Instructions to use lysandre/tiny-tapas-random-sqa with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use lysandre/tiny-tapas-random-sqa with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("table-question-answering", model="lysandre/tiny-tapas-random-sqa")# Load model directly from transformers import AutoTokenizer, AutoModelForTableQuestionAnswering tokenizer = AutoTokenizer.from_pretrained("lysandre/tiny-tapas-random-sqa") model = AutoModelForTableQuestionAnswering.from_pretrained("lysandre/tiny-tapas-random-sqa") - Notebooks
- Google Colab
- Kaggle
Commit ·
e57c2d7
1
Parent(s): 03a5d99
SQA is not WTQ
Browse files- config.json +0 -7
config.json
CHANGED
|
@@ -1,11 +1,5 @@
|
|
| 1 |
{
|
| 2 |
"_name_or_path": "/content/drive/MyDrive/Tapas/Models saved using .save_pretrained()/tapas_base_finetuned_sqa_with_reset_and_intermediate_pretraining",
|
| 3 |
-
"aggregation_labels": {
|
| 4 |
-
"0": "NONE",
|
| 5 |
-
"1": "SUM",
|
| 6 |
-
"2": "AVERAGE",
|
| 7 |
-
"3": "COUNT"
|
| 8 |
-
},
|
| 9 |
"aggregation_loss_weight": 1.0,
|
| 10 |
"aggregation_temperature": 1.0,
|
| 11 |
"allow_empty_column_selection": false,
|
|
@@ -32,7 +26,6 @@
|
|
| 32 |
"max_num_rows": 64,
|
| 33 |
"max_position_embeddings": 32,
|
| 34 |
"model_type": "tapas",
|
| 35 |
-
"no_aggregation_label_index": 0,
|
| 36 |
"num_aggregation_labels": 0,
|
| 37 |
"num_attention_heads": 4,
|
| 38 |
"num_hidden_layers": 1,
|
|
|
|
| 1 |
{
|
| 2 |
"_name_or_path": "/content/drive/MyDrive/Tapas/Models saved using .save_pretrained()/tapas_base_finetuned_sqa_with_reset_and_intermediate_pretraining",
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
"aggregation_loss_weight": 1.0,
|
| 4 |
"aggregation_temperature": 1.0,
|
| 5 |
"allow_empty_column_selection": false,
|
|
|
|
| 26 |
"max_num_rows": 64,
|
| 27 |
"max_position_embeddings": 32,
|
| 28 |
"model_type": "tapas",
|
|
|
|
| 29 |
"num_aggregation_labels": 0,
|
| 30 |
"num_attention_heads": 4,
|
| 31 |
"num_hidden_layers": 1,
|