clincolnoz commited on
Commit
bd571b8
1 Parent(s): 53198c5

epoch 100 of 100

Browse files
Files changed (7) hide show
  1. README.md +37 -37
  2. optimizer.pt +1 -1
  3. pytorch_model.bin +1 -1
  4. rng_state.pth +1 -1
  5. scaler.pt +1 -1
  6. scheduler.pt +1 -1
  7. trainer_state.json +0 -0
README.md CHANGED
@@ -84,26 +84,26 @@ You can use this model directly with a pipeline for masked language modeling:
84
  >>> unmasker = pipeline('fill-mask', model='clincolnoz/LessSexistBERT')
85
  >>> unmasker("Hello I'm a [MASK] model.")
86
 
87
- [{'score': 0.6694316267967224,
88
  'token': 3287,
89
  'token_str': 'male',
90
  'sequence': "hello i'm a male model."},
91
- {'score': 0.07414254546165466,
92
- 'token': 10516,
93
- 'token_str': 'fitness',
94
- 'sequence': "hello i'm a fitness model."},
95
- {'score': 0.039137206971645355,
96
  'token': 2931,
97
  'token_str': 'female',
98
  'sequence': "hello i'm a female model."},
99
- {'score': 0.015867002308368683,
 
 
 
 
100
  'token': 3565,
101
  'token_str': 'super',
102
  'sequence': "hello i'm a super model."},
103
- {'score': 0.013910580426454544,
104
- 'token': 2402,
105
- 'token_str': 'young',
106
- 'sequence': "hello i'm a young model."}]
107
  ```
108
 
109
  Here is how to use this model to get the features of a given text in PyTorch:
@@ -112,11 +112,11 @@ Here is how to use this model to get the features of a given text in PyTorch:
112
  from transformers import BertTokenizer, BertModel
113
  tokenizer = BertTokenizer.from_pretrained(
114
  'clincolnoz/LessSexistBERT',
115
- revision='v0.91' # tag name, or branch name, or commit hash
116
  )
117
  model = BertModel.from_pretrained(
118
  'clincolnoz/LessSexistBERT',
119
- revision='v0.91' # tag name, or branch name, or commit hash
120
  )
121
  text = "Replace me by any text you'd like."
122
  encoded_input = tokenizer(text, return_tensors='pt')
@@ -129,12 +129,12 @@ and in TensorFlow:
129
  from transformers import BertTokenizer, TFBertModel
130
  tokenizer = BertTokenizer.from_pretrained(
131
  'clincolnoz/LessSexistBERT',
132
- revision='v0.91' # tag name, or branch name, or commit hash
133
  )
134
  model = TFBertModel.from_pretrained(
135
  'clincolnoz/LessSexistBERT',
136
  from_pt=True,
137
- revision='v0.91' # tag name, or branch name, or commit hash
138
  )
139
  text = "Replace me by any text you'd like."
140
  encoded_input = tokenizer(text, return_tensors='tf')
@@ -151,49 +151,49 @@ neutral, this model can have biased predictions:
151
  >>> unmasker = pipeline('fill-mask', model='clincolnoz/LessSexistBERT')
152
  >>> unmasker("The man worked as a [MASK].")
153
 
154
- [{'score': 0.1168212816119194,
155
- 'token': 7155,
156
- 'token_str': 'scientist',
157
- 'sequence': 'the man worked as a scientist.'},
158
- {'score': 0.11011917889118195,
159
  'token': 3836,
160
  'token_str': 'teacher',
161
  'sequence': 'the man worked as a teacher.'},
162
- {'score': 0.09386853873729706,
163
- 'token': 15893,
164
- 'token_str': 'mechanic',
165
- 'sequence': 'the man worked as a mechanic.'},
166
- {'score': 0.05859819054603577,
167
  'token': 19294,
168
  'token_str': 'therapist',
169
  'sequence': 'the man worked as a therapist.'},
170
- {'score': 0.04985320568084717,
171
- 'token': 3460,
172
- 'token_str': 'doctor',
173
- 'sequence': 'the man worked as a doctor.'}]
 
 
 
 
174
 
175
  >>> unmasker("The woman worked as a [MASK].")
176
 
177
- [{'score': 0.16592034697532654,
178
  'token': 6821,
179
  'token_str': 'nurse',
180
  'sequence': 'the woman worked as a nurse.'},
181
- {'score': 0.1295347660779953,
182
  'token': 3836,
183
  'token_str': 'teacher',
184
  'sequence': 'the woman worked as a teacher.'},
185
- {'score': 0.12351243197917938,
186
  'token': 15812,
187
  'token_str': 'bartender',
188
  'sequence': 'the woman worked as a bartender.'},
189
- {'score': 0.0773676186800003,
190
  'token': 15610,
191
  'token_str': 'waiter',
192
  'sequence': 'the woman worked as a waiter.'},
193
- {'score': 0.05898765102028847,
194
- 'token': 19294,
195
- 'token_str': 'therapist',
196
- 'sequence': 'the woman worked as a therapist.'}]
197
  ```
198
 
199
  This bias may also affect all fine-tuned versions of this model.
 
84
  >>> unmasker = pipeline('fill-mask', model='clincolnoz/LessSexistBERT')
85
  >>> unmasker("Hello I'm a [MASK] model.")
86
 
87
+ [{'score': 0.8502375483512878,
88
  'token': 3287,
89
  'token_str': 'male',
90
  'sequence': "hello i'm a male model."},
91
+ {'score': 0.024909328669309616,
 
 
 
 
92
  'token': 2931,
93
  'token_str': 'female',
94
  'sequence': "hello i'm a female model."},
95
+ {'score': 0.010792124085128307,
96
+ 'token': 2402,
97
+ 'token_str': 'young',
98
+ 'sequence': "hello i'm a young model."},
99
+ {'score': 0.009874437935650349,
100
  'token': 3565,
101
  'token_str': 'super',
102
  'sequence': "hello i'm a super model."},
103
+ {'score': 0.006058268249034882,
104
+ 'token': 2450,
105
+ 'token_str': 'woman',
106
+ 'sequence': "hello i'm a woman model."}]
107
  ```
108
 
109
  Here is how to use this model to get the features of a given text in PyTorch:
 
112
  from transformers import BertTokenizer, BertModel
113
  tokenizer = BertTokenizer.from_pretrained(
114
  'clincolnoz/LessSexistBERT',
115
+ revision='v1.00' # tag name, or branch name, or commit hash
116
  )
117
  model = BertModel.from_pretrained(
118
  'clincolnoz/LessSexistBERT',
119
+ revision='v1.00' # tag name, or branch name, or commit hash
120
  )
121
  text = "Replace me by any text you'd like."
122
  encoded_input = tokenizer(text, return_tensors='pt')
 
129
  from transformers import BertTokenizer, TFBertModel
130
  tokenizer = BertTokenizer.from_pretrained(
131
  'clincolnoz/LessSexistBERT',
132
+ revision='v1.00' # tag name, or branch name, or commit hash
133
  )
134
  model = TFBertModel.from_pretrained(
135
  'clincolnoz/LessSexistBERT',
136
  from_pt=True,
137
+ revision='v1.00' # tag name, or branch name, or commit hash
138
  )
139
  text = "Replace me by any text you'd like."
140
  encoded_input = tokenizer(text, return_tensors='tf')
 
151
  >>> unmasker = pipeline('fill-mask', model='clincolnoz/LessSexistBERT')
152
  >>> unmasker("The man worked as a [MASK].")
153
 
154
+ [{'score': 0.1330045461654663,
 
 
 
 
155
  'token': 3836,
156
  'token_str': 'teacher',
157
  'sequence': 'the man worked as a teacher.'},
158
+ {'score': 0.07733502238988876,
159
+ 'token': 3460,
160
+ 'token_str': 'doctor',
161
+ 'sequence': 'the man worked as a doctor.'},
162
+ {'score': 0.07027694582939148,
163
  'token': 19294,
164
  'token_str': 'therapist',
165
  'sequence': 'the man worked as a therapist.'},
166
+ {'score': 0.06408169120550156,
167
+ 'token': 15893,
168
+ 'token_str': 'mechanic',
169
+ 'sequence': 'the man worked as a mechanic.'},
170
+ {'score': 0.05363383889198303,
171
+ 'token': 8872,
172
+ 'token_str': 'cop',
173
+ 'sequence': 'the man worked as a cop.'}]
174
 
175
  >>> unmasker("The woman worked as a [MASK].")
176
 
177
+ [{'score': 0.3286875784397125,
178
  'token': 6821,
179
  'token_str': 'nurse',
180
  'sequence': 'the woman worked as a nurse.'},
181
+ {'score': 0.1587877869606018,
182
  'token': 3836,
183
  'token_str': 'teacher',
184
  'sequence': 'the woman worked as a teacher.'},
185
+ {'score': 0.058329254388809204,
186
  'token': 15812,
187
  'token_str': 'bartender',
188
  'sequence': 'the woman worked as a bartender.'},
189
+ {'score': 0.040849827229976654,
190
  'token': 15610,
191
  'token_str': 'waiter',
192
  'sequence': 'the woman worked as a waiter.'},
193
+ {'score': 0.04005971923470497,
194
+ 'token': 8872,
195
+ 'token_str': 'cop',
196
+ 'sequence': 'the woman worked as a cop.'}]
197
  ```
198
 
199
  This bias may also affect all fine-tuned versions of this model.
optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:46c5f98fb19784adada7a9ef8a1d66ed8e7715c3eb3bc4ee1c45d15e33a01f00
3
  size 881735429
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e97fbe7a4412d07507baffe07ea41f31af182aaefdf59313bf00e1b4c76d03ca
3
  size 881735429
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2062c3056310d75caee48a33a1de187bc604641b10137bb851006b87d63e3ce4
3
  size 440881865
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ab16e07330bdd4a130b2636fe8e8f817a3aecd5d22f8fe6b2cba94520bed5e04
3
  size 440881865
rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f806318b27c20828409f0dccd94ad9b71022c89b751e20c4f4f36d806af5efa0
3
  size 14575
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:435b4fdbc966ef05734edf53760fa68580c27ab1d93be6afb4e3b5621cec1729
3
  size 14575
scaler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a4a79fbab6f2b22f7975c65cd0526f1901dab93d703d1a305e962b1611ff9571
3
  size 557
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9fc6303fd66a9b8036ce7a71c51d2d8e49b7fdb7672ac6a5e7ce1cde4f120f4f
3
  size 557
scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1b19b6538885f59e299ff3ddd6aaeed9e3ef93d51f3b3d4efc4418fb1f1fd847
3
  size 627
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1244c975ed3dd0a1771d1c9bcf32a1c6267ae79cefa7a2f5acd19ac22198d019
3
  size 627
trainer_state.json CHANGED
The diff for this file is too large to render. See raw diff