first commit

app.py

@@ -0,0 +1,83 @@
+from transformers import AutoTokenizer, AutoModelForSequenceClassification
+import torch
+
+
+device = torch.device('cuda') if torch.cuda.is_available() else torch.device('cpu')
+device
+
+
+
+download = False
+save_model_locally= False
+if download:
+    tokenizer = AutoTokenizer.from_pretrained("MilaNLProc/feel-it-italian-sentiment", cache_dir="data/")
+    model = AutoModelForSequenceClassification.from_pretrained("MilaNLProc/feel-it-italian-sentiment", cache_dir="data/")
+    model.eval()
+    tokenizer_emo = AutoTokenizer.from_pretrained("MilaNLProc/feel-it-italian-emotion", cache_dir="data/")
+    model_emo = AutoModelForSequenceClassification.from_pretrained("MilaNLProc/feel-it-italian-emotion", cache_dir="data/")
+    model_emo.eval()
+    if save_model_locally:
+        model.save_pretrained('./local_models/sentiment_ITA')
+        tokenizer.save_pretrained('./local_models/sentiment_ITA')
+        model_emo.save_pretrained('./local_models/emotion_ITA')
+        tokenizer_emo.save_pretrained('./local_models/emotion_ITA')
+else:
+    tokenizer = AutoTokenizer.from_pretrained("./local_models/sentiment_ITA/")
+    model = AutoModelForSequenceClassification.from_pretrained("./local_models/sentiment_ITA/", num_labels=2)
+    model.eval()
+    tokenizer_emo = AutoTokenizer.from_pretrained("./local_models/emotion_ITA/")
+    model_emo = AutoModelForSequenceClassification.from_pretrained("./local_models/emotion_ITA/", num_labels=4)
+    model_emo.eval()
+
+
+#%%
+
+from transformers import pipeline
+import re
+
+generator = pipeline(task="text-classification", model=model, tokenizer=tokenizer, return_all_scores =True)
+generator_emo = pipeline(task="text-classification", model=model_emo, tokenizer=tokenizer_emo, return_all_scores =True)
+
+def sentiment_emoji(input_abs):
+
+    if(input_abs ==""):
+        return "🤷‍♂️"
+        
+    res = generator(input_abs)[0]
+    res = {res[x]["label"]: res[x]["score"] for x in range(len(res))}
+    res["🙂"] = res.pop("positive")
+    res["🙁"] = res.pop("negative")
+    return res
+
+
+def emotion_emoji(input_abs):
+    if(input_abs ==""):
+        return "🤷‍♂️"
+
+    res = generator_emo(input_abs)[0]
+    res = {res[x]["label"]: res[x]["score"] for x in range(len(res))}
+    res["😃"] = res.pop("joy")
+    res["😡"] = res.pop("anger")
+    res["😨"] = res.pop("fear")
+    res["😟"] = res.pop("sadness")
+   
+    return res
+#%%
+
+import gradio as gr
+demo = gr.Blocks()
+with demo:
+   gr.Markdown("# Analisi sentimento/emozioni del testo italiano")
+   with gr.Row():
+      with gr.Column():
+         text_input = gr.Textbox(placeholder="Scrivi qui")
+         button_1 = gr.Button("Invia")
+      with gr.Column():
+         label_sem = gr.Label()
+         label_emo = gr.Label()
+      #   gr.Interface(fn=emotion_emoji, inputs=text_input, outputs="label")
+   button_1.click(sentiment_emoji, inputs=text_input, outputs=label_sem)
+   button_1.click(emotion_emoji, inputs=text_input, outputs=label_emo)
+                
+
+demo.launch()

local_models/emotion_ITA/config.json

@@ -0,0 +1,41 @@
+{
+  "_name_or_path": "MilaNLProc/feel-it-italian-emotion",
+  "architectures": [
+    "CamembertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 5,
+  "classifier_dropout": null,
+  "eos_token_id": 6,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "anger",
+    "1": "fear",
+    "2": "joy",
+    "3": "sadness"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "anger": 0,
+    "fear": 1,
+    "joy": 2,
+    "sadness": 3
+  },
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "camembert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "output_past": true,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.11.3",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 32005
+}

local_models/emotion_ITA/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aae2a4c143248032471a56a42cf9e2afa9930c08535e02df8cbece685fcf5f9c
+size 442585813

local_models/emotion_ITA/special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": false}, "additional_special_tokens": ["<s>NOTUSED", "</s>NOTUSED"]}

local_models/emotion_ITA/tokenizer_config.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "sep_token": "</s>", "cls_token": "<s>", "unk_token": "<unk>", "pad_token": "<pad>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "additional_special_tokens": ["<s>NOTUSED", "</s>NOTUSED"], "special_tokens_map_file": null, "name_or_path": "MilaNLProc/feel-it-italian-emotion", "tokenizer_class": "CamembertTokenizer"}

local_models/sentiment_ITA/config.json

@@ -0,0 +1,37 @@
+{
+  "_name_or_path": "MilaNLProc/feel-it-italian-sentiment",
+  "architectures": [
+    "CamembertForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 5,
+  "classifier_dropout": null,
+  "eos_token_id": 6,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "negative",
+    "1": "positive"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "negative": 0,
+    "positive": 1
+  },
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "camembert",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "output_past": true,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.11.3",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 32005
+}

local_models/sentiment_ITA/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6f779a009987fcd3392dedad71923c8c4b8c88ff74898eeae4b30d5c94601844
+size 442579735

local_models/sentiment_ITA/special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": false}, "additional_special_tokens": ["<s>NOTUSED", "</s>NOTUSED"]}

local_models/sentiment_ITA/tokenizer_config.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "sep_token": "</s>", "cls_token": "<s>", "unk_token": "<unk>", "pad_token": "<pad>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "additional_special_tokens": ["<s>NOTUSED", "</s>NOTUSED"], "special_tokens_map_file": null, "name_or_path": "MilaNLProc/feel-it-italian-sentiment", "tokenizer_class": "CamembertTokenizer"}