End of training
Browse files- pytorch_model.bin +1 -1
- runs/{May17_01-31-58_0d0f38c34c7b → xlsrDe-d}/1652751127.5893614/events.out.tfevents.1652751127.0d0f38c34c7b.1404.1 +0 -0
- runs/{May17_01-31-58_0d0f38c34c7b → xlsrDe-d}/events.out.tfevents.1652751127.0d0f38c34c7b.1404.0 +2 -2
- xlsrDe-d/config.json +86 -0
- xlsrDe-d/optimizer.pt +3 -0
- xlsrDe-d/preprocessor_config.json +9 -0
- xlsrDe-d/pytorch_model.bin +3 -0
- xlsrDe-d/rng_state.pth +3 -0
- xlsrDe-d/scaler.pt +3 -0
- xlsrDe-d/scheduler.pt +3 -0
- xlsrDe-d/trainer_state.json +256 -0
- xlsrDe-d/training_args.bin +3 -0
pytorch_model.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 1262079473
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8abdd1bbd27fa6778d8c7b7df3d12088c8c513356992eb0f7424a65a023ce0b3
|
3 |
size 1262079473
|
runs/{May17_01-31-58_0d0f38c34c7b → xlsrDe-d}/1652751127.5893614/events.out.tfevents.1652751127.0d0f38c34c7b.1404.1
RENAMED
File without changes
|
runs/{May17_01-31-58_0d0f38c34c7b → xlsrDe-d}/events.out.tfevents.1652751127.0d0f38c34c7b.1404.0
RENAMED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e3ff1acc1702f8b2671555185da39db837589d401e1628e7a8984a6f415d03f9
|
3 |
+
size 12275
|
xlsrDe-d/config.json
ADDED
@@ -0,0 +1,86 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "jonatasgrosman/wav2vec2-large-xlsr-53-german",
|
3 |
+
"activation_dropout": 0.05,
|
4 |
+
"apply_spec_augment": true,
|
5 |
+
"architectures": [
|
6 |
+
"Wav2Vec2ForCTC"
|
7 |
+
],
|
8 |
+
"attention_dropout": 0.5,
|
9 |
+
"bos_token_id": 1,
|
10 |
+
"classifier_proj_size": 256,
|
11 |
+
"codevector_dim": 768,
|
12 |
+
"contrastive_logits_temperature": 0.1,
|
13 |
+
"conv_bias": true,
|
14 |
+
"conv_dim": [
|
15 |
+
512,
|
16 |
+
512,
|
17 |
+
512,
|
18 |
+
512,
|
19 |
+
512,
|
20 |
+
512,
|
21 |
+
512
|
22 |
+
],
|
23 |
+
"conv_kernel": [
|
24 |
+
10,
|
25 |
+
3,
|
26 |
+
3,
|
27 |
+
3,
|
28 |
+
3,
|
29 |
+
2,
|
30 |
+
2
|
31 |
+
],
|
32 |
+
"conv_stride": [
|
33 |
+
5,
|
34 |
+
2,
|
35 |
+
2,
|
36 |
+
2,
|
37 |
+
2,
|
38 |
+
2,
|
39 |
+
2
|
40 |
+
],
|
41 |
+
"ctc_loss_reduction": "mean",
|
42 |
+
"ctc_zero_infinity": true,
|
43 |
+
"diversity_loss_weight": 0.1,
|
44 |
+
"do_stable_layer_norm": true,
|
45 |
+
"eos_token_id": 2,
|
46 |
+
"feat_extract_activation": "gelu",
|
47 |
+
"feat_extract_dropout": 0.0,
|
48 |
+
"feat_extract_norm": "layer",
|
49 |
+
"feat_proj_dropout": 0.0,
|
50 |
+
"feat_quantizer_dropout": 0.0,
|
51 |
+
"final_dropout": 0.0,
|
52 |
+
"hidden_act": "gelu",
|
53 |
+
"hidden_dropout": 0.0,
|
54 |
+
"hidden_size": 1024,
|
55 |
+
"initializer_range": 0.02,
|
56 |
+
"intermediate_size": 4096,
|
57 |
+
"layer_norm_eps": 1e-05,
|
58 |
+
"layerdrop": 0.0,
|
59 |
+
"mask_channel_length": 10,
|
60 |
+
"mask_channel_min_space": 1,
|
61 |
+
"mask_channel_other": 0.0,
|
62 |
+
"mask_channel_prob": 0.0,
|
63 |
+
"mask_channel_selection": "static",
|
64 |
+
"mask_feature_length": 10,
|
65 |
+
"mask_feature_prob": 0.0,
|
66 |
+
"mask_time_length": 10,
|
67 |
+
"mask_time_min_space": 1,
|
68 |
+
"mask_time_other": 0.0,
|
69 |
+
"mask_time_prob": 0.05,
|
70 |
+
"mask_time_selection": "static",
|
71 |
+
"model_type": "wav2vec2",
|
72 |
+
"num_attention_heads": 16,
|
73 |
+
"num_codevector_groups": 2,
|
74 |
+
"num_codevectors_per_group": 320,
|
75 |
+
"num_conv_pos_embedding_groups": 16,
|
76 |
+
"num_conv_pos_embeddings": 128,
|
77 |
+
"num_feat_extract_layers": 7,
|
78 |
+
"num_hidden_layers": 24,
|
79 |
+
"num_negatives": 100,
|
80 |
+
"pad_token_id": 24,
|
81 |
+
"proj_codevector_dim": 768,
|
82 |
+
"torch_dtype": "float32",
|
83 |
+
"transformers_version": "4.11.3",
|
84 |
+
"use_weighted_layer_sum": false,
|
85 |
+
"vocab_size": 38
|
86 |
+
}
|
xlsrDe-d/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:faa60cada6e65ce1c3d5a6a916782d003fc163fb3a48895c8783d6cc3f66a6d5
|
3 |
+
size 2490370577
|
xlsrDe-d/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": 0.0,
|
7 |
+
"return_attention_mask": true,
|
8 |
+
"sampling_rate": 16000
|
9 |
+
}
|
xlsrDe-d/pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3d79e5fd2754fa90ea4955d2e1eb959dac8e0055d81761ad729d3ffbfd110abc
|
3 |
+
size 1262079473
|
xlsrDe-d/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:43b183631dae7312fa06471905adfa9a1e0795c11a6c4540a4697d12e37e632d
|
3 |
+
size 14567
|
xlsrDe-d/scaler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:378256d35f3510dc9e76f4371bc34b02a30c7a0699ed25a2ad3696ff0f439dc5
|
3 |
+
size 559
|
xlsrDe-d/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f8fd6b64173700554787bd96b3a97002f204277035a984524a8a564f4da1aa28
|
3 |
+
size 623
|
xlsrDe-d/trainer_state.json
ADDED
@@ -0,0 +1,256 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 799.8,
|
5 |
+
"global_step": 1600,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 49.8,
|
12 |
+
"learning_rate": 5.6999999999999996e-05,
|
13 |
+
"loss": 63.303,
|
14 |
+
"step": 100
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 49.8,
|
18 |
+
"eval_loss": 3.881598949432373,
|
19 |
+
"eval_runtime": 1.6151,
|
20 |
+
"eval_samples_per_second": 33.434,
|
21 |
+
"eval_steps_per_second": 4.334,
|
22 |
+
"eval_wer": 1.0,
|
23 |
+
"step": 100
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 99.8,
|
27 |
+
"learning_rate": 0.000117,
|
28 |
+
"loss": 1.0894,
|
29 |
+
"step": 200
|
30 |
+
},
|
31 |
+
{
|
32 |
+
"epoch": 99.8,
|
33 |
+
"eval_loss": 6.306798458099365,
|
34 |
+
"eval_runtime": 1.6201,
|
35 |
+
"eval_samples_per_second": 33.33,
|
36 |
+
"eval_steps_per_second": 4.321,
|
37 |
+
"eval_wer": 1.0289855072463767,
|
38 |
+
"step": 200
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 149.8,
|
42 |
+
"learning_rate": 0.00017699999999999997,
|
43 |
+
"loss": 0.2105,
|
44 |
+
"step": 300
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 149.8,
|
48 |
+
"eval_loss": 7.806252479553223,
|
49 |
+
"eval_runtime": 1.6345,
|
50 |
+
"eval_samples_per_second": 33.037,
|
51 |
+
"eval_steps_per_second": 4.283,
|
52 |
+
"eval_wer": 1.0,
|
53 |
+
"step": 300
|
54 |
+
},
|
55 |
+
{
|
56 |
+
"epoch": 199.8,
|
57 |
+
"learning_rate": 0.000237,
|
58 |
+
"loss": 0.1022,
|
59 |
+
"step": 400
|
60 |
+
},
|
61 |
+
{
|
62 |
+
"epoch": 199.8,
|
63 |
+
"eval_loss": 8.66800594329834,
|
64 |
+
"eval_runtime": 1.6359,
|
65 |
+
"eval_samples_per_second": 33.01,
|
66 |
+
"eval_steps_per_second": 4.279,
|
67 |
+
"eval_wer": 1.0289855072463767,
|
68 |
+
"step": 400
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 249.8,
|
72 |
+
"learning_rate": 0.00029699999999999996,
|
73 |
+
"loss": 0.0992,
|
74 |
+
"step": 500
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 249.8,
|
78 |
+
"eval_loss": 10.373652458190918,
|
79 |
+
"eval_runtime": 1.6295,
|
80 |
+
"eval_samples_per_second": 33.139,
|
81 |
+
"eval_steps_per_second": 4.296,
|
82 |
+
"eval_wer": 1.0144927536231885,
|
83 |
+
"step": 500
|
84 |
+
},
|
85 |
+
{
|
86 |
+
"epoch": 299.8,
|
87 |
+
"learning_rate": 0.0002740909090909091,
|
88 |
+
"loss": 0.0732,
|
89 |
+
"step": 600
|
90 |
+
},
|
91 |
+
{
|
92 |
+
"epoch": 299.8,
|
93 |
+
"eval_loss": 8.226058006286621,
|
94 |
+
"eval_runtime": 1.6292,
|
95 |
+
"eval_samples_per_second": 33.146,
|
96 |
+
"eval_steps_per_second": 4.297,
|
97 |
+
"eval_wer": 1.0,
|
98 |
+
"step": 600
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 349.8,
|
102 |
+
"learning_rate": 0.0002468181818181818,
|
103 |
+
"loss": 0.0419,
|
104 |
+
"step": 700
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 349.8,
|
108 |
+
"eval_loss": 9.197142601013184,
|
109 |
+
"eval_runtime": 1.6246,
|
110 |
+
"eval_samples_per_second": 33.239,
|
111 |
+
"eval_steps_per_second": 4.309,
|
112 |
+
"eval_wer": 1.0289855072463767,
|
113 |
+
"step": 700
|
114 |
+
},
|
115 |
+
{
|
116 |
+
"epoch": 399.8,
|
117 |
+
"learning_rate": 0.00021954545454545452,
|
118 |
+
"loss": 0.0273,
|
119 |
+
"step": 800
|
120 |
+
},
|
121 |
+
{
|
122 |
+
"epoch": 399.8,
|
123 |
+
"eval_loss": 9.849931716918945,
|
124 |
+
"eval_runtime": 1.6419,
|
125 |
+
"eval_samples_per_second": 32.888,
|
126 |
+
"eval_steps_per_second": 4.263,
|
127 |
+
"eval_wer": 1.0434782608695652,
|
128 |
+
"step": 800
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 449.8,
|
132 |
+
"learning_rate": 0.00019227272727272723,
|
133 |
+
"loss": 0.0226,
|
134 |
+
"step": 900
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 449.8,
|
138 |
+
"eval_loss": 10.22618293762207,
|
139 |
+
"eval_runtime": 1.6076,
|
140 |
+
"eval_samples_per_second": 33.591,
|
141 |
+
"eval_steps_per_second": 4.354,
|
142 |
+
"eval_wer": 1.0144927536231885,
|
143 |
+
"step": 900
|
144 |
+
},
|
145 |
+
{
|
146 |
+
"epoch": 499.8,
|
147 |
+
"learning_rate": 0.000165,
|
148 |
+
"loss": 0.0182,
|
149 |
+
"step": 1000
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 499.8,
|
153 |
+
"eval_loss": 10.209068298339844,
|
154 |
+
"eval_runtime": 1.609,
|
155 |
+
"eval_samples_per_second": 33.562,
|
156 |
+
"eval_steps_per_second": 4.351,
|
157 |
+
"eval_wer": 1.0289855072463767,
|
158 |
+
"step": 1000
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 549.8,
|
162 |
+
"learning_rate": 0.0001377272727272727,
|
163 |
+
"loss": 0.0099,
|
164 |
+
"step": 1100
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 549.8,
|
168 |
+
"eval_loss": 10.943706512451172,
|
169 |
+
"eval_runtime": 1.6307,
|
170 |
+
"eval_samples_per_second": 33.115,
|
171 |
+
"eval_steps_per_second": 4.293,
|
172 |
+
"eval_wer": 1.0,
|
173 |
+
"step": 1100
|
174 |
+
},
|
175 |
+
{
|
176 |
+
"epoch": 599.8,
|
177 |
+
"learning_rate": 0.00011045454545454545,
|
178 |
+
"loss": 0.0046,
|
179 |
+
"step": 1200
|
180 |
+
},
|
181 |
+
{
|
182 |
+
"epoch": 599.8,
|
183 |
+
"eval_loss": 10.346452713012695,
|
184 |
+
"eval_runtime": 1.613,
|
185 |
+
"eval_samples_per_second": 33.478,
|
186 |
+
"eval_steps_per_second": 4.34,
|
187 |
+
"eval_wer": 1.0,
|
188 |
+
"step": 1200
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 649.8,
|
192 |
+
"learning_rate": 8.318181818181818e-05,
|
193 |
+
"loss": 0.0108,
|
194 |
+
"step": 1300
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 649.8,
|
198 |
+
"eval_loss": 9.559195518493652,
|
199 |
+
"eval_runtime": 1.6215,
|
200 |
+
"eval_samples_per_second": 33.302,
|
201 |
+
"eval_steps_per_second": 4.317,
|
202 |
+
"eval_wer": 1.0,
|
203 |
+
"step": 1300
|
204 |
+
},
|
205 |
+
{
|
206 |
+
"epoch": 699.8,
|
207 |
+
"learning_rate": 5.590909090909091e-05,
|
208 |
+
"loss": 0.006,
|
209 |
+
"step": 1400
|
210 |
+
},
|
211 |
+
{
|
212 |
+
"epoch": 699.8,
|
213 |
+
"eval_loss": 10.173698425292969,
|
214 |
+
"eval_runtime": 1.6283,
|
215 |
+
"eval_samples_per_second": 33.164,
|
216 |
+
"eval_steps_per_second": 4.299,
|
217 |
+
"eval_wer": 1.0144927536231885,
|
218 |
+
"step": 1400
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 749.8,
|
222 |
+
"learning_rate": 2.8636363636363634e-05,
|
223 |
+
"loss": 0.003,
|
224 |
+
"step": 1500
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 749.8,
|
228 |
+
"eval_loss": 9.914641380310059,
|
229 |
+
"eval_runtime": 1.6245,
|
230 |
+
"eval_samples_per_second": 33.241,
|
231 |
+
"eval_steps_per_second": 4.309,
|
232 |
+
"eval_wer": 1.0144927536231885,
|
233 |
+
"step": 1500
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 799.8,
|
237 |
+
"learning_rate": 1.3636363636363634e-06,
|
238 |
+
"loss": 0.0023,
|
239 |
+
"step": 1600
|
240 |
+
},
|
241 |
+
{
|
242 |
+
"epoch": 799.8,
|
243 |
+
"eval_loss": 10.248388290405273,
|
244 |
+
"eval_runtime": 1.6125,
|
245 |
+
"eval_samples_per_second": 33.488,
|
246 |
+
"eval_steps_per_second": 4.341,
|
247 |
+
"eval_wer": 1.0144927536231885,
|
248 |
+
"step": 1600
|
249 |
+
}
|
250 |
+
],
|
251 |
+
"max_steps": 1600,
|
252 |
+
"num_train_epochs": 800,
|
253 |
+
"total_flos": 6.256122688468838e+18,
|
254 |
+
"trial_name": null,
|
255 |
+
"trial_params": null
|
256 |
+
}
|
xlsrDe-d/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e9eb099522910f5f0deff4d2bd83132cc3a06dd1e0148901d06b58b6760016e1
|
3 |
+
size 2799
|