ShiWarai commited on
Commit
f7a69f0
·
verified ·
1 Parent(s): 9a86e72

CI: модель обучена и загружена (21d6143)

Browse files
2_Dense/model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d145071cdcd6b749cdaf5753f52e3578e7b0f32154d7c03526c5c9263cc52c80
3
  size 9437272
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1d3013cb8d9c37ce23876ff03e69e7e54f8a54e33b23cb073a50bd26eec3a283
3
  size 9437272
3_Dense/model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6bd10e3240af2263eedc7a91fbedfdfb8b2c0454892f425ccebbcbb20b713368
3
  size 9437272
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:108ce65cf9204f98d306d3e0562e6d8cc9817e7e8a9f0d423e0e26b4dc033445
3
  size 9437272
README.md CHANGED
@@ -5,11 +5,11 @@ tags:
5
  - text-classification
6
  - generated_from_setfit_trainer
7
  widget:
8
- - text: второй робот
9
- - text: давай лапу
10
  - text: привяжи робота 1
11
- - text: быстро перевернись
12
- - text: покрутишься
13
  metrics:
14
  - accuracy
15
  pipeline_tag: text-classification
@@ -34,7 +34,7 @@ The model has been trained using an efficient few-shot learning technique that i
34
  - **Sentence Transformer body:** [google/embeddinggemma-300M](https://huggingface.co/google/embeddinggemma-300M)
35
  - **Classification head:** a [LogisticRegression](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html) instance
36
  - **Maximum Sequence Length:** 2048 tokens
37
- - **Number of Classes:** 13 classes
38
  <!-- - **Training Dataset:** [Unknown](https://huggingface.co/datasets/unknown) -->
39
  <!-- - **Language:** Unknown -->
40
  <!-- - **License:** Unknown -->
@@ -52,6 +52,7 @@ The model has been trained using an efficient few-shot learning technique that i
52
  | silence | <ul><li>'тишина'</li><li>'молчи'</li><li>'молчите'</li></ul> |
53
  | bind | <ul><li>'привяжи робота'</li><li>'привяжи панду'</li><li>'привяжи робота 1'</li></ul> |
54
  | unbind | <ul><li>'отвяжи робота'</li><li>'отвяжи панду'</li><li>'отвяжите робота'</li></ul> |
 
55
  | give_paw | <ul><li>'лапу'</li><li>'дай лапу'</li><li>'дать лапу'</li></ul> |
56
  | stand_at_attention | <ul><li>'равняйсь'</li><li>'равняйся'</li><li>'равняться'</li></ul> |
57
  | dismiss | <ul><li>'отставить'</li><li>'отставь'</li><li>'встать'</li></ul> |
@@ -78,9 +79,9 @@ Then you can load this model and run inference.
78
  from setfit import SetFitModel
79
 
80
  # Download from the 🤗 Hub
81
- model = SetFitModel.from_pretrained("tmp2xh3zlis/panda_commands")
82
  # Run inference
83
- preds = model("давай лапу")
84
  ```
85
 
86
  <!--
@@ -112,7 +113,7 @@ preds = model("давай лапу")
112
  ### Training Set Metrics
113
  | Training set | Min | Median | Max |
114
  |:-------------|:----|:-------|:----|
115
- | Word count | 1 | 2.3566 | 7 |
116
 
117
  | Label | Training Sample Count |
118
  |:-------------------|:----------------------|
@@ -122,13 +123,14 @@ preds = model("давай лапу")
122
  | help | 22 |
123
  | lie_down | 110 |
124
  | reconnect_joystick | 135 |
 
125
  | rotate | 137 |
126
  | run | 106 |
127
  | silence | 27 |
128
  | stand_at_attention | 88 |
129
  | stop_running | 135 |
130
  | unbind | 37 |
131
- | unknown | 477 |
132
 
133
  ### Training Hyperparameters
134
  - batch_size: (256, 256)
@@ -152,11 +154,12 @@ preds = model("давай лапу")
152
  ### Training Results
153
  | Epoch | Step | Training Loss | Validation Loss |
154
  |:------:|:----:|:-------------:|:---------------:|
155
- | 0.0040 | 1 | 0.2421 | - |
156
- | 0.2008 | 50 | 0.0777 | - |
157
- | 0.4016 | 100 | 0.0133 | - |
158
- | 0.6024 | 150 | 0.0069 | - |
159
- | 0.8032 | 200 | 0.0044 | - |
 
160
 
161
  ### Framework Versions
162
  - Python: 3.11.14
 
5
  - text-classification
6
  - generated_from_setfit_trainer
7
  widget:
8
+ - text: заставляю давать лапу
9
+ - text: пора показать лапу
10
  - text: привяжи робота 1
11
+ - text: робот поднимись
12
+ - text: примите положение лежа
13
  metrics:
14
  - accuracy
15
  pipeline_tag: text-classification
 
34
  - **Sentence Transformer body:** [google/embeddinggemma-300M](https://huggingface.co/google/embeddinggemma-300M)
35
  - **Classification head:** a [LogisticRegression](https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LogisticRegression.html) instance
36
  - **Maximum Sequence Length:** 2048 tokens
37
+ - **Number of Classes:** 14 classes
38
  <!-- - **Training Dataset:** [Unknown](https://huggingface.co/datasets/unknown) -->
39
  <!-- - **Language:** Unknown -->
40
  <!-- - **License:** Unknown -->
 
52
  | silence | <ul><li>'тишина'</li><li>'молчи'</li><li>'молчите'</li></ul> |
53
  | bind | <ul><li>'привяжи робота'</li><li>'привяжи панду'</li><li>'привяжи робота 1'</li></ul> |
54
  | unbind | <ul><li>'отвяжи робота'</li><li>'отвяжи панду'</li><li>'отвяжите робота'</li></ul> |
55
+ | report_command | <ul><li>'исправить команду'</li><li>'исправь команду'</li><li>'исправьте команду'</li></ul> |
56
  | give_paw | <ul><li>'лапу'</li><li>'дай лапу'</li><li>'дать лапу'</li></ul> |
57
  | stand_at_attention | <ul><li>'равняйсь'</li><li>'равняйся'</li><li>'равняться'</li></ul> |
58
  | dismiss | <ul><li>'отставить'</li><li>'отставь'</li><li>'встать'</li></ul> |
 
79
  from setfit import SetFitModel
80
 
81
  # Download from the 🤗 Hub
82
+ model = SetFitModel.from_pretrained("tmp8a80_nwv/panda_commands")
83
  # Run inference
84
+ preds = model("робот поднимись")
85
  ```
86
 
87
  <!--
 
113
  ### Training Set Metrics
114
  | Training set | Min | Median | Max |
115
  |:-------------|:----|:-------|:----|
116
+ | Word count | 1 | 2.3696 | 7 |
117
 
118
  | Label | Training Sample Count |
119
  |:-------------------|:----------------------|
 
123
  | help | 22 |
124
  | lie_down | 110 |
125
  | reconnect_joystick | 135 |
126
+ | report_command | 50 |
127
  | rotate | 137 |
128
  | run | 106 |
129
  | silence | 27 |
130
  | stand_at_attention | 88 |
131
  | stop_running | 135 |
132
  | unbind | 37 |
133
+ | unknown | 479 |
134
 
135
  ### Training Hyperparameters
136
  - batch_size: (256, 256)
 
154
  ### Training Results
155
  | Epoch | Step | Training Loss | Validation Loss |
156
  |:------:|:----:|:-------------:|:---------------:|
157
+ | 0.0039 | 1 | 0.2227 | - |
158
+ | 0.1938 | 50 | 0.0799 | - |
159
+ | 0.3876 | 100 | 0.012 | - |
160
+ | 0.5814 | 150 | 0.0052 | - |
161
+ | 0.7752 | 200 | 0.004 | - |
162
+ | 0.9690 | 250 | 0.0041 | - |
163
 
164
  ### Framework Versions
165
  - Python: 3.11.14
config_setfit.json CHANGED
@@ -1,4 +1,5 @@
1
  {
 
2
  "labels": [
3
  "bind",
4
  "dismiss",
@@ -6,6 +7,7 @@
6
  "help",
7
  "lie_down",
8
  "reconnect_joystick",
 
9
  "rotate",
10
  "run",
11
  "silence",
@@ -13,6 +15,5 @@
13
  "stop_running",
14
  "unbind",
15
  "unknown"
16
- ],
17
- "normalize_embeddings": false
18
  }
 
1
  {
2
+ "normalize_embeddings": false,
3
  "labels": [
4
  "bind",
5
  "dismiss",
 
7
  "help",
8
  "lie_down",
9
  "reconnect_joystick",
10
+ "report_command",
11
  "rotate",
12
  "run",
13
  "silence",
 
15
  "stop_running",
16
  "unbind",
17
  "unknown"
18
+ ]
 
19
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1257f0d36413c7f4c09ed585111c026475a8489165d5f4b17e050ba9bf7ab7e5
3
  size 1211486072
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:51077e0b476f30623ed3350f170621b30e45b1b8b3319cc46b4f3b368d298ece
3
  size 1211486072
model_head.pkl CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:81ae3433585d876377807bf1d53a0b881aeb06789c5dc568079c125841e1e781
3
- size 81743
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4ccc026d69f5a499954b81aa5ff6783d0a9add5f3f6c1e200ca58b4d28aeb49d
3
+ size 87959