0.1
|
127 |
+
matplotlib-inline==0.1.3
|
128 |
+
matplotlib==3.5.1
|
129 |
+
mccabe==0.6.1
|
130 |
+
mistune==0.8.4
|
131 |
+
msgpack==1.0.3
|
132 |
+
multidict==6.0.2
|
133 |
+
multiprocess==0.70.12.2
|
134 |
+
mypy-extensions==0.4.3
|
135 |
+
nbclassic==0.3.5
|
136 |
+
nbclient==0.5.10
|
137 |
+
nbconvert==6.4.1
|
138 |
+
nbformat==5.1.3
|
139 |
+
nest-asyncio==1.5.4
|
140 |
+
nltk==3.7
|
141 |
+
notebook==6.4.8
|
142 |
+
numba==0.55.1
|
143 |
+
numpy==1.21.5
|
144 |
+
oauthlib==3.2.0
|
145 |
+
onnx==1.11.0
|
146 |
+
onnxconverter-common==1.9.0
|
147 |
+
opt-einsum==3.3.0
|
148 |
+
optax==0.1.0
|
149 |
+
optuna==2.10.0
|
150 |
+
packaging==21.3
|
151 |
+
pandas==1.4.0
|
152 |
+
pandocfilters==1.5.0
|
153 |
+
parameterized==0.8.1
|
154 |
+
parso==0.8.3
|
155 |
+
pathspec==0.9.0
|
156 |
+
pathtools==0.1.2
|
157 |
+
pbr==5.8.1
|
158 |
+
pexpect==4.8.0
|
159 |
+
phonemizer==3.0.1
|
160 |
+
pickleshare==0.7.5
|
161 |
+
pillow==9.0.0
|
162 |
+
pint==0.16.1
|
163 |
+
pip==22.0.2
|
164 |
+
pkg-resources==0.0.0
|
165 |
+
plac==1.3.5
|
166 |
+
platformdirs==2.4.1
|
167 |
+
plotly==5.6.0
|
168 |
+
pluggy==1.0.0
|
169 |
+
pooch==1.6.0
|
170 |
+
portalocker==2.0.0
|
171 |
+
poyo==0.5.0
|
172 |
+
prettytable==3.2.0
|
173 |
+
prometheus-client==0.13.1
|
174 |
+
promise==2.3
|
175 |
+
prompt-toolkit==3.0.26
|
176 |
+
protobuf==3.19.4
|
177 |
+
psutil==5.9.0
|
178 |
+
ptyprocess==0.7.0
|
179 |
+
pure-eval==0.2.2
|
180 |
+
py-cpuinfo==8.0.0
|
181 |
+
py==1.11.0
|
182 |
+
pyarrow==6.0.1
|
183 |
+
pyasn1-modules==0.2.8
|
184 |
+
pyasn1==0.4.8
|
185 |
+
pycodestyle==2.8.0
|
186 |
+
pycparser==2.21
|
187 |
+
pyctcdecode==0.3.0
|
188 |
+
pyflakes==2.4.0
|
189 |
+
pygments==2.11.2
|
190 |
+
pygtrie==2.4.2
|
191 |
+
pynvml==11.4.1
|
192 |
+
pyopenssl==22.0.0
|
193 |
+
pyparsing==3.0.7
|
194 |
+
pyperclip==1.8.2
|
195 |
+
pypng==0.0.21
|
196 |
+
pyrsistent==0.18.1
|
197 |
+
pytest-forked==1.4.0
|
198 |
+
pytest-timeout==2.1.0
|
199 |
+
pytest-xdist==2.5.0
|
200 |
+
pytest==7.1.1
|
201 |
+
python-dateutil==2.8.2
|
202 |
+
python-levenshtein==0.12.2
|
203 |
+
python-slugify==6.1.1
|
204 |
+
pytz-deprecation-shim==0.1.0.post0
|
205 |
+
pytz==2021.3
|
206 |
+
pyyaml==5.4.1
|
207 |
+
pyzmq==22.3.0
|
208 |
+
qtconsole==5.2.2
|
209 |
+
qtpy==2.0.1
|
210 |
+
ray==1.11.0
|
211 |
+
redis==4.2.2
|
212 |
+
regex==2022.1.18
|
213 |
+
requests-oauthlib==1.3.1
|
214 |
+
requests==2.27.1
|
215 |
+
resampy==0.2.2
|
216 |
+
responses==0.18.0
|
217 |
+
rfc3986==2.0.0
|
218 |
+
rouge-score==0.0.4
|
219 |
+
rsa==4.8
|
220 |
+
s3transfer==0.3.7
|
221 |
+
sacrebleu==1.5.1
|
222 |
+
sacremoses==0.0.47
|
223 |
+
scikit-learn==1.0.2
|
224 |
+
scipy==1.7.3
|
225 |
+
segments==2.2.0
|
226 |
+
send2trash==1.8.0
|
227 |
+
sentencepiece==0.1.96
|
228 |
+
sentry-sdk==1.5.6
|
229 |
+
setuptools==44.1.1
|
230 |
+
shortuuid==1.0.8
|
231 |
+
sigopt==8.3.0
|
232 |
+
six==1.16.0
|
233 |
+
smmap==5.0.0
|
234 |
+
sniffio==1.2.0
|
235 |
+
sortedcontainers==2.4.0
|
236 |
+
soundfile==0.10.3.post1
|
237 |
+
sqlalchemy==1.4.34
|
238 |
+
stack-data==0.1.4
|
239 |
+
stevedore==3.5.0
|
240 |
+
tabulate==0.8.9
|
241 |
+
tenacity==8.0.1
|
242 |
+
tensorboard-data-server==0.6.1
|
243 |
+
tensorboard-plugin-wit==1.8.1
|
244 |
+
tensorboard==2.8.0
|
245 |
+
tensorboardx==2.5
|
246 |
+
tensorflow-io-gcs-filesystem==0.24.0
|
247 |
+
tensorflow==2.8.0
|
248 |
+
termcolor==1.1.0
|
249 |
+
terminado==0.13.1
|
250 |
+
testpath==0.5.0
|
251 |
+
text-unidecode==1.3
|
252 |
+
tf-estimator-nightly==2.8.0.dev2021122109
|
253 |
+
tf2onnx==1.9.3
|
254 |
+
threadpoolctl==3.1.0
|
255 |
+
timeout-decorator==0.5.0
|
256 |
+
timm==0.5.4
|
257 |
+
tokenizers==0.11.4
|
258 |
+
toml==0.10.2
|
259 |
+
tomli==2.0.0
|
260 |
+
toolz==0.11.2
|
261 |
+
torch==1.10.2+cu113
|
262 |
+
torchaudio==0.10.2+cu113
|
263 |
+
torchvision==0.11.3
|
264 |
+
tornado==6.1
|
265 |
+
tqdm==4.62.3
|
266 |
+
traitlets==5.1.1
|
267 |
+
transformers==4.18.0.dev0
|
268 |
+
typing-extensions==3.10.0.2
|
269 |
+
tzdata==2022.1
|
270 |
+
tzlocal==4.2
|
271 |
+
unidic-lite==1.0.8
|
272 |
+
unidic==1.1.0
|
273 |
+
uritemplate==4.1.1
|
274 |
+
urllib3==1.26.8
|
275 |
+
wandb==0.12.10
|
276 |
+
wasabi==0.9.1
|
277 |
+
wcwidth==0.2.5
|
278 |
+
webencodings==0.5.1
|
279 |
+
websocket-client==1.2.3
|
280 |
+
werkzeug==2.0.2
|
281 |
+
wheel==0.37.1
|
282 |
+
widgetsnbextension==3.5.2
|
283 |
+
wrapt==1.14.0
|
284 |
+
xxhash==2.0.2
|
285 |
+
yarl==1.7.2
|
286 |
+
yaspin==2.1.0
|
287 |
+
zipp==3.7.0
|
wandb/run-20220503_171959-a6039xud/files/wandb-metadata.json
ADDED
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"os": "Linux-5.11.0-1028-gcp-x86_64-with-glibc2.33",
|
3 |
+
"python": "3.9.5",
|
4 |
+
"heartbeatAt": "2022-05-03T17:20:04.283739",
|
5 |
+
"startedAt": "2022-05-03T17:19:59.978222",
|
6 |
+
"docker": null,
|
7 |
+
"gpu": "Tesla V100-SXM2-16GB",
|
8 |
+
"gpu_count": 2,
|
9 |
+
"cpu_count": 16,
|
10 |
+
"cuda": null,
|
11 |
+
"args": [
|
12 |
+
"--overwrite_output_dir",
|
13 |
+
"--freeze_feature_encoder",
|
14 |
+
"--gradient_checkpointing",
|
15 |
+
"--predict_with_generate",
|
16 |
+
"--fp16",
|
17 |
+
"--group_by_length",
|
18 |
+
"--do_train",
|
19 |
+
"--do_eval",
|
20 |
+
"--load_best_model_at_end",
|
21 |
+
"--push_to_hub",
|
22 |
+
"--use_auth_token",
|
23 |
+
"--eval_split_name=test",
|
24 |
+
"--eval_steps=500",
|
25 |
+
"--evaluation_strategy=steps",
|
26 |
+
"--generation_max_length=40",
|
27 |
+
"--generation_num_beams=1",
|
28 |
+
"--gradient_accumulation_steps=8",
|
29 |
+
"--greater_is_better=True",
|
30 |
+
"--hidden_dropout=0.06862889720223829",
|
31 |
+
"--language=fr.en",
|
32 |
+
"--learning_rate=0.0004848089062550082",
|
33 |
+
"--logging_steps=1",
|
34 |
+
"--max_duration_in_seconds=20",
|
35 |
+
"--metric_for_best_model=bleu",
|
36 |
+
"--model_name_or_path=./",
|
37 |
+
"--num_train_epochs=3",
|
38 |
+
"--output_dir=./",
|
39 |
+
"--per_device_eval_batch_size=4",
|
40 |
+
"--per_device_train_batch_size=4",
|
41 |
+
"--save_steps=500",
|
42 |
+
"--task=covost2",
|
43 |
+
"--warmup_steps=500"
|
44 |
+
],
|
45 |
+
"state": "running",
|
46 |
+
"program": "/home/sanchit_huggingface_co/xtreme_s_xlsr_2_bart_covost2_fr_en/run_xtreme_s.py",
|
47 |
+
"codePath": "run_xtreme_s.py",
|
48 |
+
"git": {
|
49 |
+
"remote": "https://huggingface.co/sanchit-gandhi/xtreme_s_xlsr_2_bart_covost2_fr_en",
|
50 |
+
"commit": "c787ad4df48cedb3c1492e0593442fbabf32819f"
|
51 |
+
},
|
52 |
+
"email": "sanchit@huggingface.co",
|
53 |
+
"root": "/home/sanchit_huggingface_co/xtreme_s_xlsr_2_bart_covost2_fr_en",
|
54 |
+
"host": "sanchit--v100",
|
55 |
+
"username": "sanchit_huggingface_co",
|
56 |
+
"executable": "/home/sanchit_huggingface_co/gcp/bin/python3"
|
57 |
+
}
|
wandb/run-20220503_171959-a6039xud/files/wandb-summary.json
ADDED
The diff for this file is too large to render.
See raw diff
|
wandb/run-20220503_171959-a6039xud/logs/debug-internal.log
ADDED
The diff for this file is too large to render.
See raw diff
|
wandb/run-20220503_171959-a6039xud/logs/debug.log
ADDED
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
2022-05-03 17:19:59,980 INFO MainThread:42111 [wandb_setup.py:_flush():75] Loading settings from /home/sanchit_huggingface_co/.config/wandb/settings
|
2 |
+
2022-05-03 17:19:59,980 INFO MainThread:42111 [wandb_setup.py:_flush():75] Loading settings from wandb/settings
|
3 |
+
2022-05-03 17:19:59,980 INFO MainThread:42111 [wandb_setup.py:_flush():75] Loading settings from environment variables: {'entity': 'sanchit-gandhi', 'project': 'xtreme_s_xlsr_2_bart_covost2_fr_en', 'sweep_id': '39ci3gkf', 'root_dir': '/home/sanchit_huggingface_co/xtreme_s_xlsr_2_bart_covost2_fr_en', 'run_id': 'a6039xud', 'sweep_param_path': '/home/sanchit_huggingface_co/xtreme_s_xlsr_2_bart_covost2_fr_en/wandb/sweep-39ci3gkf/config-a6039xud.yaml'}
|
4 |
+
2022-05-03 17:19:59,980 INFO MainThread:42111 [wandb_setup.py:_flush():75] Inferring run settings from compute environment: {'program_relpath': 'run_xtreme_s.py', 'program': '/home/sanchit_huggingface_co/xtreme_s_xlsr_2_bart_covost2_fr_en/run_xtreme_s.py'}
|
5 |
+
2022-05-03 17:19:59,980 INFO MainThread:42111 [wandb_init.py:_log_setup():386] Logging user logs to /home/sanchit_huggingface_co/xtreme_s_xlsr_2_bart_covost2_fr_en/wandb/run-20220503_171959-a6039xud/logs/debug.log
|
6 |
+
2022-05-03 17:19:59,980 INFO MainThread:42111 [wandb_init.py:_log_setup():387] Logging internal logs to /home/sanchit_huggingface_co/xtreme_s_xlsr_2_bart_covost2_fr_en/wandb/run-20220503_171959-a6039xud/logs/debug-internal.log
|
7 |
+
2022-05-03 17:19:59,980 INFO MainThread:42111 [wandb_init.py:init():420] calling init triggers
|
8 |
+
2022-05-03 17:19:59,980 INFO MainThread:42111 [wandb_init.py:init():425] wandb.init called with sweep_config: {'eval_split_name': 'test', 'eval_steps': 500, 'evaluation_strategy': 'steps', 'generation_max_length': 40, 'generation_num_beams': 1, 'gradient_accumulation_steps': 8, 'greater_is_better': True, 'hidden_dropout': 0.06862889720223829, 'language': 'fr.en', 'learning_rate': 0.0004848089062550082, 'logging_steps': 1, 'max_duration_in_seconds': 20, 'metric_for_best_model': 'bleu', 'model_name_or_path': './', 'num_train_epochs': 3, 'output_dir': './', 'per_device_eval_batch_size': 4, 'per_device_train_batch_size': 4, 'save_steps': 500, 'task': 'covost2', 'warmup_steps': 500}
|
9 |
+
config: {}
|
10 |
+
2022-05-03 17:19:59,981 INFO MainThread:42111 [wandb_init.py:init():471] starting backend
|
11 |
+
2022-05-03 17:19:59,981 INFO MainThread:42111 [backend.py:_multiprocessing_setup():99] multiprocessing start_methods=fork,spawn,forkserver, using: spawn
|
12 |
+
2022-05-03 17:20:00,059 INFO MainThread:42111 [backend.py:ensure_launched():219] starting backend process...
|
13 |
+
2022-05-03 17:20:00,134 INFO MainThread:42111 [backend.py:ensure_launched():224] started backend process with pid: 42356
|
14 |
+
2022-05-03 17:20:00,137 INFO MainThread:42111 [wandb_init.py:init():480] backend started and connected
|
15 |
+
2022-05-03 17:20:00,140 INFO MainThread:42111 [wandb_run.py:_config_callback():966] config_cb None None {'eval_split_name': 'test', 'eval_steps': 500, 'evaluation_strategy': 'steps', 'generation_max_length': 40, 'generation_num_beams': 1, 'gradient_accumulation_steps': 8, 'greater_is_better': True, 'hidden_dropout': 0.06862889720223829, 'language': 'fr.en', 'learning_rate': 0.0004848089062550082, 'logging_steps': 1, 'max_duration_in_seconds': 20, 'metric_for_best_model': 'bleu', 'model_name_or_path': './', 'num_train_epochs': 3, 'output_dir': './', 'per_device_eval_batch_size': 4, 'per_device_train_batch_size': 4, 'save_steps': 500, 'task': 'covost2', 'warmup_steps': 500}
|
16 |
+
2022-05-03 17:20:00,155 INFO MainThread:42111 [wandb_init.py:init():550] updated telemetry
|
17 |
+
2022-05-03 17:20:00,359 INFO MainThread:42111 [wandb_init.py:init():581] communicating current version
|
18 |
+
2022-05-03 17:20:01,247 INFO MainThread:42111 [wandb_init.py:init():586] got version response upgrade_message: "wandb version 0.12.16 is available! To upgrade, please run:\n $ pip install wandb --upgrade"
|
19 |
+
|
20 |
+
2022-05-03 17:20:01,248 INFO MainThread:42111 [wandb_init.py:init():596] communicating run to backend with 30 second timeout
|
21 |
+
2022-05-03 17:20:01,757 INFO MainThread:42111 [wandb_init.py:init():624] starting run threads in backend
|
22 |
+
2022-05-03 17:20:04,342 INFO MainThread:42111 [wandb_run.py:_console_start():1827] atexit reg
|
23 |
+
2022-05-03 17:20:04,342 INFO MainThread:42111 [wandb_run.py:_redirect():1701] redirect: SettingsConsole.REDIRECT
|
24 |
+
2022-05-03 17:20:04,343 INFO MainThread:42111 [wandb_run.py:_redirect():1706] Redirecting console.
|
25 |
+
2022-05-03 17:20:04,345 INFO MainThread:42111 [wandb_run.py:_redirect():1762] Redirects installed.
|
26 |
+
2022-05-03 17:20:04,345 INFO MainThread:42111 [wandb_init.py:init():651] run started, returning control to user process
|
27 |
+
2022-05-03 17:20:04,348 INFO MainThread:42111 [wandb_run.py:_config_callback():966] config_cb None None {'return_dict': True, 'output_hidden_states': False, 'output_attentions': False, 'torchscript': False, 'torch_dtype': 'torch.float32', 'use_bfloat16': False, 'pruned_heads': {}, 'tie_word_embeddings': False, 'is_encoder_decoder': True, 'is_decoder': False, 'cross_attention_hidden_size': None, 'add_cross_attention': False, 'tie_encoder_decoder': False, 'max_length': 40, 'min_length': 0, 'do_sample': False, 'early_stopping': False, 'num_beams': 1, 'num_beam_groups': 1, 'diversity_penalty': 0.0, 'temperature': 1.0, 'top_k': 50, 'top_p': 1.0, 'typical_p': 1.0, 'repetition_penalty': 1.0, 'length_penalty': 1.0, 'no_repeat_ngram_size': 0, 'encoder_no_repeat_ngram_size': 0, 'bad_words_ids': None, 'num_return_sequences': 1, 'chunk_size_feed_forward': 0, 'output_scores': False, 'return_dict_in_generate': False, 'forced_bos_token_id': None, 'forced_eos_token_id': None, 'remove_invalid_values': False, 'exponential_decay_length_penalty': None, 'architectures': ['SpeechEncoderDecoderModel'], 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': None, 'pad_token_id': 1, 'eos_token_id': 2, 'sep_token_id': None, 'decoder_start_token_id': 0, 'task_specific_params': None, 'problem_type': None, '_name_or_path': './', 'transformers_version': None, 'decoder': {'vocab_size': 50265, 'max_position_embeddings': 1024, 'd_model': 1024, 'encoder_ffn_dim': 4096, 'encoder_layers': 12, 'encoder_attention_heads': 16, 'decoder_ffn_dim': 4096, 'decoder_layers': 12, 'decoder_attention_heads': 16, 'dropout': 0.1, 'attention_dropout': 0.1, 'activation_dropout': 0.1, 'activation_function': 'gelu', 'init_std': 0.02, 'encoder_layerdrop': 0.0, 'decoder_layerdrop': 0.0, 'classifier_dropout': 0.0, 'use_cache': True, 'num_hidden_layers': 12, 'scale_embedding': False, 'return_dict': True, 'output_hidden_states': False, 'output_attentions': False, 'torchscript': False, 'torch_dtype': None, 'use_bfloat16': False, 'pruned_heads': {}, 'tie_word_embeddings': True, 'is_encoder_decoder': False, 'is_decoder': True, 'cross_attention_hidden_size': None, 'add_cross_attention': True, 'tie_encoder_decoder': False, 'max_length': 20, 'min_length': 0, 'do_sample': False, 'early_stopping': True, 'num_beams': 4, 'num_beam_groups': 1, 'diversity_penalty': 0.0, 'temperature': 1.0, 'top_k': 50, 'top_p': 1.0, 'typical_p': 1.0, 'repetition_penalty': 1.0, 'length_penalty': 1.0, 'no_repeat_ngram_size': 3, 'encoder_no_repeat_ngram_size': 0, 'bad_words_ids': None, 'num_return_sequences': 1, 'chunk_size_feed_forward': 0, 'output_scores': False, 'return_dict_in_generate': False, 'forced_bos_token_id': 0, 'forced_eos_token_id': 2, 'remove_invalid_values': False, 'exponential_decay_length_penalty': None, 'architectures': ['BartModel'], 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1', 2: 'LABEL_2'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1, 'LABEL_2': 2}, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': 0, 'pad_token_id': 1, 'eos_token_id': 2, 'sep_token_id': None, 'decoder_start_token_id': 2, 'task_specific_params': {'summarization': {'length_penalty': 1.0, 'max_length': 128, 'min_length': 12, 'num_beams': 4}, 'summarization_cnn': {'length_penalty': 2.0, 'max_length': 142, 'min_length': 56, 'num_beams': 4}, 'summarization_xsum': {'length_penalty': 1.0, 'max_length': 62, 'min_length': 11, 'num_beams': 6}}, 'problem_type': None, '_name_or_path': 'facebook/bart-large', 'transformers_version': '4.19.0.dev0', 'add_bias_logits': False, 'add_final_layer_norm': False, 'classif_dropout': 0.1, 'gradient_checkpointing': False, 'normalize_before': False, 'model_type': 'bart'}, 'encoder': {'return_dict': True, 'output_hidden_states': False, 'output_attentions': False, 'torchscript': False, 'torch_dtype': 'float32', 'use_bfloat16': False, 'pruned_heads': {}, 'tie_word_embeddings': True, 'is_encoder_decoder': False, 'is_decoder': False, 'cross_attention_hidden_size': None, 'add_cross_attention': False, 'tie_encoder_decoder': False, 'max_length': 20, 'min_length': 0, 'do_sample': False, 'early_stopping': False, 'num_beams': 1, 'num_beam_groups': 1, 'diversity_penalty': 0.0, 'temperature': 1.0, 'top_k': 50, 'top_p': 1.0, 'typical_p': 1.0, 'repetition_penalty': 1.0, 'length_penalty': 1.0, 'no_repeat_ngram_size': 0, 'encoder_no_repeat_ngram_size': 0, 'bad_words_ids': None, 'num_return_sequences': 1, 'chunk_size_feed_forward': 0, 'output_scores': False, 'return_dict_in_generate': False, 'forced_bos_token_id': None, 'forced_eos_token_id': None, 'remove_invalid_values': False, 'exponential_decay_length_penalty': None, 'architectures': ['Wav2Vec2ForPreTraining'], 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': 1, 'pad_token_id': 0, 'eos_token_id': 2, 'sep_token_id': None, 'decoder_start_token_id': None, 'task_specific_params': None, 'problem_type': None, '_name_or_path': 'facebook/wav2vec2-xls-r-300m', 'transformers_version': '4.19.0.dev0', 'feat_extract_dropout': 0.0, 'gradient_checkpointing': False, 'num_feat_extract_layers': 7, 'hidden_size': 1024, 'feat_extract_norm': 'layer', 'feat_extract_activation': 'gelu', 'conv_dim': [512, 512, 512, 512, 512, 512, 512], 'conv_stride': [5, 2, 2, 2, 2, 2, 2], 'conv_kernel': [10, 3, 3, 3, 3, 2, 2], 'conv_bias': True, 'num_conv_pos_embeddings': 128, 'num_conv_pos_embedding_groups': 16, 'num_hidden_layers': 24, 'intermediate_size': 4096, 'hidden_act': 'gelu', 'num_attention_heads': 16, 'hidden_dropout': 0.06862889720223829, 'attention_dropout': 0.1, 'activation_dropout': 0.0, 'feat_proj_dropout': 0.0, 'final_dropout': 0.0, 'layerdrop': 0.0, 'layer_norm_eps': 1e-05, 'initializer_range': 0.02, 'vocab_size': 32, 'do_stable_layer_norm': True, 'use_weighted_layer_sum': False, 'apply_spec_augment': True, 'mask_time_prob': 0.1, 'mask_time_length': 10, 'mask_time_min_masks': 2, 'mask_feature_prob': 0.0, 'mask_feature_length': 10, 'mask_feature_min_masks': 0, 'num_codevectors_per_group': 320, 'num_codevector_groups': 2, 'contrastive_logits_temperature': 0.1, 'feat_quantizer_dropout': 0.0, 'num_negatives': 100, 'codevector_dim': 768, 'proj_codevector_dim': 768, 'diversity_loss_weight': 0.1, 'ctc_loss_reduction': 'sum', 'ctc_zero_infinity': False, 'add_adapter': True, 'adapter_kernel_size': 3, 'adapter_stride': 2, 'num_adapter_layers': 3, 'output_hidden_size': 1024, 'classifier_proj_size': 256, 'tdnn_dim': [512, 512, 512, 512, 1500], 'tdnn_kernel': [5, 3, 3, 1, 1], 'tdnn_dilation': [1, 2, 3, 1, 1], 'xvector_output_dim': 512, 'model_type': 'wav2vec2'}, 'model_type': 'speech-encoder-decoder', 'processor_class': 'Wav2Vec2Processor', 'use_cache': False, 'overwrite_output_dir': True, 'do_train': True, 'do_eval': True, 'do_predict': False, 'prediction_loss_only': False, 'per_gpu_train_batch_size': 'None', 'per_gpu_eval_batch_size': 'None', 'eval_accumulation_steps': 'None', 'eval_delay': 0, 'weight_decay': 0.0, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'max_grad_norm': 1.0, 'max_steps': -1, 'lr_scheduler_type': 'linear', 'warmup_ratio': 0.0, 'log_level': -1, 'log_level_replica': -1, 'log_on_each_node': True, 'logging_dir': './runs/May03_17-15-22_sanchit--v100', 'logging_strategy': 'steps', 'logging_first_step': False, 'logging_nan_inf_filter': True, 'save_strategy': 'steps', 'save_total_limit': 'None', 'save_on_each_node': False, 'no_cuda': False, 'seed': 42, 'data_seed': 'None', 'bf16': False, 'fp16': True, 'fp16_opt_level': 'O1', 'half_precision_backend': 'amp', 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': 'None', 'local_rank': -1, 'xpu_backend': 'None', 'tpu_num_cores': 'None', 'tpu_metrics_debug': False, 'debug': '[]', 'dataloader_drop_last': False, 'dataloader_num_workers': 0, 'past_index': -1, 'run_name': './', 'disable_tqdm': False, 'remove_unused_columns': True, 'label_names': 'None', 'load_best_model_at_end': True, 'ignore_data_skip': False, 'sharded_ddp': '[]', 'deepspeed': 'None', 'label_smoothing_factor': 0.0, 'optim': 'adamw_hf', 'adafactor': False, 'group_by_length': True, 'length_column_name': 'length', 'report_to': "['tensorboard', 'wandb', 'codecarbon']", 'ddp_find_unused_parameters': 'None', 'ddp_bucket_cap_mb': 'None', 'dataloader_pin_memory': True, 'skip_memory_metrics': True, 'use_legacy_prediction_loop': False, 'push_to_hub': True, 'resume_from_checkpoint': 'None', 'hub_model_id': 'None', 'hub_strategy': 'every_save', 'hub_token': '<HUB_TOKEN>', 'hub_private_repo': False, 'gradient_checkpointing': True, 'include_inputs_for_metrics': False, 'fp16_backend': 'auto', 'push_to_hub_model_id': 'None', 'push_to_hub_organization': 'None', 'push_to_hub_token': '<PUSH_TO_HUB_TOKEN>', '_n_gpu': 1, 'mp_parameters': '', 'sortish_sampler': False, 'predict_with_generate': True, 'train_batch_size': 4, 'eval_batch_size': 4}
|
28 |
+
2022-05-03 17:20:04,351 INFO MainThread:42111 [wandb_watch.py:watch():43] Watching
|
wandb/run-20220503_171959-a6039xud/run-a6039xud.wandb
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a2d3034118c3e235104590124be355a3a0b9fcb07d29b2485f05cfac95b555bf
|
3 |
+
size 54318872
|
wandb/sweep-39ci3gkf/config-a6039xud.yaml
ADDED
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
wandb_version: 1
|
2 |
+
|
3 |
+
eval_split_name:
|
4 |
+
value: test
|
5 |
+
eval_steps:
|
6 |
+
value: 500
|
7 |
+
evaluation_strategy:
|
8 |
+
value: steps
|
9 |
+
generation_max_length:
|
10 |
+
value: 40
|
11 |
+
generation_num_beams:
|
12 |
+
value: 1
|
13 |
+
gradient_accumulation_steps:
|
14 |
+
value: 8
|
15 |
+
greater_is_better:
|
16 |
+
value: true
|
17 |
+
hidden_dropout:
|
18 |
+
value: 0.06862889720223829
|
19 |
+
language:
|
20 |
+
value: fr.en
|
21 |
+
learning_rate:
|
22 |
+
value: 0.0004848089062550082
|
23 |
+
logging_steps:
|
24 |
+
value: 1
|
25 |
+
max_duration_in_seconds:
|
26 |
+
value: 20
|
27 |
+
metric_for_best_model:
|
28 |
+
value: bleu
|
29 |
+
model_name_or_path:
|
30 |
+
value: ./
|
31 |
+
num_train_epochs:
|
32 |
+
value: 3
|
33 |
+
output_dir:
|
34 |
+
value: ./
|
35 |
+
per_device_eval_batch_size:
|
36 |
+
value: 4
|
37 |
+
per_device_train_batch_size:
|
38 |
+
value: 4
|
39 |
+
save_steps:
|
40 |
+
value: 500
|
41 |
+
task:
|
42 |
+
value: covost2
|
43 |
+
warmup_steps:
|
44 |
+
value: 500
|
wandb/sweep-y3ak427l/config-irggvkgd.yaml
ADDED
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
wandb_version: 1
|
2 |
+
|
3 |
+
eval_split_name:
|
4 |
+
value: test
|
5 |
+
eval_steps:
|
6 |
+
value: 500
|
7 |
+
evaluation_strategy:
|
8 |
+
value: steps
|
9 |
+
generation_max_length:
|
10 |
+
value: 40
|
11 |
+
generation_num_beams:
|
12 |
+
value: 1
|
13 |
+
gradient_accumulation_steps:
|
14 |
+
value: 8
|
15 |
+
greater_is_better:
|
16 |
+
value: true
|
17 |
+
hidden_dropout:
|
18 |
+
value: 0.1875094322808032
|
19 |
+
language:
|
20 |
+
value: fr.en
|
21 |
+
learning_rate:
|
22 |
+
value: 0.00024438201183496223
|
23 |
+
logging_steps:
|
24 |
+
value: 1
|
25 |
+
max_duration_in_seconds:
|
26 |
+
value: 20
|
27 |
+
metric_for_best_model:
|
28 |
+
value: bleu
|
29 |
+
model_name_or_path:
|
30 |
+
value: ./
|
31 |
+
num_train_epochs:
|
32 |
+
value: 3
|
33 |
+
output_dir:
|
34 |
+
value: ./output_dir
|
35 |
+
per_device_eval_batch_size:
|
36 |
+
value: 4
|
37 |
+
per_device_train_batch_size:
|
38 |
+
value: 4
|
39 |
+
save_steps:
|
40 |
+
value: 500
|
41 |
+
task:
|
42 |
+
value: covost2
|
43 |
+
warmup_steps:
|
44 |
+
value: 500
|
wandb/sweep-y3ak427l/config-ldsojzle.yaml
ADDED
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
wandb_version: 1
|
2 |
+
|
3 |
+
eval_split_name:
|
4 |
+
value: test
|
5 |
+
eval_steps:
|
6 |
+
value: 500
|
7 |
+
evaluation_strategy:
|
8 |
+
value: steps
|
9 |
+
generation_max_length:
|
10 |
+
value: 40
|
11 |
+
generation_num_beams:
|
12 |
+
value: 1
|
13 |
+
gradient_accumulation_steps:
|
14 |
+
value: 8
|
15 |
+
greater_is_better:
|
16 |
+
value: true
|
17 |
+
hidden_dropout:
|
18 |
+
value: 0.055722391000930585
|
19 |
+
language:
|
20 |
+
value: fr.en
|
21 |
+
learning_rate:
|
22 |
+
value: 0.0006457481677728278
|
23 |
+
logging_steps:
|
24 |
+
value: 1
|
25 |
+
max_duration_in_seconds:
|
26 |
+
value: 20
|
27 |
+
metric_for_best_model:
|
28 |
+
value: bleu
|
29 |
+
model_name_or_path:
|
30 |
+
value: ./
|
31 |
+
num_train_epochs:
|
32 |
+
value: 3
|
33 |
+
output_dir:
|
34 |
+
value: ./output_dir
|
35 |
+
per_device_eval_batch_size:
|
36 |
+
value: 4
|
37 |
+
per_device_train_batch_size:
|
38 |
+
value: 4
|
39 |
+
save_steps:
|
40 |
+
value: 500
|
41 |
+
task:
|
42 |
+
value: covost2
|
43 |
+
warmup_steps:
|
44 |
+
value: 500
|
wandb/sweep-y3ak427l/config-qv3vjr6j.yaml
ADDED
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
wandb_version: 1
|
2 |
+
|
3 |
+
eval_split_name:
|
4 |
+
value: test
|
5 |
+
eval_steps:
|
6 |
+
value: 500
|
7 |
+
evaluation_strategy:
|
8 |
+
value: steps
|
9 |
+
generation_max_length:
|
10 |
+
value: 40
|
11 |
+
generation_num_beams:
|
12 |
+
value: 1
|
13 |
+
gradient_accumulation_steps:
|
14 |
+
value: 8
|
15 |
+
greater_is_better:
|
16 |
+
value: true
|
17 |
+
hidden_dropout:
|
18 |
+
value: 0.056807662149569525
|
19 |
+
language:
|
20 |
+
value: fr.en
|
21 |
+
learning_rate:
|
22 |
+
value: 0.0005558468401613797
|
23 |
+
logging_steps:
|
24 |
+
value: 1
|
25 |
+
max_duration_in_seconds:
|
26 |
+
value: 20
|
27 |
+
metric_for_best_model:
|
28 |
+
value: bleu
|
29 |
+
model_name_or_path:
|
30 |
+
value: ./
|
31 |
+
num_train_epochs:
|
32 |
+
value: 3
|
33 |
+
output_dir:
|
34 |
+
value: ./output_dir
|
35 |
+
per_device_eval_batch_size:
|
36 |
+
value: 4
|
37 |
+
per_device_train_batch_size:
|
38 |
+
value: 4
|
39 |
+
save_steps:
|
40 |
+
value: 500
|
41 |
+
task:
|
42 |
+
value: covost2
|
43 |
+
warmup_steps:
|
44 |
+
value: 500
|
wandb/sweep-y3ak427l/config-vz5ppd75.yaml
ADDED
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
wandb_version: 1
|
2 |
+
|
3 |
+
eval_split_name:
|
4 |
+
value: test
|
5 |
+
eval_steps:
|
6 |
+
value: 500
|
7 |
+
evaluation_strategy:
|
8 |
+
value: steps
|
9 |
+
generation_max_length:
|
10 |
+
value: 40
|
11 |
+
generation_num_beams:
|
12 |
+
value: 1
|
13 |
+
gradient_accumulation_steps:
|
14 |
+
value: 8
|
15 |
+
greater_is_better:
|
16 |
+
value: true
|
17 |
+
hidden_dropout:
|
18 |
+
value: 0.036619638921206475
|
19 |
+
language:
|
20 |
+
value: fr.en
|
21 |
+
learning_rate:
|
22 |
+
value: 0.00024391819705381628
|
23 |
+
logging_steps:
|
24 |
+
value: 1
|
25 |
+
max_duration_in_seconds:
|
26 |
+
value: 20
|
27 |
+
metric_for_best_model:
|
28 |
+
value: bleu
|
29 |
+
model_name_or_path:
|
30 |
+
value: ./
|
31 |
+
num_train_epochs:
|
32 |
+
value: 3
|
33 |
+
output_dir:
|
34 |
+
value: ./output_dir
|
35 |
+
per_device_eval_batch_size:
|
36 |
+
value: 4
|
37 |
+
per_device_train_batch_size:
|
38 |
+
value: 4
|
39 |
+
save_steps:
|
40 |
+
value: 500
|
41 |
+
task:
|
42 |
+
value: covost2
|
43 |
+
warmup_steps:
|
44 |
+
value: 500
|
wandb/sweep-y3ak427l/config-xur584bd.yaml
ADDED
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
wandb_version: 1
|
2 |
+
|
3 |
+
eval_split_name:
|
4 |
+
value: test
|
5 |
+
eval_steps:
|
6 |
+
value: 500
|
7 |
+
evaluation_strategy:
|
8 |
+
value: steps
|
9 |
+
generation_max_length:
|
10 |
+
value: 40
|
11 |
+
generation_num_beams:
|
12 |
+
value: 1
|
13 |
+
gradient_accumulation_steps:
|
14 |
+
value: 8
|
15 |
+
greater_is_better:
|
16 |
+
value: true
|
17 |
+
hidden_dropout:
|
18 |
+
value: 0.03413483050532159
|
19 |
+
language:
|
20 |
+
value: fr.en
|
21 |
+
learning_rate:
|
22 |
+
value: 0.00022086866790135088
|
23 |
+
logging_steps:
|
24 |
+
value: 1
|
25 |
+
max_duration_in_seconds:
|
26 |
+
value: 20
|
27 |
+
metric_for_best_model:
|
28 |
+
value: bleu
|
29 |
+
model_name_or_path:
|
30 |
+
value: ./
|
31 |
+
num_train_epochs:
|
32 |
+
value: 3
|
33 |
+
output_dir:
|
34 |
+
value: ./output_dir
|
35 |
+
per_device_eval_batch_size:
|
36 |
+
value: 4
|
37 |
+
per_device_train_batch_size:
|
38 |
+
value: 4
|
39 |
+
save_steps:
|
40 |
+
value: 500
|
41 |
+
task:
|
42 |
+
value: covost2
|
43 |
+
warmup_steps:
|
44 |
+
value: 500
|