Upload 10 files

1_Pooling/config.json

@@ -1,9 +1,9 @@
 {
   "word_embedding_dimension": 1024,
   "pooling_mode_cls_token": false,
-  "pooling_mode_mean_tokens": false,
+  "pooling_mode_mean_tokens": true,
   "pooling_mode_max_tokens": false,
   "pooling_mode_mean_sqrt_len_tokens": false,
-  "pooling_mode_weightedmean_tokens": true,
+  "pooling_mode_weightedmean_tokens": false,
   "pooling_mode_lasttoken": false
 }

README.md

@@ -4,6 +4,7 @@ tags:
 - sentence-transformers
 - feature-extraction
 - sentence-similarity
+- transformers
 ---
 
 # {MODEL_NAME}
@@ -33,6 +34,44 @@ print(embeddings)
 
 
 
+## Usage (HuggingFace Transformers)
+Without [sentence-transformers](https://www.SBERT.net), you can use the model like this: First, you pass your input through the transformer model, then you have to apply the right pooling-operation on-top of the contextualized word embeddings.
+
+```python
+from transformers import AutoTokenizer, AutoModel
+import torch
+
+
+#Mean Pooling - Take attention mask into account for correct averaging
+def mean_pooling(model_output, attention_mask):
+    token_embeddings = model_output[0] #First element of model_output contains all token embeddings
+    input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float()
+    return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9)
+
+
+# Sentences we want sentence embeddings for
+sentences = ['This is an example sentence', 'Each sentence is converted']
+
+# Load model from HuggingFace Hub
+tokenizer = AutoTokenizer.from_pretrained('{MODEL_NAME}')
+model = AutoModel.from_pretrained('{MODEL_NAME}')
+
+# Tokenize sentences
+encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt')
+
+# Compute token embeddings
+with torch.no_grad():
+    model_output = model(**encoded_input)
+
+# Perform pooling. In this case, mean pooling.
+sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask'])
+
+print("Sentence embeddings:")
+print(sentence_embeddings)
+```
+
+
+
 ## Evaluation Results
 
 <!--- Describe how your model was evaluated -->
@@ -45,29 +84,31 @@ The model was trained with the parameters:
 
 **DataLoader**:
 
-`sentence_transformers.datasets.NoDuplicatesDataLoader.NoDuplicatesDataLoader` of length 905 with parameters:
+`torch.utils.data.dataloader.DataLoader` of length 3076 with parameters:
 ```
-{'batch_size': 624}
+{'batch_size': 16, 'sampler': 'torch.utils.data.sampler.RandomSampler', 'batch_sampler': 'torch.utils.data.sampler.BatchSampler'}
 ```
 
 **Loss**:
 
-`sentence_transformers.losses.MultipleNegativesRankingLoss.MNRLGradCache` 
+`sentence_transformers.losses.MSELoss.MSELoss` 
 
 Parameters of the fit()-Method:
 ```
 {
     "epochs": 1,
-    "evaluation_steps": 90,
-    "evaluator": "sentence_transformers.evaluation.EmbeddingSimilarityEvaluator.EmbeddingSimilarityEvaluator",
+    "evaluation_steps": 500,
+    "evaluator": "sentence_transformers.evaluation.SequentialEvaluator.SequentialEvaluator",
     "max_grad_norm": 1,
     "optimizer_class": "<class 'transformers.optimization.AdamW'>",
     "optimizer_params": {
-        "lr": 0.00032
+        "correct_bias": false,
+        "eps": 1e-06,
+        "lr": 2e-05
     },
     "scheduler": "WarmupLinear",
     "steps_per_epoch": null,
-    "warmup_steps": 91,
+    "warmup_steps": 1000,
     "weight_decay": 0.01
 }
 ```
@@ -77,7 +118,7 @@ Parameters of the fit()-Method:
 ```
 SentenceTransformer(
   (0): Transformer({'max_seq_length': 150, 'do_lower_case': False}) with Transformer model: BloomModel 
-  (1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': True, 'pooling_mode_lasttoken': False})
+  (1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False})
 )
 ```
 

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "bigscience/bloom-560m",
+  "_name_or_path": "Mayhem50/sgpt-bloom-560M-nli-v3",
   "apply_residual_connection_post_layernorm": false,
   "architectures": [
     "BloomModel"

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:140ef9bcc4f8c817983d623387f36750ec34935d61d2b2e053a65cfbb2ad0094
+oid sha256:7ed607694a81d990d4e8e5a98818a12945cfc1c8d5b60e9b3656c372a0c18b6d
 size 2236953889

tokenizer_config.json

@@ -3,7 +3,7 @@
   "bos_token": "<s>",
   "eos_token": "</s>",
   "model_max_length": 1000000000000000019884624838656,
-  "name_or_path": "bigscience/bloom-560m",
+  "name_or_path": "Mayhem50/sgpt-bloom-560M-nli-v3",
   "pad_token": "<pad>",
   "padding_side": "left",
   "special_tokens_map_file": null,