asadnaqvi commited on
Commit
cf99a0b
1 Parent(s): d9d851e

Add SetFit ABSA model

Browse files
Files changed (4) hide show
  1. README.md +21 -23
  2. config_setfit.json +2 -2
  3. model.safetensors +1 -1
  4. model_head.pkl +1 -1
README.md CHANGED
@@ -10,22 +10,20 @@ base_model: BAAI/bge-small-en-v1.5
10
  metrics:
11
  - accuracy
12
  widget:
13
- - text: to be a nightmare, said retired:The Type 096s are going to be a nightmare,
14
- said retired submariner and naval technical intelligence analyst Christopher Carlson,
15
- one of the researchers.
16
- - text: census as an independent exercise.:In fact, the government of Bihar has recently
17
- taken up the caste census as an independent exercise.
18
- - text: to Moscow's Improved Akula boats.:Carlson told Reuters he did not believe
19
- China had obtained Russia's 'crown jewels' - its very latest technology - but
20
- would be producing a submarine stealthy enough to compare to Moscow's Improved
21
- Akula boats.
22
- - text: staging fully armed nuclear deterrence patrols with its older:The Chinese
23
- navy is routinely staging fully armed nuclear deterrence patrols with its older
24
- Type 094 boats out of Hainan Island in the South China Sea, the Pentagon said
25
- in November, much like patrols operated for years by the United States, Britain,
26
- Russia, and France.
27
- - text: Sanjeev Chopra is a former:Sanjeev Chopra is a former IAS officer and Festival
28
- Director of Valley of Words.
29
  pipeline_tag: text-classification
30
  inference: false
31
  model-index:
@@ -40,7 +38,7 @@ model-index:
40
  split: test
41
  metrics:
42
  - type: accuracy
43
- value: 0.8387096774193549
44
  name: Accuracy
45
  ---
46
 
@@ -93,7 +91,7 @@ This model was trained within the context of a larger system for ABSA, which loo
93
  ### Metrics
94
  | Label | Accuracy |
95
  |:--------|:---------|
96
- | **all** | 0.8387 |
97
 
98
  ## Uses
99
 
@@ -177,11 +175,11 @@ preds = model("The food was great, but the venue is just way too busy.")
177
  ### Training Results
178
  | Epoch | Step | Training Loss | Validation Loss |
179
  |:----------:|:------:|:-------------:|:---------------:|
180
- | 0.0217 | 1 | 0.2604 | - |
181
- | **1.0870** | **50** | **0.0642** | **0.3479** |
182
- | 2.1739 | 100 | 0.0249 | 0.3974 |
183
- | 3.2609 | 150 | 0.0146 | 0.4341 |
184
- | 4.3478 | 200 | 0.0033 | 0.4358 |
185
 
186
  * The bold row denotes the saved checkpoint.
187
  ### Framework Versions
 
10
  metrics:
11
  - accuracy
12
  widget:
13
+ - text: mine employment is starker at mid-:The future of coal mine employment is starker
14
+ at mid-century.
15
+ - text: are considered relatively noisy - a major:But the Type 094s, which carry China's
16
+ most advanced submarine-launched JL-3 missile, are considered relatively noisy
17
+ - a major handicap for military submarines.
18
+ - text: March this year raised an objection, saying that:Initially, the registrar's
19
+ office of the IHC in March this year raised an objection, saying that the high
20
+ court was not an appropriate forum and asked the petitioner to approach the relevant
21
+ authorities.
22
+ - text: mine closures is West Bengal.:Runa Sarkar, a professor at the Indian Institute
23
+ of Management Calcutta, said the coal mining region most affected by mine closures
24
+ is West Bengal.
25
+ - text: FIA DG to proceed in accordance with the law.:When the petition was heard
26
+ by the chief justice, he asked the FIA DG to proceed in accordance with the law.
 
 
27
  pipeline_tag: text-classification
28
  inference: false
29
  model-index:
 
38
  split: test
39
  metrics:
40
  - type: accuracy
41
+ value: 0.7065217391304348
42
  name: Accuracy
43
  ---
44
 
 
91
  ### Metrics
92
  | Label | Accuracy |
93
  |:--------|:---------|
94
+ | **all** | 0.7065 |
95
 
96
  ## Uses
97
 
 
175
  ### Training Results
176
  | Epoch | Step | Training Loss | Validation Loss |
177
  |:----------:|:------:|:-------------:|:---------------:|
178
+ | 0.0217 | 1 | 0.2599 | - |
179
+ | **1.0870** | **50** | **0.0608** | **0.3526** |
180
+ | 2.1739 | 100 | 0.0253 | 0.4091 |
181
+ | 3.2609 | 150 | 0.0159 | 0.4497 |
182
+ | 4.3478 | 200 | 0.0035 | 0.4437 |
183
 
184
  * The bold row denotes the saved checkpoint.
185
  ### Framework Versions
config_setfit.json CHANGED
@@ -5,7 +5,7 @@
5
  "Negative",
6
  "Positive"
7
  ],
8
- "spacy_model": "en_core_web_lg",
9
  "normalize_embeddings": false,
10
- "span_context": 3
 
11
  }
 
5
  "Negative",
6
  "Positive"
7
  ],
 
8
  "normalize_embeddings": false,
9
+ "span_context": 3,
10
+ "spacy_model": "en_core_web_lg"
11
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d516dab61975def5ed7af94623f6e76b0cf8eb7b8332ef31a6ea942eed896cc9
3
  size 133462128
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8b50fb1bc040ac672ac71f796586a6d07bc3d9f0d110083cc4af858de9828b5b
3
  size 133462128
model_head.pkl CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:904885f87a2ed02f4cbd2bd8e8bca759d569cc80af225a44d28a852dea535c42
3
  size 13335
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d3f74a1b2f81d58e52070703c6970427dbf0197a5248fafb502dc9bf36046414
3
  size 13335