jonathanagustin commited on
Commit
dafadaa
1 Parent(s): ef76789

Model save

Browse files
Files changed (3) hide show
  1. README.md +48 -215
  2. model_card.md +234 -31
  3. trainer_state.json +8 -8
README.md CHANGED
@@ -1,235 +1,68 @@
1
  ---
2
- language: en
3
- license: mit
 
 
4
  model-index:
5
- - name: bert-finetuned-uncased
6
- results:
7
- - task:
8
- type: question-answering
9
- dataset:
10
- name: SQuAD v2
11
- type: squad_v2
12
- metrics:
13
- - type: Exact
14
- value: 27.878379516550154
15
- - type: F1
16
- value: 32.12991628283337
17
- - type: Total
18
- value: 11873
19
- - type: Hasans Exact
20
- value: 50.40485829959514
21
- - type: Hasans F1
22
- value: 58.920124160944766
23
- - type: Hasans Total
24
- value: 5928
25
- - type: Noans Exact
26
- value: 5.416316232127839
27
- - type: Noans F1
28
- value: 5.416316232127839
29
- - type: Noans Total
30
- value: 5945
31
- - type: Best Exact
32
- value: 50.11370336056599
33
- - type: Best Exact Thresh
34
- value: 0.0
35
- - type: Best F1
36
- value: 50.11370336056599
37
- - type: Best F1 Thresh
38
- value: 0.0
39
  ---
40
 
41
- # Model Card for Model ID
 
42
 
43
- <!-- Provide a quick summary of what the model is/does. -->
44
 
 
 
 
45
 
 
46
 
47
- ## Model Details
48
 
49
- ### Model Description
50
 
51
- <!-- Provide a longer summary of what this model is. -->
52
 
 
53
 
 
54
 
55
- - **Developed by:** [More Information Needed]
56
- - **Shared by [optional]:** [More Information Needed]
57
- - **Model type:** [More Information Needed]
58
- - **Language(s) (NLP):** en
59
- - **License:** mit
60
- - **Finetuned from model [optional]:** [More Information Needed]
61
 
62
- ### Model Sources [optional]
63
 
64
- <!-- Provide the basic links for the model. -->
 
 
 
 
 
 
 
 
 
65
 
66
- - **Repository:** [More Information Needed]
67
- - **Paper [optional]:** [More Information Needed]
68
- - **Demo [optional]:** [More Information Needed]
69
 
70
- ## Uses
 
 
 
 
 
 
 
 
 
 
 
71
 
72
- <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
73
-
74
- ### Direct Use
75
-
76
- <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
77
-
78
- [More Information Needed]
79
-
80
- ### Downstream Use [optional]
81
-
82
- <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
83
-
84
- [More Information Needed]
85
-
86
- ### Out-of-Scope Use
87
-
88
- <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
89
-
90
- [More Information Needed]
91
-
92
- ## Bias, Risks, and Limitations
93
-
94
- <!-- This section is meant to convey both technical and sociotechnical limitations. -->
95
-
96
- [More Information Needed]
97
-
98
- ### Recommendations
99
-
100
- <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
101
-
102
- Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
103
-
104
- ## How to Get Started with the Model
105
-
106
- Use the code below to get started with the model.
107
-
108
- [More Information Needed]
109
-
110
- ## Training Details
111
-
112
- ### Training Data
113
-
114
- <!-- This should link to a Data Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
115
-
116
- [More Information Needed]
117
-
118
- ### Training Procedure
119
-
120
- <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
121
-
122
- #### Preprocessing [optional]
123
-
124
- [More Information Needed]
125
-
126
-
127
- #### Training Hyperparameters
128
-
129
- - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
130
-
131
- #### Speeds, Sizes, Times [optional]
132
-
133
- <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
134
-
135
- [More Information Needed]
136
-
137
- ## Evaluation
138
-
139
- <!-- This section describes the evaluation protocols and provides the results. -->
140
-
141
- ### Testing Data, Factors & Metrics
142
-
143
- #### Testing Data
144
-
145
- <!-- This should link to a Data Card if possible. -->
146
-
147
- [More Information Needed]
148
-
149
- #### Factors
150
-
151
- <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
152
-
153
- [More Information Needed]
154
-
155
- #### Metrics
156
-
157
- <!-- These are the evaluation metrics being used, ideally with a description of why. -->
158
-
159
- [More Information Needed]
160
-
161
- ### Results
162
-
163
- [More Information Needed]
164
-
165
- #### Summary
166
-
167
-
168
-
169
- ## Model Examination [optional]
170
-
171
- <!-- Relevant interpretability work for the model goes here -->
172
-
173
- [More Information Needed]
174
-
175
- ## Environmental Impact
176
-
177
- <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
178
-
179
- Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
180
-
181
- - **Hardware Type:** [More Information Needed]
182
- - **Hours used:** [More Information Needed]
183
- - **Cloud Provider:** [More Information Needed]
184
- - **Compute Region:** [More Information Needed]
185
- - **Carbon Emitted:** [More Information Needed]
186
-
187
- ## Technical Specifications [optional]
188
-
189
- ### Model Architecture and Objective
190
-
191
- [More Information Needed]
192
-
193
- ### Compute Infrastructure
194
-
195
- [More Information Needed]
196
-
197
- #### Hardware
198
-
199
- [More Information Needed]
200
-
201
- #### Software
202
-
203
- [More Information Needed]
204
-
205
- ## Citation [optional]
206
-
207
- <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
208
-
209
- **BibTeX:**
210
-
211
- [More Information Needed]
212
-
213
- **APA:**
214
-
215
- [More Information Needed]
216
-
217
- ## Glossary [optional]
218
-
219
- <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
220
-
221
- [More Information Needed]
222
-
223
- ## More Information [optional]
224
-
225
- [More Information Needed]
226
-
227
- ## Model Card Authors [optional]
228
-
229
- [More Information Needed]
230
-
231
- ## Model Card Contact
232
-
233
- [More Information Needed]
234
 
 
235
 
 
 
 
 
 
1
  ---
2
+ tags:
3
+ - generated_from_trainer
4
+ datasets:
5
+ - squad_v2
6
  model-index:
7
+ - name: bert-finetuned-uncased-squad_v2
8
+ results: []
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
9
  ---
10
 
11
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
12
+ should probably proofread and complete it, then remove this comment. -->
13
 
14
+ # bert-finetuned-uncased-squad_v2
15
 
16
+ This model was trained from scratch on the squad_v2 dataset.
17
+ It achieves the following results on the evaluation set:
18
+ - Loss: 1.2041
19
 
20
+ ## Model description
21
 
22
+ More information needed
23
 
24
+ ## Intended uses & limitations
25
 
26
+ More information needed
27
 
28
+ ## Training and evaluation data
29
 
30
+ More information needed
31
 
32
+ ## Training procedure
 
 
 
 
 
33
 
34
+ ### Training hyperparameters
35
 
36
+ The following hyperparameters were used during training:
37
+ - learning_rate: 2e-05
38
+ - train_batch_size: 128
39
+ - eval_batch_size: 128
40
+ - seed: 42
41
+ - gradient_accumulation_steps: 4
42
+ - total_train_batch_size: 512
43
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
44
+ - lr_scheduler_type: linear
45
+ - num_epochs: 1
46
 
47
+ ### Training results
 
 
48
 
49
+ | Training Loss | Epoch | Step | Validation Loss |
50
+ |:-------------:|:-----:|:----:|:---------------:|
51
+ | 3.2307 | 0.2 | 100 | 1.8959 |
52
+ | 1.9581 | 0.39 | 200 | 1.4856 |
53
+ | 1.6358 | 0.59 | 300 | 1.3948 |
54
+ | 1.4964 | 0.78 | 400 | 1.2934 |
55
+ | 1.4169 | 0.98 | 500 | 1.2605 |
56
+ | 1.327 | 1.18 | 600 | 1.2218 |
57
+ | 1.2763 | 1.37 | 700 | 1.2539 |
58
+ | 1.2755 | 1.57 | 800 | 1.2090 |
59
+ | 1.251 | 1.76 | 900 | 1.2041 |
60
+ | 1.229 | 1.96 | 1000 | 1.2159 |
61
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
62
 
63
+ ### Framework versions
64
 
65
+ - Transformers 4.34.1
66
+ - Pytorch 2.1.0+cu118
67
+ - Datasets 2.14.5
68
+ - Tokenizers 0.14.1
model_card.md CHANGED
@@ -1,32 +1,235 @@
1
-
2
- ---
3
- language:
4
- - en
5
- tags:
6
- - question-answering
7
- - fine-tuned
8
- datasets:
9
- - squad_v2
 
 
10
  metrics:
11
- - squad
12
- ---
13
-
14
- ## bert-finetuned-uncased
15
-
16
- This model is a fine-tuned version of bert-base-uncased for Question Answering on the SQuAD v2 dataset.
17
-
18
- ## Evaluation Results
19
- - Exact Match: 27.878379516550154
20
- - F1 Score: 32.12991628283337
21
- - Total: 11873
22
- - Has Answer Exact: 50.40485829959514
23
- - Has Answer F1: 58.920124160944766
24
- - Has Answer Total: 5928
25
- - No Answer Exact: 5.416316232127839
26
- - No Answer F1: 5.416316232127839
27
- - No Answer Total: 5945
28
- - Best Exact: 50.11370336056599
29
- - Best Exact Threshold: 0.0
30
- - Best F1: 50.11370336056599
31
- - Best F1 Threshold: 0.0
32
-
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language: en
3
+ license: mit
4
+ model-index:
5
+ - name: bert-finetuned-uncased
6
+ results:
7
+ - task:
8
+ type: question-answering
9
+ dataset:
10
+ name: SQuAD v2
11
+ type: squad_v2
12
  metrics:
13
+ - type: Exact
14
+ value: 27.878379516550154
15
+ - type: F1
16
+ value: 32.12991628283337
17
+ - type: Total
18
+ value: 11873
19
+ - type: Hasans Exact
20
+ value: 50.40485829959514
21
+ - type: Hasans F1
22
+ value: 58.920124160944766
23
+ - type: Hasans Total
24
+ value: 5928
25
+ - type: Noans Exact
26
+ value: 5.416316232127839
27
+ - type: Noans F1
28
+ value: 5.416316232127839
29
+ - type: Noans Total
30
+ value: 5945
31
+ - type: Best Exact
32
+ value: 50.11370336056599
33
+ - type: Best Exact Thresh
34
+ value: 0.0
35
+ - type: Best F1
36
+ value: 50.11370336056599
37
+ - type: Best F1 Thresh
38
+ value: 0.0
39
+ ---
40
+
41
+ # Model Card for Model ID
42
+
43
+ <!-- Provide a quick summary of what the model is/does. -->
44
+
45
+
46
+
47
+ ## Model Details
48
+
49
+ ### Model Description
50
+
51
+ <!-- Provide a longer summary of what this model is. -->
52
+
53
+
54
+
55
+ - **Developed by:** [More Information Needed]
56
+ - **Shared by [optional]:** [More Information Needed]
57
+ - **Model type:** [More Information Needed]
58
+ - **Language(s) (NLP):** en
59
+ - **License:** mit
60
+ - **Finetuned from model [optional]:** [More Information Needed]
61
+
62
+ ### Model Sources [optional]
63
+
64
+ <!-- Provide the basic links for the model. -->
65
+
66
+ - **Repository:** [More Information Needed]
67
+ - **Paper [optional]:** [More Information Needed]
68
+ - **Demo [optional]:** [More Information Needed]
69
+
70
+ ## Uses
71
+
72
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
73
+
74
+ ### Direct Use
75
+
76
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
77
+
78
+ [More Information Needed]
79
+
80
+ ### Downstream Use [optional]
81
+
82
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
83
+
84
+ [More Information Needed]
85
+
86
+ ### Out-of-Scope Use
87
+
88
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
89
+
90
+ [More Information Needed]
91
+
92
+ ## Bias, Risks, and Limitations
93
+
94
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
95
+
96
+ [More Information Needed]
97
+
98
+ ### Recommendations
99
+
100
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
101
+
102
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
103
+
104
+ ## How to Get Started with the Model
105
+
106
+ Use the code below to get started with the model.
107
+
108
+ [More Information Needed]
109
+
110
+ ## Training Details
111
+
112
+ ### Training Data
113
+
114
+ <!-- This should link to a Data Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
115
+
116
+ [More Information Needed]
117
+
118
+ ### Training Procedure
119
+
120
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
121
+
122
+ #### Preprocessing [optional]
123
+
124
+ [More Information Needed]
125
+
126
+
127
+ #### Training Hyperparameters
128
+
129
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
130
+
131
+ #### Speeds, Sizes, Times [optional]
132
+
133
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
134
+
135
+ [More Information Needed]
136
+
137
+ ## Evaluation
138
+
139
+ <!-- This section describes the evaluation protocols and provides the results. -->
140
+
141
+ ### Testing Data, Factors & Metrics
142
+
143
+ #### Testing Data
144
+
145
+ <!-- This should link to a Data Card if possible. -->
146
+
147
+ [More Information Needed]
148
+
149
+ #### Factors
150
+
151
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
152
+
153
+ [More Information Needed]
154
+
155
+ #### Metrics
156
+
157
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
158
+
159
+ [More Information Needed]
160
+
161
+ ### Results
162
+
163
+ [More Information Needed]
164
+
165
+ #### Summary
166
+
167
+
168
+
169
+ ## Model Examination [optional]
170
+
171
+ <!-- Relevant interpretability work for the model goes here -->
172
+
173
+ [More Information Needed]
174
+
175
+ ## Environmental Impact
176
+
177
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
178
+
179
+ Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
180
+
181
+ - **Hardware Type:** [More Information Needed]
182
+ - **Hours used:** [More Information Needed]
183
+ - **Cloud Provider:** [More Information Needed]
184
+ - **Compute Region:** [More Information Needed]
185
+ - **Carbon Emitted:** [More Information Needed]
186
+
187
+ ## Technical Specifications [optional]
188
+
189
+ ### Model Architecture and Objective
190
+
191
+ [More Information Needed]
192
+
193
+ ### Compute Infrastructure
194
+
195
+ [More Information Needed]
196
+
197
+ #### Hardware
198
+
199
+ [More Information Needed]
200
+
201
+ #### Software
202
+
203
+ [More Information Needed]
204
+
205
+ ## Citation [optional]
206
+
207
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
208
+
209
+ **BibTeX:**
210
+
211
+ [More Information Needed]
212
+
213
+ **APA:**
214
+
215
+ [More Information Needed]
216
+
217
+ ## Glossary [optional]
218
+
219
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
220
+
221
+ [More Information Needed]
222
+
223
+ ## More Information [optional]
224
+
225
+ [More Information Needed]
226
+
227
+ ## Model Card Authors [optional]
228
+
229
+ [More Information Needed]
230
+
231
+ ## Model Card Contact
232
+
233
+ [More Information Needed]
234
+
235
+
trainer_state.json CHANGED
@@ -153,22 +153,22 @@
153
  "step": 1000,
154
  "total_flos": 6.687707580928205e+16,
155
  "train_loss": 0.0,
156
- "train_runtime": 0.5139,
157
- "train_samples_per_second": 507861.208,
158
- "train_steps_per_second": 992.35
159
  },
160
  {
161
  "epoch": 1.96,
162
  "eval_loss": 1.204106092453003,
163
- "eval_runtime": 17.1369,
164
- "eval_samples_per_second": 698.436,
165
- "eval_steps_per_second": 5.485,
166
  "step": 1000
167
  }
168
  ],
169
  "logging_steps": 100,
170
- "max_steps": 510,
171
- "num_train_epochs": 2,
172
  "save_steps": 100,
173
  "total_flos": 6.687707580928205e+16,
174
  "trial_name": null,
 
153
  "step": 1000,
154
  "total_flos": 6.687707580928205e+16,
155
  "train_loss": 0.0,
156
+ "train_runtime": 0.4966,
157
+ "train_samples_per_second": 262788.244,
158
+ "train_steps_per_second": 513.482
159
  },
160
  {
161
  "epoch": 1.96,
162
  "eval_loss": 1.204106092453003,
163
+ "eval_runtime": 17.1537,
164
+ "eval_samples_per_second": 697.752,
165
+ "eval_steps_per_second": 5.48,
166
  "step": 1000
167
  }
168
  ],
169
  "logging_steps": 100,
170
+ "max_steps": 255,
171
+ "num_train_epochs": 1,
172
  "save_steps": 100,
173
  "total_flos": 6.687707580928205e+16,
174
  "trial_name": null,