Training in progress, step 1000
Browse files- config.json +4 -4
- evalonlyhindi_indicwav2vec_MUCS_warmup500_s300shuff100_2142429.out +40 -0
- indicwav2vec_MUCS_warmup500_s300shuff100_2143522.out +0 -0
- language_segregated_prediction_texts/evalpredictions_hindi_indicw2v_ad0_3_hd_02_featd_0_2_lr6e-4_warmup2000_s300_shuf500.txt +0 -0
- language_segregated_prediction_texts/evalpredictions_hindi_indicw2v_ad0_3_hd_02_featd_0_2_lr6e-4_warmup2000_s400_shuf42.txt +0 -0
- model.safetensors +1 -1
- predictions_indicw2v_ad0_3_hd_02_featd_0_2_lr6e-4_warmup500_s300_shuff100.txt +0 -0
- training_args.bin +1 -1
config.json
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
{
|
2 |
-
"_name_or_path": "/scratch/elec/puhe/p/palp3/MUCS/
|
3 |
"activation_dropout": 0.0,
|
4 |
"adapter_attn_dim": null,
|
5 |
"adapter_kernel_size": 3,
|
@@ -9,7 +9,7 @@
|
|
9 |
"architectures": [
|
10 |
"Wav2Vec2ForCTC"
|
11 |
],
|
12 |
-
"attention_dropout": 0.
|
13 |
"bos_token_id": 1,
|
14 |
"classifier_proj_size": 256,
|
15 |
"codevector_dim": 256,
|
@@ -50,11 +50,11 @@
|
|
50 |
"feat_extract_activation": "gelu",
|
51 |
"feat_extract_dropout": 0.0,
|
52 |
"feat_extract_norm": "layer",
|
53 |
-
"feat_proj_dropout": 0.
|
54 |
"feat_quantizer_dropout": 0.0,
|
55 |
"final_dropout": 0.0,
|
56 |
"hidden_act": "gelu",
|
57 |
-
"hidden_dropout": 0.
|
58 |
"hidden_dropout_prob": 0.1,
|
59 |
"hidden_size": 1024,
|
60 |
"initializer_range": 0.02,
|
|
|
1 |
{
|
2 |
+
"_name_or_path": "/m/triton/scratch/elec/puhe/p/palp3/MUCS/indicwav2vec-hindi",
|
3 |
"activation_dropout": 0.0,
|
4 |
"adapter_attn_dim": null,
|
5 |
"adapter_kernel_size": 3,
|
|
|
9 |
"architectures": [
|
10 |
"Wav2Vec2ForCTC"
|
11 |
],
|
12 |
+
"attention_dropout": 0.3,
|
13 |
"bos_token_id": 1,
|
14 |
"classifier_proj_size": 256,
|
15 |
"codevector_dim": 256,
|
|
|
50 |
"feat_extract_activation": "gelu",
|
51 |
"feat_extract_dropout": 0.0,
|
52 |
"feat_extract_norm": "layer",
|
53 |
+
"feat_proj_dropout": 0.3,
|
54 |
"feat_quantizer_dropout": 0.0,
|
55 |
"final_dropout": 0.0,
|
56 |
"hidden_act": "gelu",
|
57 |
+
"hidden_dropout": 0.2,
|
58 |
"hidden_dropout_prob": 0.1,
|
59 |
"hidden_size": 1024,
|
60 |
"initializer_range": 0.02,
|
evalonlyhindi_indicwav2vec_MUCS_warmup500_s300shuff100_2142429.out
CHANGED
@@ -153,3 +153,43 @@ last prediction string
|
|
153 |
eval_steps_per_second = 1.164
|
154 |
eval_wer = 1.0
|
155 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
153 |
eval_steps_per_second = 1.164
|
154 |
eval_wer = 1.0
|
155 |
|
156 |
+
wandb: - 0.005 MB of 0.005 MB uploaded
|
157 |
+
wandb: Run history:
|
158 |
+
wandb: eval/cer β
|
159 |
+
wandb: eval/model_preparation_time β
|
160 |
+
wandb: eval/runtime β
|
161 |
+
wandb: eval/samples_per_second β
|
162 |
+
wandb: eval/steps_per_second β
|
163 |
+
wandb: eval/wer β
|
164 |
+
wandb: eval_cer β
|
165 |
+
wandb: eval_model_preparation_time β
|
166 |
+
wandb: eval_runtime β
|
167 |
+
wandb: eval_samples β
|
168 |
+
wandb: eval_samples_per_second β
|
169 |
+
wandb: eval_steps_per_second β
|
170 |
+
wandb: eval_wer β
|
171 |
+
wandb: train/global_step ββ
|
172 |
+
wandb:
|
173 |
+
wandb: Run summary:
|
174 |
+
wandb: eval/cer 1.0
|
175 |
+
wandb: eval/loss nan
|
176 |
+
wandb: eval/model_preparation_time 0.0045
|
177 |
+
wandb: eval/runtime 30.9217
|
178 |
+
wandb: eval/samples_per_second 18.498
|
179 |
+
wandb: eval/steps_per_second 1.164
|
180 |
+
wandb: eval/wer 1.0
|
181 |
+
wandb: eval_cer 1.0
|
182 |
+
wandb: eval_loss nan
|
183 |
+
wandb: eval_model_preparation_time 0.0045
|
184 |
+
wandb: eval_runtime 30.9217
|
185 |
+
wandb: eval_samples 572
|
186 |
+
wandb: eval_samples_per_second 18.498
|
187 |
+
wandb: eval_steps_per_second 1.164
|
188 |
+
wandb: eval_wer 1.0
|
189 |
+
wandb: train/global_step 0
|
190 |
+
wandb:
|
191 |
+
wandb: π View run eval_pd2000_s300_shuff100_hindi at: https://wandb.ai/priyanshipal/huggingface/runs/64250v6u
|
192 |
+
wandb: βοΈ View project at: https://wandb.ai/priyanshipal/huggingface
|
193 |
+
wandb: Synced 6 W&B file(s), 0 media file(s), 0 artifact file(s) and 0 other file(s)
|
194 |
+
wandb: Find logs at: ./wandb/run-20240822_152047-64250v6u/logs
|
195 |
+
wandb: WARNING The new W&B backend becomes opt-out in version 0.18.0; try it out with `wandb.require("core")`! See https://wandb.me/wandb-core for more information.
|
indicwav2vec_MUCS_warmup500_s300shuff100_2143522.out
ADDED
The diff for this file is too large to render.
See raw diff
|
|
language_segregated_prediction_texts/evalpredictions_hindi_indicw2v_ad0_3_hd_02_featd_0_2_lr6e-4_warmup2000_s300_shuf500.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
language_segregated_prediction_texts/evalpredictions_hindi_indicw2v_ad0_3_hd_02_featd_0_2_lr6e-4_warmup2000_s400_shuf42.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 1262426580
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:08de7da709cb1c87182ea525062fcd525beeade5ce32613467ec0bcc59e40f04
|
3 |
size 1262426580
|
predictions_indicw2v_ad0_3_hd_02_featd_0_2_lr6e-4_warmup500_s300_shuff100.txt
CHANGED
The diff for this file is too large to render.
See raw diff
|
|
training_args.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 5432
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6414775e5e3dbcfd7bebd7f7c9e2b10fb4f3cb3532a487bef20431df5d198f8a
|
3 |
size 5432
|